Lockheed Martin Corporation v. United States

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
__________________________________________ 
)
LOCKHEED MARTIN CORPORATION, )
)
Plaintiff, )
)
v. ) Civil Action No. 08-1160 (ESH)
) UNDER SEAL
UNITED STATES, )
)
Defendant. )
_______________________________________)

MEMORANDUM OPINION

Lockheed Martin Corporation brings this action against the United States under the

Comprehensive Environmental Response, Compensation, and Liability Act (“CERCLA”), 42

U.S.C. § 9601 et seq., for recovery of past and future response costs to remediate the

environmental contamination caused by its corporate predecessor’s operation of three rocket

motor-production facilities – Redlands, Potrero Canyon, and LaBorde Canyon – in California

from 1954 to 1975. Both parties admit to being liable as potentially responsible parties (“PRPs”)

for the contamination at the three facilities under CERCLA § 107(a). The Court held a twelve-

day bench trial from February 10 to March 14, 2014, to determine the equitable allocation of

response costs between the parties. Having considered the evidence, the controlling law, and all

relevant equitable factors, the Court has determined that an equitable allocation for the past

response costs for all three facilities is 0% liability to the United States and 100% liability to

Lockheed. In contrast, the Court will equitably allocate future response costs between the parties

differently for each facility: 29% to the United States and 71% to Lockheed for the Redlands

facility; 24% to the United States and 76% to Lockheed for the Potrero Canyon facility; and 19%

to the United States and 81% to Lockheed for the LaBorde Canyon facility.

1
 TABLE OF CONTENTS

BACKGROUND .............................................................................................................................3
I. History of the Sites...................................................................................................3
A. Government contracts for solid propellant rockets at the Sites ...................3
B. Solid propellant rocket operations at the Sites .............................................7
C. General waste disposal practices at the Sites .............................................14
II. Cleanup of the Sites ...............................................................................................15
A. Redlands facility ........................................................................................15
1. Trichloroethylene ...........................................................................16
2. Perchlorate .....................................................................................18
B. Potrero Canyon facility ..............................................................................19
C. LaBorde Canyon facility ............................................................................21
III. Lockheed’s indirect recovery of response costs through U.S.-government
contracts .................................................................................................................22
A. The Federal Acquisition Regulations.........................................................23
B. The Discontinued Operations Settlement Agreement (“DOSA”) .............25
C. Lockheed’s treatment of response costs for the Sites under the DOSA ....27
IV. Procedural history ..................................................................................................29
A. Related actions while the CERCLA statute of limitations was tolled .......29
B. The government’s partial motion for summary judgment .........................33
C. Trial on equitable allocation ......................................................................36
LEGAL FRAMEWORK ...............................................................................................................40
FINDINGS OF FACTS AND CONCLUSIONS OF LAW ..........................................................46
I. Sources of contamination at the Sites ....................................................................47
A. Redlands facility ........................................................................................47
1. Perchlorate .....................................................................................47
2. Trichloroethylene ...........................................................................52
B. Potrero Canyon facility ..............................................................................56
C. LaBorde Canyon facility ............................................................................58
II. Traditional equitable allocation .............................................................................59
A. Limited value of certain equitable factors .................................................59
1. Waste attributable to each party .....................................................59
2. Parties’ relative benefits from waste disposal activities ................60
3. Degree of cooperation ....................................................................60
4. The government’s ownership of waste ..........................................62
5. The government’s ownership of facilities......................................64
6. Knowledge of risk of pollution from AP and TCE ........................65
7. Violation of California water quality laws.....................................68
8. Ability to pay .................................................................................74
9. Indemnification provisions ............................................................74
B. LPC exercised significantly more control than the government over the
day-to-day hazardous waste disposal operations at the Sites.....................79
C. The government acquiesced in many of LPC’s disposal operations at the
Sites ............................................................................................................88

2
 D.
Some of LPC’s disposals at the Sites violated internal LPC rules or
government requirements...........................................................................90
E. Conclusion under traditional equitable allocation .....................................93
III. Effect of indirect recoveries on equitable allocation .............................................93
A. Lockheed’s recovery of past response costs would unfairly burden the
taxpayer. .....................................................................................................94
B. Lockheed’s recovery of future response costs would not unfairly burden
the taxpayer. .............................................................................................106
CONCLUSION ............................................................................................................................107

BACKGROUND1

I. HISTORY OF THE SITES

The environmental contamination that underlies this action arose from the operation of

solid propellant rocket production facilities by Lockheed Propulsion Company (“LPC”)2 at three

locations in Redlands and Beaumont, California – the Redlands facility, the Potrero Canyon

facility, and the LaBorde Canyon facility (collectively the “Sites”) – between 1954 and 1975.

A. Government contracts for solid propellant rockets at the Sites

LPC researched, developed, and manufactured state-of-the-art solid propellant rocket

technologies at the Sites in support of military and scientific programs critical to the United

States’ Cold War efforts. (Roman Decl. ¶¶ 21, 23.) Government interest in the development of

solid propellant rocket technologies grew significantly in the 1950s following the Soviet Union’s

successful nuclear tests in 1949 and the Sputnik launches in 1957. (Id. ¶¶ 19, 24-25.) Rocket

motors using solid propellants offered, at a lower cost, several benefits over their liquid-based

counterparts – greater safety, operational readiness, and reliability. (Id. ¶ 24, 27.) The

1
What follows in this section, although labeled “Background” and generally undisputed by the
parties, constitutes findings of fact in accordance with Fed. R. Civ. P. 52.
2
Grand Central Rocket operated the Sites from 1954 to 1963, when LPC purchased the company.
(USX88.0044.) For simplicity’s sake, the Court will refer to GCR and Lockheed Propulsion Company as
“LPC.” During its existence, LPC was an operating division of Lockheed Aircraft Corporation, which, in
1977, became the Lockheed Corporation. The Lockheed Corporation merged with the Martin Marietta
Corporation in 1995 to form the Lockheed Martin Corporation.

3
Eisenhower administration’s decision to initiate the development of several large missile

programs and to prioritize the research and development of solid propellants created the market

for a private solid propellant industry. (Id. ¶ 33.) As the only purchaser of advanced solid

propellant rockets at the time, the United States held monopsonistic control over the solid

propellant industry. (Trial Tr. 65-66 (Roman); Roman Decl. ¶¶ 5, 34.)

LPC was one of the largest participants in that industry, and as a contractor, it developed

or manufactured rockets for eight major Cold War programs: the Vanguard artificial satellite,

the Explorer artificial satellite, the Nike-Zeus anti-ballistic missile system, Project Mercury, the

Apollo Program, the Large Solid Propellant Motor Program, the TAGBOARD reconnaissance

drone, and the Short Range Attack Missile (“SRAM”) program. (Roman Decl. ¶¶ 7, 35.)

President Eisenhower designated four of these programs – Vanguard, Nike-Zeus, Mercury, and

Apollo – as of the “highest national priority.” (Id. ¶¶ 7, 23.)

As described by three aeronautics scholars, “[t]he brief life of the Lockheed Propulsion

Company was marked by rather modest, but notable historical and technical achievements in

solid rocket development.” (PX0088 at 14.) LPC contributed to four major Cold War space

programs both as a developer and manufacturer of solid propellant rocket motors. In the 1950s,

LPC developed and manufactured the third-stage motor for the Vanguard satellite program and

loaded solid propellants into motors for the Explorer satellite program. Following the first of

many successful launches into orbit in 1958, both Vanguard and Explorer were foundational

space race programs. (Roman Decl. ¶¶ 8, 36-37.) LPC later developed and manufactured

launch escape motors for Project Mercury, the United States’ first manned space program. The

escape motors – critical for the safety of the astronaut in the event of an emergency during a

launch – had a 100% reliability rate over numerous tests and missions. (Id. ¶¶ 10, 40-41.)

4
Finally, LPC developed and manufactured the launch escape and pitch control motors for the

Apollo manned lunar exploration program. The motors were part of the space capsule for eight

Apollo lunar missions, including Apollo 8, the first manned orbit of the moon in December 1968,

and Apollo 11, the first manned lunar landing in July 1969. (Id. ¶¶ 11, 42-43.)

From 1958 to 1974, LPC also researched, developed, and tested large solid propellant

motors for NASA and the Department of Defense (“DOD”). Large solid propellant motors were

necessary to generate enough thrust to lift large vehicles into space. Under the Large Solid

Propellant Motor Program, LPC designed, fabricated, and tested the first 120-inch and 156-inch

solid propellant motors and contributed to numerous technological advances later incorporated

by competitors in the Space Shuttle and ballistic missile programs. For instance, the “Lockseal”

device developed by LPC as a solution to solid propellant rocket thrust vector control became a

mainstay in the solid propellant rocket industry. (Id. ¶¶ 12-13, 44-45.)

LPC also contributed as a developer and manufacturer to the conventional Cold War

arms race. LPC developed the second-stage motor for the Nike-Zeus missile, a surface-to-air

missile designed to destroy incoming nuclear warheads. The Nike-Zeus missiles were

successfully tested in 1958 and 1959. (Id. ¶¶ 38-39.) LPC also developed and produced forty-

five motors for the then-highly classified TAGBOARD reconnaissance drone program in the late

1960s. LPC designed motors capable of allowing the unmanned drone to reach an altitude and

speed that would ignite the drone’s ramjet. Once the ramjet ignited, the drone could reach

speeds in excess of Mach 3 and photograph an area sixty miles wide and 3000 miles long in a

single flight. The government ultimately discontinued TAGBOARD in 1971 for technical and

political reasons. (Id. ¶¶ 49-50; see also PX0088 at 13.)

5
 LPC’s largest contracts, however, were for the SRAM program. The SRAM was a

nuclear-armed air-to-ground missile designed for use on the aerial front lines in the case of an

American invasion into Soviet territory. (Roman Decl. ¶ 51.) LPC developed and manufactured

a revolutionary two-pulse solid propellant rocket motor, which enabled an individual SRAM to

shut down and then restart mid-flight. This innovation created the possibility for three distinct

flight profiles and an “omni-directional” striking capability for each missile. (Id. ¶¶ 51, 54, 70-

71; see also Trial Tr. at 68 (Roman).)

The Air Force developed the technical requirements for the SRAM program in the spring

of 1964. (Roman Decl. ¶ 53) The Air Force Rocket Propulsion Lab (“AFRPL”) initiated SRAM

research and development contracts with LPC, as well as several of its competitors, that same

year. (Trial Tr. at 1340-41 (Dull).) LPC was the only contractor to successfully demonstrate the

feasibility of a two-pulse motor during its research and development contracts. (Roman Decl. ¶

54.) In 1966, Boeing won the development contract for the SRAM program (Trial Tr. at 1339,

1341 (Dull)) and awarded the subcontract for the development of the propulsion system – i.e.,

the solid propellant rocket motor – to LPC. (Roman Decl. ¶ 55.)

Numerous technical difficulties and rocket failures plagued the SRAM program. (Trial

Tr. at 1341 (Dull); Roman Decl. ¶¶ 56-60.) The September 1969 initial qualification tests for the

SRAM motor were unsuccessful. (Roman Decl. ¶ 61.) Shortly thereafter, Boeing and LPC filed

a $54.2 million claim against the Air Force asserting that the SRAM propulsion system

requirements were “unattainable” and “grossly impracticable.” The Air Force settled the claim

for $20 million before it could be adjudicated. (Id. ¶ 63.)

Setbacks notwithstanding, the Air Force deemed the SRAM motor fit for production in

1971. The Air Force awarded Boeing the first production contract, and Boeing awarded LPC a

6
one-year production subcontract worth $27.6 million. (Id. ¶ 64.) The Air Force awarded Boeing

four more production contracts, and Boeing and LPC entered into four additional production

subcontracts between 1972 and 1975. During that period, LPC produced 1500 SRAM solid

propellant rocket motors and completed 107 consecutive successful test fires. (Trial Tr. at 1395-

96 (Dull); Roman Decl. ¶ 68.) Each SRAM was 150 inches long, nearly 18 inches in diameter

(see Trial Tr. at 1337 (Dull)), and carried a thermonuclear weapon with a yield of approximately

170 kilotons, roughly ten times the yield of the atomic bomb the United States dropped over

Hiroshima during World War II. (Roman Decl. ¶ 69.) Thus, a B-52 bomber with the capacity to

carry 20 SRAMs had a destructive power equivalent to 200 Hiroshima bombs. (Id.)

In light of decreased government focus on solid propellant rocketry for defense and space

exploration purposes, LPC ceased its operations in the summer of 1975 at the end of its fifth

SRAM subcontract. (Id. ¶ 68.) Nonetheless, SRAMs – all of which contained solid propellant

motors produced by LPC – remained a mainstay in the United States’ arsenal through the 1980s

and were considered critical to the deterrent effect of the United States’ strategic bomber force.

(Id. ¶¶ 70-72.) Even twenty years after the first successful SRAM test firing, the Soviet Union

had not developed an air defense system capable of neutralizing the omni-directional capabilities

of SRAMs provided by LPC’s two-pulse rocket system. (Id. ¶ 71.)

B. Solid propellant rocket operations at the Sites

While the parties dispute what happened on a day-to-day basis with respect to operations

at the Sites, particularly relating to the disposal of the hazardous substances, many of the

background facts are undisputed.

LPC began its rocket motor production operations at the nearly 500-acre Redlands

facility in 1954 when it leased the facility from the City of Redlands. (USX15; see also

7
USX11.0020-21.) Between 1958 and 1962, LPC acquired the parcels comprising the much-

larger 9,100-acre Potrero Canyon and 2,500-acre LaBorde Canyon facilities located near

Beaumont, California. (USX12.0044.)3 Lockheed researched and manufactured solid propellant

rockets at the Redlands facility from 1954 to 1975. (USX11.0018-20.) LPC used the Potrero

Canyon facility for manufacturing solid propellant rockets, testing rocket motors, and washing

out defective rocket motors for re-use from 1958 to 1974. (Sterrett Decl. ¶ 213; USX12.0046.)

LPC used the LaBorde Canyon facility for assembling small rocket motors, testing rocket

motors, and washing out defective motors for reuse from 1958 to 1974. (Sterrett Decl. ¶ 214;

USX13.0011-12.)

The design, testing, and production processes for LPC’s contracts followed a general

protocol. The government provided requirements regarding the performance, dimensions, and

interface points of the solid propellant rocket motor. (Trial Tr. at 78 (Oppliger).) From those

specifications, LPC’s engineering analysts manipulated the composition of the propellant and the

physical design of the propellant grain within the rocket casing to achieve desired rocket

performance. (Id. at 78-79.) LPC engineers then designed the rocket casing and casing

insulation based on the characteristics of the propellant grain design. (Id. at 80.) From there,

LPC began the iterative process of testing subscale (and eventually to-scale) rockets against the

government-provided specifications. (See id. at 80-83.)

Although LPC’s applications of solid propellant rocket technology were state-of-the-art,

solid propellant rocket motors themselves are “simple units” with “no moving parts.” (PX91 at

3
Today, the San Bernardino Valley Water Conservation District owns the majority of the
Redlands facility and uses portions of it for water spreading to recharge groundwater. Operators of an
industrial park own the remaining sixty-six acres of the Redlands facility. (USX11.0021.) California
owns and manages as a wildlife area all but nearly 600 acres of the Potrero Canyon facility. Lockheed
owns the remaining acreage under a conservation easement. (USX12.0047.) The County of Riverside
has owned the LaBorde Canyon facility since 2006. (USX13.0011.)

8
934.) A solid propellant rocket “motor consists of an encased energy supply, which is a

combustible mixture of all of the elements required for the generation of propulsive energy” – it

is self-contained and ready to fire when it leaves the factory. (Id.)

A solid propellant is composed of three basic components – an oxidizer and two fuels, a

“rubber-like binder” and powdered aluminum. (Id. at 935.) Because solid propellant

combustion occurs in the closed environment of a rocket’s core, a solid propellant requires much

more oxidizer than fuel. (Id. at 955.) The oxidizer used by LPC consisted of precise proportions

of ground and unground ammonium perchlorate (“AP”). LPC used several grinders at the Sites

to grind raw AP down from 200 microns in diameter (unground) to precise sizes of less than

eight microns in diameter. (Id.) LPC combined the ground and unground oxidizer with a fuel

slurry in a large mixing machine to produce a homogenous fluid propellant. (Id. at 958.)

LPC then “cast” the fluid propellant into the rocket motor case under vacuum conditions,

and “cured” the rocket by heating the cast for nearly a week to “stiffen[] it into a rubbery, shape-

retaining mass.” (Id. at 935, 961.) LPC produced these solid propellant motors in short,

cylindrical segments that could be joined to complete the rocket motor. (Id. at 940.) Once the

nozzle and igniter were added, the rockets were ready for either testing or shipment. (Id. at 938.)

Aside from providing the initial specifications for a particular solid propellant rocket

motor, the government played several additional roles in the design, testing, and production of a

rocket.4 First, as a government contractor or subcontractor, LPC had to comply with any military

and federal specifications incorporated in a particular contract. (PX1057 ¶¶ 13-14 (Speer

Decl.).) Deviation from government specifications required government review and approval.

(Id. ¶ 14.) Second, LPC’s contracts often incorporated by reference specific safety and

4
Because of the lack of contracts and documents from the LPC’s earlier years, this section
focuses primarily on LPC’s later – and biggest – subcontracts under the SRAM program.

9
production manuals. (See Trial Tr. at 941 (Nagle); see also, e.g., PX0001-0003, 0005, 0007,

0009.) The manuals provided guidance for best practices regarding safety and production

processes. Although LPC was not required to comply with all aspects of the referenced and

applicable manuals, LPC was required to comply with, or to seek a waiver or deviation from, any

directives in the manuals that used the words “shall” or “must.” (See Trial Tr. at 439 (Delaney);

id. at 941 (Nagle); PX0007 § 102). Waivers or deviations, however, were commonly sought and

granted. (Trial Tr. at 941 (Nagle); see, e.g., PX400.)

Specifications aside, the government had limited input into LPC’s technical development

of solid propellant rocket motors under government contracts. In the mid-1960s, the government

adopted the so-called Total System Performance Responsibility (“TSPR”) for the SRAM

contracts, meaning that the prime contractor – Boeing – and its subcontractor –LPC – had “total

system responsibility to build th[e SRAM] without any more direction from the Air Force.”

(Trial Tr. at 1338 (Dull); see USX222.0003.) “The Air Force provide[d] a statement of work,

what [it] want[ed] to be built, and a system spec that goes into technical details” regarding

performance, but it did not “dictate the design” of the rocket. (Trial Tr. at 1338 (Dull).)

Given the technical complexity of the SRAM program, LPC held daily early morning

meetings to discuss pertinent program details. (See Trial Tr. at 86 (Oppliger).) Government

representatives from the Air Force Systems Program Office (“SPO”)5 and the AFRPL attended

these meetings, as well as other technical interchange meetings scheduled by Boeing, only on

invitation by Boeing. (Id. at 1341, 1345-47 (Dull).) Although the frequency of these meetings

increased over time due to the repeated rocket motor failures that plagued SRAM, government

representatives at the meetings were instructed not to provide direction, but to only “observe,

5
The SPO, located at Wright-Patterson Air Force Base near Dayton, Ohio, was the Air Force
division responsible for procuring weapons systems, including SRAM. (Trial Tr. at 1335-36 (Dull).)

10
take notes, and report back to . . . supervisors.” (Id. at 1347-48.) Even at SRAM’s preliminary

design review, which included high-ranking officials from LPC, Boeing, and the Air Force, the

Air Force representatives attended only “as observers to witness the progress at [LPC] at that

time.” (Id. at 1349.) Waste disposal practices were not discussed at these technical meetings.

(Id. at 1352.)

At the beginning of its SRAM development subcontract, LPC formed the SRAM

Propulsion Program Review Committee to provide LPC monthly guidance on meeting the

SRAM program’s technical and schedule objectives. (PX0571 at 443.) The Committee was

initially comprised of Lockheed Aircraft Corporation employees who did not generally work for

LPC at the Sites. (Id.) In early 1968, pending the commencement of SRAM rocket motor test

firing, LPC , with the permission of Boeing, expanded the Committee to include several outside

experts, including Boeing engineers, scholars, and Donald Ross, “Mr. Solid Rocket,” from the

AFRPL. (Id. at 444, 446; see Trial Tr. at 1369-70 (Dull).) Although Mr. Ross was generally an

“observer” under the TSPR (Trial Tr. at 1370 (Dull)), he provided some technical input to LPC

through the Committee. (See PX0571 at 0439.) LPC promptly implemented many of the

improvements recommended by the expanded Committee. (See Roman Decl. ¶ 60.)

In 1970, at the request of the SPO, the AFRPL reviewed the design of the SRAM rocket

motor, the test firing results, and LPC’s tooling and manufacturing processes and internal

procedures, to ensure that LPC “w[as] ready for production” and to “document the things that

needed to be completed” before SRAM rocket motor production could begin. (Trial Tr. at 101,

104-05 (Oppliger); see also Trial Tr. at 1356 (Dull); PX0577 at 110.) Pursuant to this review,

the AFRPL provided eight recommendations to Boeing regarding LPC’s readiness for SRAM

rocket motor production. (See PX0577 at 115-18.) These recommendations ranged from the

11
procedural – recommending that Boeing conduct a First Article Contractual Inspection following

the eight test motor firings – to the prudent – suggesting that LPC’s engineering work force

could be reduced by fifty percent once development ended and production began. (Id. at 116,

118.) Although the government recommendations did not bind either Boeing or LPC (Trial Tr.

at 1376-77 (Dull)), LPC ultimately adopted many of them. (Id. at 106 (Oppliger).)

The government’s quality assurance presence at the Sites was much larger than its

technical development presence described above. Although LPC had its own safety and quality

control inspectors (id. at 128), both Boeing (as prime contractor for SRAM) and the Defense

Contract Administration Service (“DCAS”) had offices at the Redlands facility and had the

“right[]” to “roam the production floor and see anything they wanted to see and watch anything

they wanted to watch.” (Id. at 90-92, 94, 110-11.) DCAS officials, Boeing officials, and LPC

engineers reviewed the process specifications and manufacturing process standards before

production could begin and inspected the processes once underway. (Id. at 90.)6 DCAS officials

determined which steps in the production process they wanted to inspect and, in theory, a

production process could not proceed beyond a dedicated inspection point until a DCAS official

had inspected and “stamped off” the process. (Id. at 90-92.)

During the SRAM development and production contracts, DCAS had between four and

five full-time representatives and Boeing had around twenty full-time employees stationed at the

Redlands facility. (Trial Tr. at 111 (Oppliger); id. at 1346, 1357 (Dull).) By comparison, LPC

6
“Process specifications” identified the applicable government specifications for a contract and
provided a high-level description of the production process. (See PX1057 ¶¶ 16-17 (Speer Decl.); see,
e.g., PX328.) Manufacturing process standards, on the other hand, were “how-to” documents that
provided step-by-step instructions for the completion of certain processes in compliance with the process
specifications. (PX1057 ¶ 22; see, e.g., PX913.) That is, “[m]anufacturing process standards were used to
tell the operators how to do or how to use a certain piece of equipment or how to do a certain thing.”
(Trial Tr. at 117 (Oppliger).)

12
had around four hundred employees at the Redlands facility. (Trial Tr. at 139 (Oppliger);

PX1202 at 133.)

The frequency and quality of inspections at the Sites varied over time. “The amount and

kind of inspection to be performed by the Government [wa]s at the discretion of DCAS.”

(USX270.) In its 1970 review of the SRAM Rocket Motor, the AFRPL criticized all three

parties – LPC, Boeing, and DCAS – for practicing a laissez-faire “‘call us and we’ll come and

inspect’” approach to inspecting production processes. (PX577 at 117, 151-52; cf. PX576 at 48-

49, 81.) The AFRPL therefore concluded that that the SRAM program had “not been receiving

the on-station witnessing of work that the program has and continues to deserve” (PX577 at 151-

52) and suggested that SPO recommend to Boeing and DCAS to “re-evaluate their inspection

philosophy and inspection operations with a view to substantially increasing their on-station

witness of work operations besides work results.” (Id. at 117; see also Trial Tr. at 1375-76

(Dull).) After this review, both inspection points and the frequency of inspections increased at

the Sites as the first SRAM production contract got underway. (Trial Tr. at 138 (Oppliger).)

Nonetheless, DCAS continued to “lean[] on Boeing for engineering judgment and decisions

essential to the quality assurance function.” (USX221.0004.)

In addition to their daily quality assurance inspections, DCAS representatives also

occasionally undertook safety inspections at the Sites. (See, e.g., PX476-77, 482-84.)7

However, these inspections were limited to safety risks involving fires and explosions inherent in

the production, testing, and disposal of propellants and did not address safety risks regarding

environmental pollution. (See, e.g., Trial Tr. at 87 (Oppliger).)

7
Prior to the creation of DCAS in 1964, the Los Angeles Ordnance District undertook safety
inspections. (Trial Tr. at 960-61, 1013-14 (Nagle).)

13
 C. General waste disposal practices at the Sites

Solid propellant rocket motor research, design, testing, and production processes produce

myriad and voluminous waste streams including waste propellant (used, unused, and defective),

waste solvents containing propellants, and scrap motors. (See PX457 at 801.) As a result,

disposal of waste was “regarded as an integral part of solid propellant rocket operations.”

(PX0009 § 7-1.1.)

During its operations of the Sites, LPC used several organic solvents – including

trichloroethylene (“TCE”) and 1,1,1-trichloroethane (“TCA”) – to clean the equipment used to

produce the solid propellant rocket motors. At the Redlands facility, AP and propellant-laden

wastewaters, AP and propellant-laden solvents, and solvents in general, were first piped,

pumped, or transported via drums to “evaporation pits.” (Trial Tr. at 669 (Feenstra); Feenstra

Decl. ¶¶ 50-51.) These shallow, concrete-lined basins allowed the organic solvents and water to

evaporate, leaving behind a residual sludge containing large amounts of AP. (See Trial Tr. at

700 (Feenstra).)

LPC ultimately disposed of most of its propellant wastes – including sludge taken from

evaporation pits – by burning them in earthen “burn pits” at the Redlands and Potrero Canyon

facilities. (Feenstra Decl. ¶¶ 51, 156.) In some instances, propellant wastes were disposed of

off-site at Camp Irwin – a military facility in the Mojave Desert (now Fort Irwin). (See id. ¶ 42;

PX431; PX440.)

An exception to disposal-by-burning existed, however, for defective rocket motor

casings. Rather than burning the entire rocket motor – and ruining the expensive metal casing –

LPC often attempted to reuse the motor casing by removing the propellant with water. In the

1950s, LPC did this by “soaking out” the scrapped rocket motors in evaporation pits at the

14
Redlands facility. (Feenstra Decl. ¶¶ 47, 152; see also PX429.) Later, LPC began “hogging out”

defective rocket motors at the Potrero Canyon facility using high-pressure water jets to remove

the propellant from the motor casing. (See Trial Tr. at 393 (Delaney); USX49.0068.)

II. CLEANUP OF THE SITES

As of the beginning of the trial, Lockheed had incurred environmental response costs for

the Sites totaling nearly $287 million. Lockheed estimates it will incur another $124 million in

future response costs for the Sites. (See Meyer Decl. ¶ 50 fig. 29.) Because each facility has a

distinct operational and cleanup history, the Court considers them separately below.

A. Redlands facility

TCE8 and perchlorate9 are the principal contaminants driving the cleanup costs relating to

the Redlands facility. Due to a combination of hydrogeologic factors around Redlands, the

majority of the TCE and AP disposed of during LPC’s operations is no longer located at the

facility. (Sterrett Decl. ¶ 44.) Rather, the substances percolated through the soil and into the

groundwater and have travelled downgradient to form the “Redlands plumes,” 10 approximately

four miles away. (See generally id. ¶¶ 29-45; see also Trial Tr. at 286-87 (Blackman).) Because

8
TCE is now recognized as a probable carcinogen for humans (PX1621 at 263) and the ingestion
of TCE-contaminated water has been associated with a wide variety of other health problems.
(USX826.0023, 0067-0111.)
9
During the trial, at the Court’s direction, the parties and witnesses did not distinguish between
“AP” and “perchlorate.” However, the Court must distinguish between the two in this Memorandum
Opinion: while AP is the oxidizer LPC used and disposed of at the Sites, perchlorate is the component of
AP that contaminates the soil and groundwater at the Sites.
AP is an inorganic salt that “dissociates” in water into its constituent cation and anion:
ammonium (NH4+) and perchlorate (ClO4-), respectfully. (See PX1224 at 1; PX1685 ¶ 3.) Perchlorate is
highly soluble and mobile in water and chemically stable. As a result, perchlorate forms persistent
contaminant plumes when it is released into surface or groundwaters. (PX1224 at 1.) Perchlorate is
harmful to human health because, even in the low µg/L range, it can interfere with iodide uptake by the
thyroid gland, thereby resulting in decreased thyroid hormone production. (PX1224 at 4, 35-36.)
10
Although substantially overlapping, the perchlorate and TCE plumes are distinct. (Sterrett
Decl. ¶ 46.)

15
the plumes are “detached” from the Redlands facility, environmental investigation of the facility

cannot demonstrate a “direct connection” between specific locations (and thus production

activities) at the facility and the plumes. (Sterrett Decl. ¶ 52.) That is, simply by looking at the

plumes and the facility, one cannot reliably discern where in the facility the TCE or perchlorate

originated.

1. Trichloroethylene

The California Department of Health Services first discovered the Redlands TCE plume

in 1980. (PX1677 ¶ 1.) Throughout the 1980s, several municipal wells were shut down after

testing revealed TCE concentrations above the California drinking water maximum contaminant

level of 5 µg/L. (Id. ¶ 3.)11 Lockheed began investigating the plume in the early-to-mid 1980s,

concluding in 1985 that a local airport, and not LPC’s activities at the Redlands site, was the

“most likely source” for the TCE. (Id. ¶ 12.) Notwithstanding that conclusion, the Santa Ana

Regional Water Quality Control Board instructed Lockheed to conduct further investigations into

its operations at the Redlands site. (Id.) Lockheed and the Board continued investigating the

Redlands facility as a potential source for the TCE plume into the early 1990s. (Id. ¶¶ 13-32.)

By October 1992, Lockheed argued that continued investigations into the Redlands

facility would be fruitless because whatever TCE might have been disposed of at the facility had

either volatized or dispersed through the porous soil and formed the detached Redlands TCE

plume. (Id. ¶ 33.) In November, the Board informed Lockheed that it had concluded the

Redlands facility was the source of the Redlands TCE plume and that it would order Lockheed to

investigate and cleanup the plume. (Id. ¶ 34.) Lockheed replied that it would consider proposing

11
As of 2008, the maximum TCE concentrations in the plume are in the mid-20 µg/L range.
(USX11.0032.)

16
to the Board a remediation plan that would negate the need for any order and thereafter

undertook a detailed study of the Redlands TCE plume. (Id. ¶¶ 34-35.)

On May 6, 1993, Lockheed denied responsibility for the plume, informing the Board that

“it was [its] position that there was not substantial evidence to indicate that Lockheed was the

source of the TCE contamination in the [Redlands plume], and that Lockheed, therefore, was not

in a position where they could justifiably utilize stockholders’ funds in conducting any additional

work.” (Id. ¶ 36.)12 The Board responded on January 28, 1994, by issuing its first Cleanup and

Abatement Order for the Redlands TCE plume. The Order required, inter alia, that Lockheed

submit a workplan for using groundwater monitoring wells to define the plume and based on

data gathered from those wells, submit and implement plans to first contain and then remediate

the plume. (Id. at 10-11; see also Trial Tr. at 286-87 (Blackman).) On that same day, the Board

also issued an “investigative order” requiring Lockheed to conduct groundwater and subsurface

soil investigations at the Redlands facility. (PX1678 at 10; see also Trial Tr. at 285 (Blackman).)

On April 22, 1994, after discussions with Lockheed, the Board modified its initial Cleanup and

Abatement Order by removing the requirement that Lockheed implement any remedial actions

for the plume while investigations were still ongoing. (PX1679 at 10-11; see also Trial Tr. at

295-96 (Blackman).) Lockheed continued to deny responsibility for the Redlands TCE plume

(see USX653 (January 1994 letter from Lockheed to Board)), but otherwise complied with the

1994 Orders and worked with the Board to develop an acceptable remediation plan. (Trial Tr. at

296-99 (Blackman).)

12
Mr. Blackman explained that Lockheed initially was unconvinced that the Redlands TCE
plume originated from LPC’s operations at the Redlands facility because testing of groundwater at the
facility had not revealed the presence of TCE. (Trial Tr. at 288 (Blackman).) That is, Lockheed was
unconvinced of its liability because the plume was “detached from” the facility. (See id.)

17
 2. Perchlorate

In late April 1997, following a breakthrough in testing methodologies for perchlorate (see

Trial Tr. at 301-03 (Blackman)), the California Department of Health Services discovered

perchlorate levels in several wells within the Redlands TCE plume that exceeded California’s

provisional drinking water standard of 18 µg/L. (PX1685 ¶¶ 3-4.)13 The Board connected

LPC’s use of AP at the Redlands facility – and in particular the waste disposal practices – with

the Redlands perchlorate plume. (Id. ¶¶ 3, 5.) The Board accordingly issued a Cleanup and

Abatement Order requiring Lockheed to investigate and then develop and implement a remedial

action plan for the Redlands perchlorate plume. (Id. at 2.) Lockheed appealed the Order, but

complied during the appellate process. (Trial Tr. at 337 (Blackman).) Because the Redlands

facility was “the only source of [AP] in the” watershed, Lockheed eventually accepted

responsibility for the Redlands perchlorate plume. (See id. at 307.) Further, after the discovery

of the perchlorate plume, Lockheed “began to realize that the TCE, which was fully enveloped in

the perchlorate plume, must have also come from the” Redlands facility as well. (Id. at 308.)

Since that time Lockheed has complied with all orders of the Board regarding

investigation, containment, and remediation of the Redlands plumes, including working with

water purveyors (e.g., surrounding municipalities) to reduce TCE and perchlorate concentrations

to acceptable levels and to drill new wells for drinking water supplies. (Id. at 314-17;

USX11.0023-26.) Among other remedial steps, Lockheed treated from select wells TCE-laden

13
As of 2008, maximum perchlorate levels in the plume ranged from 60 to 90 µg/L.
(USX11.0031.)

18
groundwater with granular activated carbon and AP-laden groundwater with an ion exchange

resin. (See Trial Tr. at 319-20 (Blackman); USX11.0041-42.)14

Although Lockheed has removed large amounts of TCE and perchlorate from the

Redlands plumes, concentrations remain at levels that will require continued treatment that may

“go on for many decades.” (Trial Tr. at 321 (Blackman).) As of 2011, Lockheed had incurred

over $231 million in response costs for the Redlands plumes. 15 (See Meyer Decl. ¶ 50 fig. 29.)

Response costs for the Redlands plumes also make up the lion’s share of the more than $25

million in response costs at the Sites from January 1, 2012 to February 10, 2014, and are

expected to comprise the majority of the projected $125 million in future costs for the Sites.

(Trial Tr. at 1188 (Lockheed counsel); cf. Meyer Decl. ¶ 50 fig. 29.)

B. Potrero Canyon facility

Although TCE, TCA-related compounds,16 and polychlorinated biphenyls have been

found in the soil and groundwater at the Potrero Canyon facility, perchlorate is the principal

contaminant of concern at the facility. (Sterrett Decl. ¶¶ 22.) Unlike at the Redlands facility, the

contamination at the Potrero Canyon facility is not detached, so the perchlorate contamination is

traceable to specific locations at the facility. (Feenstra Decl. ¶ 155; Sterrett Decl. ¶ 216.)

14
Granular-activated carbon is a less costly technology than ion exchange, which was developed
specifically for perchlorate contamination. (Trial Tr. at 322 (Blackman).) While response costs for the
Redlands plume in the 1990s were primarily from the cleanup of TCE, the recent, current, and future
costs are perchlorate-driven. (Id. at 1188 (Lockheed counsel).)
15
Facility-specific data are unavailable for response costs incurred after 2011. (See Meyer Decl.
¶ 50 fig. 29; see also USX378; USX401-402.)
16
These compounds include 1,4-dioxane and 1,1-DCE. 1,4-dioxane is a chemical used to
“inhibit” – or stabilize – TCA. (Feenstra Decl. ¶ 155.) Similarly, TCA – once “uninhibited” by the
preferential removal of 1,4-dioxane in water – degrades in the presence of powdered aluminum (a fuel in
rocket propellant) to 1,1-DCE. (Trial Tr. at 674-75 (Feenstra).) Thus, the presence of 1,4-dioxane and
1,1-DCE can indicate the prior disposal of TCA.

19
 The primary sources of perchlorate soil contamination at the Potrero Canyon facility are

the Large Motor Washout Area and the Burn Pit Area (Sterrett ¶¶ 220-21),17 with perchlorate

levels as high as 302,000 µg/kg and 171,000 µg/kg, respectively. (USX12.0229-30.) The

former Rocket Motor Production Area is a secondary source area, covering a much larger portion

of the facility but with lower perchlorate soil concentrations (20,400 µg/kg). (Id. at 0229, 0244-

45.) The Sanitary Landfill is also identified as a secondary source area, with soil perchlorate

levels as high as 67,300 µg/kg. (Sterrett Decl. ¶ 222; USX12.0233.)

The Burn Pit Area is the primary source of perchlorate groundwater contamination at the

facility. (Feenstra Decl. ¶ 155; Sterrett Decl. ¶¶ 226-27; USX12.0237-38.) Secondary sources at

the facility include the Rocket Motor Production Area and, to a lesser extent, the Large Motor

Washout Area. (Feenstra Decl. ¶ 165; USX12.0237-38.)

In 1986, Lockheed conducted a historical study of the Potrero Canyon facility (along with

the LaBorde Canyon facility) to better plan later investigations into environmental

contamination. (See USX49.0011.) Following that study and a 1989 Consent Order from the

California Department of Health Services, Lockheed has undertaken further remedial

investigations and some remedial actions at the facility, the most recent and relevant beginning

in 2002. (See USX12.0064-90.)

As of 2011, Lockheed had incurred nearly $21 million in response costs for the Potrero

Canyon facility. (See Meyer Decl. ¶ 50 fig. 29.) Further remedial actions are presently under

evaluation. (Feenstra Decl. ¶ 154; see generally USX699.144-300.) As of 2012, the preferred

remedial alternative for perchlorate-impacted soils at the facility is excavation and off-site

removal. The preferred remedial alternative for perchlorate-contaminated groundwater at the

17
The TCE-contaminated soils identified at the facility – located in the Burn Pit Area – were
remediated in the 1990s. (Sterrett ¶ 224; USX12.0233-35.)

20
facility hydraulic containment through the installation of a pump-and-treat system involving ex

situ treatment of the groundwater, with the discharge of remediated water back into the ground or

into local waterways. (USX699.0299-300.)

C. LaBorde Canyon facility

The soil and groundwater at the LaBorde Canyon facility are also contaminated with

perchlorate and, to a lesser extent, TCE. (Sterrett Decl. ¶¶ 25, 238.) As at the Potrero Canyon

facility, groundwater contamination at LaBorde Canyon is not detached and is therefore

traceable to specific locations of historic operations at the facility. (Id. ¶ 235.) The primary

sources of perchlorate at the facility are the Test Bay Canyons and the Waste Discharge Area.

(Feenstra Decl. ¶ 167; Sterrett Decl. ¶ 239.) Perchlorate has been detected in soil at

concentrations exceeding 100,000 µg/kg and in groundwater at concentrations exceeding

100,000 µg/L. (USX13.0236-38, .0241.)

As was the case at Potrero Canyon, Lockheed undertook a historical study of the

LaBorde Canyon facility in 1986. (See USX49.0011.) The California Department of Health

Services issued a Consent Order requiring cleanup of contamination at the facility in 1989.

(USX13.0012.) Lockheed completed remedial investigations and removal actions from 1990 to

1993, after which the California Department of Toxic Substances Control issued a “Report of

Completion of Removal Action” stating that no further remedial action was necessary. (Id.;

USX700.0037-38.) Groundwater perchlorate contamination was first detected at the facility in

the early 1990s. (Feenstra Decl. ¶ 174.) However, the California Department of Toxic

Substances Control did not reopen the facility for further assessment until 2003. (USX13.0012.)

Since then, Lockheed has engaged in nearly ten years of remedial investigations. (USX13.0027-

31.)

21
 As of 2011, Lockheed had incurred over $10 million in response costs for the LaBorde

Canyon facility. (See Meyer Decl. ¶ 50 fig. 29.) Further remedial actions are currently under

evaluation. (USX700.0148-206.) As of 2012, the preferred remedial alternative for the facility

is limited shallow soil removal, plume containment, and institutional controls. (Id. at 0191,

0215.)

III. LOCKHEED’S INDIRECT RECOVERY OF RESPONSE COSTS THROUGH
U.S.-GOVERNMENT CONTRACTS

Lockheed has recovered and continues to recover its response costs for the Sites (and

others sites like them) from its customers by allocating its cleanup expenses to its contracts as

indirect costs. Because the U.S. government is by far Lockheed’s largest customer, to date

Lockheed has indirectly recovered18 $208 million – over 72% of its total past response costs for

the Sites – through indirect costs charged to U.S.-government (primarily DOD) contracts.19

Because the U.S.-government share of Lockheed’s business is larger than it was in the past, the

percentage of total response costs for the Sites that Lockheed has recovered through U.S.-

government contracts is expected to rise in the future.

This cost-recovery system has its foundation in the byzantine federal contracting

regulations and was formalized, as to environmental response costs in particular, by the

Discontinued Operations Settlement Agreement (“DOSA”) signed by Lockheed and the U.S.

18
In its public filings, Lockheed refers to its formalized recoupment of environmental response
costs through its contracts with the U.S. government as a “recovery.” (See USX397.0069 (Lockheed
2012 Annual Report).)
19
The vast majority of Lockheed’s U.S.-government contracts are with the DOD. (See
PX397.0017.) However, Lockheed also has contracts with NASA, the Federal Aviation Administration,
the Social Security Administration, the Department of Treasury, the Department of Justice, the
Department of Health and Human Services, the Environmental Protection Agency, and the U.S. Postal
Service. (Trial Tr. at 1654 (Gatchel); USX397.0016.)

22
Defense Contract Management Agency (“DCMA”)20 in 2000. As a result of these regulations

and the DOSA, the Court must decide, in determining the equitable allocation of response costs

between the parties under CERCLA, what weight, if any, should be given to the fact that a

government contractor (1) has already formally, though indirectly, “recovered” from the U.S.

government a significant portion of its response costs through the pricing of its goods and

services and (2) now seeks to directly recover from the U.S. government those same response

costs under CERCLA. As the parties agree, this is an issue of first impression.21 But before the

Court can address that question, it must explain the relevant regulations and the DOSA, both of

which complicate the resolution of this case.

A. The Federal Acquisition Regulations

Lockheed’s cost-recovery system has an established basis in the complex Federal

Acquisition Regulations (“FAR”) that govern government contracting. Pursuant to the FAR, the

government pays contractors both their “direct” and “indirect” costs, plus a profit. (Wright Decl.

¶ 84.) Direct costs are those costs related to a specific contract, such as materials and labor. See

48 C.F.R. § 31.202. Indirect costs are those costs not associated with a specific contract –

essentially overhead. See id. § 31.203. Profit factors are determined on a contract-by-contract

20
The DCMA is the component of the DOD that engages directly with defense contractors on
issues of contract compliance.
21
Although an issue of first impression, it is far from sui generis. Counsel for the parties
identified at least two other cases involving Lockheed and the government where the same issue will need
to be addressed. (Trial Tr. at 999-1000.) See Lockheed Martin Corp. v. United States, 06-cv-1032-RSL
(W.D. Wash. filed July 21, 2006) (“Seattle Shipyards”); Lockheed Martin v. United States, 06-cv-1438-
RJL (D.D.C. filed Aug. 15, 2006) (“Great Neck”). The issue was also raised in Raytheon Aircraft Co. v.
United States, 2007 WL 4300221 (D. Kan. Dec. 8, 2007), but that case settled before trial. Because
environmental contamination at defense contracting facilities is pervasive, the issue likely looms large in
any case where a major government contractor can sue the government for recovery of environmental
response costs under CERCLA. (Cf. Trial Tr. at 1996 (government closing).)

23
basis. (See Trial Tr. at 600-01 (Wright); id. at 1660-61 (Gatchel).) Across all contracts in 2013,

Lockheed’s operating margin (i.e., pre-tax profit) was over 9.9%.22

The government will only reimburse a contractor for “allowable” indirect costs. An

indirect cost is “allowable” if it is “reasonable,” i.e., “it does not exceed that which would be

incurred by a prudent person in the conduct of competitive business,” 48 C.F.R. § 31.201-3(a), is

“allocable,” complies applicable accounting standards, and is not otherwise disallowed by

regulation or contract. See id. §§ 31.201-2, -4. Although not specifically allowable under the

FAR, see generally id. § 31.205, environmental costs “are normal costs of doing business and are

generally allowable costs if reasonable and allocable.” (PX1862 § 7-2120.1 (Defense Contract

Audit Agency Contract Audit Manual).) 23 Environmental cleanup costs at facilities no longer in

operation are generally allocable as indirect costs. (Id. ¶ 7-7120.7.) Attorneys’ fees and legal

22
See Press Release, Lockheed Martin Corp., Lockheed Martin Reports Fourth Quarter and Full
Year 2013 Results (Jan. 23, 2014), available at http://www.lockheedmartin.com/us/news/press-
releases/2014/ january/0123hq-earnings.html.
23
There was a movement in the late 1980s to make environmental cleanup costs, except for those
at government-owned, contractor-operated facilities, unallowable; however, the proposal was ultimately
withdrawn in the face of industry resistance. See Cpt. Gerald P. Kohns et. al., A Primer on Contractor
Environmental Remediation and Compliance Costs, ARMY LAW., Nov. 1993, at 22, 28. In 1991, a
revised draft environmental cost principle – proposed FAR § 31.205-9 – was approved by Defense
Acquisition Regulation Council and by the Civilian Agency Acquisition Council. Cpt. Joshua H. Van
Eaton, A Not-So Equitable Allocation: The Need for an Environmental Cost Principle, 14 MO. ENVTL. L.
& POL’Y REV. 441, 459 (2007). The proposed principle would have disallowed environmental cleanup
costs except where the contractor could show that it was performing a government contract at the time of
the disposal that required cleanup and that performance of the government contract caused the disposal.
Id. at 460. Even in those instances, however, the contractor would also have to demonstrate
reasonableness of costs, due diligence, compliance with environmental standards of care and laws at the
time of disposal, and the exhaustion of legal remedies to defray cleanup costs. Id. at 473-74. The
regulation was never finalized, presumably because of the moratorium on new regulations in effect at the
time. Id. at 460. Even so, the U.S. Government Accountability Office continued to press for the
development of an environmental cost principle to address what it considered to be inconsistent – and
ultimately detrimental to the government – treatment of environmental costs by contractors. Id. at 461-
63. In 1997, after years of industry resistance, the DOD ultimately abandoned its efforts to develop an
environmental cost principle. Id. at 463-64.

24
costs incurred while pursuing a CERCLA action may also be treated as indirect costs. See 48

C.F.R. § 31.205-47.

For a fixed-price contract, contractors attempt to predict the incurrence of indirect costs

that will be allocated to that contract over its term. Because the price of the contract is fixed, the

contractor benefits from a relatively higher return on its contract if it overestimates the total

indirect costs; on the other hand, if the contractor underestimates the total indirect costs, it will

receive a relatively lower profit on that contract. In contrast, in cost-reimbursement contracts the

contractor is paid for all allowable direct and indirect costs allocated to the contract. Thus, over-

or under-estimating indirect costs with regard to a cost-reimbursement contract does not pose the

same potential for increased or reduced returns. (Cf. Trial Tr. at 1660, 1679 (Gatchel).)

Even after the contractor has allocated indirect costs to specific contracts and has been

paid for those costs pursuant to those contracts, the contractor must credit back to the

government “either as a cost reduction or by cash refund” any “applicable portion of any income,

rebate, allowance, or other credit relating to” those indirect costs “received by or accruing to the

contractor.” 48 U.S.C. § 31.201-5. For environmental cleanup costs in particular, this provision

requires a contractor to credit back to its indirect environmental cost pool any recoveries for

cleanup costs from insurance policies or other PRPs. (Wright Decl. ¶ 50.)

B. The Discontinued Operations Settlement Agreement (“DOSA”)

Lockheed and the DCMA recognized the allowability and appropriate accounting

treatment of environmental remediation expenses as indirect costs when they signed the DOSA

on September 6, 2000. (USX1033 at 10.) The DOSA formally allowed as indirect costs, inter

alia, environmental response costs incurred for “discontinued operations,” sites, or facilities that

25
Lockheed closed, abandoned, or sold prior to January 1, 2000, including all three Sites at issue in

this case. (Id. ¶¶ 1.7-.8.)

Under the DOSA, Lockheed collects its environmental response costs for discontinued

sites in an accounting pool at the corporate level – the Settled Discontinued Operations Pool

(“DiscOps Pool”).24 (Id. ¶ 2.4.) Environmental response costs in the DiscOps Pool are not

charged to contracts immediately upon incurrence; instead, the costs for a given year are

amortized over a five-year period. (USX407 at 5; Mateer Decl. ¶ 8.) Both parties benefit from

amortization because it smoothes costs over time, thereby increasing predictability in multi-year

contract prices and promoting uniformity and comparability in the measurements of contract

costs. (Mateer Decl. ¶ 9; see generally PX1859 at 3-4.)

Lockheed “flows down” the allocable costs for a given year from the DiscOps Pool to its

business units using a three-factor formula typically used in government contracting. (USX1033

¶ 2.8; Wright Decl. ¶ 45.) It is the business units that then develop indirect rates, based partially

on the DiscOps Pool, to use when bidding on and billing government contracts. (Wright Decl. ¶

45; Mateer Decl. ¶¶ 10-12.) The percentage of Lockheed’s indirect costs passed on to U.S.-

government contracts roughly correlates with the U.S. government’s share of Lockheed’s

business for a given year. (Wright Decl. ¶ 74.)

Thus, under DOSA, if Lockheed spent $10 million on environmental response costs at

discontinued operations in the year 2005, it would amortize those costs over the next five years,

and $2 million per year would flow down to its business units from 2006 to 2010. Assuming that

U.S.-government contracts make up 80% of Lockheed’s business, Lockheed would recover $1.6

24
The DOSA also settled disputes between the parties as to the allowability of past response costs
allocated as indirect costs through the DiscOps Pool from 1991 onward. (See USX1033 ¶¶ 2.9-.11.)
Importantly, the DOSA merely formalized past practice and did not materially affect how Lockheed had
treated or would continue to treat the recovery of response costs for the Sites.

26
million per year – or $8 million total – of the response costs through its government contracts. In

addition, Lockheed would recover the remaining $400,000 per year – or $2 million total – in an

identical manner from non-U.S. government contracts.25

The DOSA also provides that Lockheed “shall not realize a double recovery with regard

to any Settled Discontinued Operations Costs,” and to the extent that it does, it “will reimburse

the United States for any such double recovery . . . under government contracts.” (USX1033 ¶

4.7.) Pursuant to this provision, and as required under 48 C.F.R. § 31.201-5, Lockheed credits to

the DiscOps Pool any direct payment it receives for environmental remediation costs at its

discontinued operations, whether in the form of insurance proceeds, settlements, or other

recoveries. (Wright Decl. ¶ 50; Mateer Decl. ¶ 14.) Credits are treated the same as costs and are

amortized over the five years following receipt of the payment. (Wright Decl. ¶ 51.) Therefore,

credits represent a bottom line reduction to the DiscOps Pool, lowering the total indirect costs

passed on to Lockheed’s clients. Thus, again assuming that U.S.-government contracts make up

80% of Lockheed’s business, the U.S. government would receive 80% of the benefit of any

credits allocated to the DiscOps Pool in the form of reduced contract costs. (See Wright ¶ 73-74;

USX1009 at 154-55 (Reese Dep.).)

C. Lockheed’s treatment of response costs for the Sites under the DOSA

To date, the DOSA and DiscOps Pool have functioned as intended. As to the three

discontinued Sites in this case, Lockheed has incurred environmental response costs related to

those Sites, allocated them to the DiscOps Pool, amortized them over five years, and passed them

down to its business units for bidding and billing purposes. Because once individual costs enter

the DiscOps Pool, they are not “tracked” or “followed” (as individual costs) down through the

25
Beyond the U.S. government, Lockheed provides goods and services to other governmental and
some commercial entities, including foreign governments, state governments, municipalities, institutions
of higher education, and the United Parcel Service. (Trial Tr. at 1653-56 (Gatchel).)

27
business unit or specific contracts, it is impossible to identify or trace response costs for the Sites

to any particular government contract. (See Mateer Decl. ¶ 13.) Nonetheless, it is clear that

Lockheed has indirectly recovered, and continues to indirectly recover, response costs for the

Sites from the U.S. government through the indirect costs charged to U.S.-government contracts.

(Wright Decl. ¶ 57-59; USX407 at 4-5.) Indeed, Lockheed boasts that its “underlying tenet in

pricing [its] contracts with the U.S. government” is its “ability to recover [its] costs . . .

regardless of the type of contract.” (USX407 at 4-5.)

As of December 31, 2013, Lockheed had indirectly recovered through its U.S.-

government contracts more than $208 million – or 72% – of its estimated $287 million in total

nominal response costs at the Sites. (Meyer Decl. ¶ 29 & fig. 5.)26 In the future, this “effective

government share” of response costs at the site is expected to rise because U.S.-government

contracts will make up an even higher percentage of Lockheed’s business than in the past. (See

Trial Tr. 581 (Wright); Meyer Decl. ¶ 82 & fig. 30; USX 402.0002.) Indeed, in the past few

years, Lockheed’s recovery rate through U.S.-government contracts has been approximately 87%

(Wright Decl. ¶ 66), and it is predicted that this recovery rate will continue to fluctuate around

that percentage in the near future. (See Trial Tr. 591-93 (Wright); USX 402.0002.) Accounting

for the increase in Lockheed’s U.S.-government contract base, and considering the projected

future cleanup costs for the Sites, the government’s expert economist estimates that Lockheed

will ultimately indirectly recover through U.S.-government contracts nearly 83% of its total

nominal response costs at the Sites. (Meyer Decl. ¶ 30 & fig. 6.)

26
Lockheed also earned, and continues to earn, a profit factor on every dollar of response costs
passed through to its contracts, U.S.-government or otherwise. (See Meyer Decl. ¶ 39; Wright Decl. ¶¶
77, 83.) Thus, the higher the indirect costs charged to the DiscOps Pool, the more profit Lockheed earns
from its clients. (See Wright Decl. ¶ 84.)

28
 Aside from its indirect recovery of over 72% of its response costs for the Sites through

U.S.-government contracts to date, Lockheed has also allocated as an indirect cost and

subsequently recovered (though not through the DiscOps Pool) a substantial portion of its legal

fees and costs associated with bringing this action. The government’s expert economist

estimates that Lockheed indirectly recovered through government contracts, in nominal dollars,

over 85% of the $7.35 million in legal fees and costs it incurred between 2007 and 2012. (Meyer

Decl. ¶¶ 163-75.) Lockheed incurred an additional $3.2 million in legal fees and costs through

November 2013 (Gov’t Closing Presentation [Dkt. No. 144] at 245), and undoubtedly, it has

incurred several million dollars more in fees and costs for the extensive pretrial preparation and

trial briefing since December 2013, the expert fees for the six experts who provided testimony at

trial, and the fees and costs associated with the five counsel who attended a twelve-day trial.

According to the government’s expert economist, Lockheed will indirectly recover over 85% of

its total legal fees and costs (presumably at least $10 million) through U.S.-government

contracts. (See Meyer Decl. ¶ 172.)

IV. PROCEDURAL HISTORY

A. Related actions while the CERCLA statute of limitations was tolled

Lockheed did not bring this action for recovery of response costs under CERCLA until

2008, fourteen years after it began remediation efforts at the Redlands facility. The parties,

through Lockheed’s in-house counsel and attorneys within the Environmental and Natural

Resources Division of the Department of Justice, agreed to a tolling of the CERCLA statute of

limitations, see 42 U.S.C. § 9613(g)(2), beginning in 1997. (See PX1788, 1823, 1838, 1849,

2078.) These tolling agreements, specifically for the Redlands site, were in effect from January

27, 1997 to November 1, 2003. (PX 1788, 2078.) The parties renegotiated the agreements on a

29
semi-annual basis during this time period, including negotiating a one-year extension of the

tolling agreement on July 10, 2000 (PX1849), less than two months before Lockheed and the

DCMA entered into the DOSA on September 6, 2000. (USX1033 at 10.)

Although the DOSA addressed environmental cleanup costs, it did not purport to settle

CERCLA liabilities between Lockheed and the government. To the contrary, the DOSA

explicitly provides that it “does not settle any claims, if any, arising under CERCLA.” (Id. ¶

4.18.) Indeed, the DOSA implicitly recognized the potential coexistence of direct recoveries

from the government under CERCLA and indirect recoveries from the government through costs

charged to U.S.-government contracts. In particular, the parties devoted an entire section of the

DOSA to the treatment of the January 20, 2000 Consent Decree signed by Lockheed and the

Department of Justice in the Burbank CERCLA litigation.27 (USX1033 ¶¶ 3.1-.5.) In the

Burbank Consent Decree, the United States agreed to pay Lockheed directly for over forty-four

percent of past response costs and for fifty percent of most future response costs for the site.

(PX1844 ¶¶ 3.1-.6, 4.1-.3.) While the Burbank Consent Decree acknowledged that Lockheed

was already allocating “certain environmental remediation costs” to its discontinued operations

pool, it did not purport to “resolve[]” the “allowability and allocability of these costs.” (Id. ¶

2.14.) Nonetheless, the Consent Decree did specify that under no circumstance could the United

States determine that the agreed-upon past costs were “not properly subject to payment under the

terms of the Decree because the costs were previously reimbursed by the United States or

27
The Burbank case arose when, in 1991, the United States sued Lockheed, among others, under
CERCLA § 107(a) for the recovery of response costs associated with the cleanup of TCE at a former
Lockheed aircraft and aeronautic production facility in Burbank, California. (PX1844 ¶¶ 2.1, 2.5.) The
parties signed an initial consent decree in 1992 requiring Lockheed to operate a groundwater treatment
plant at the Burbank facility and to reimburse the government for response costs it had incurred. (Id. ¶
2.5.) In 1997, Lockheed filed a counterclaim against the government under CERCLA § 113(f), seeking
contribution for its past and future response costs at the Burbank facility. (Id. ¶ 2.9.) The 2000 Burbank
Consent Decree settled Lockheed’s counterclaim against the United States.

30
another party through an overhead pool” like the discontinued operations pool. (Id. ¶ 3.16.) To

counterbalance that provision, the Consent Decree required that “Lockheed Martin shall not

realize a double recovery with regard to any” response costs and “shall credit its continued

operations pool with amounts received from the United States pursuant to this Decree in

accordance with an agreement of Lockheed Martin and the United States.” (Id. ¶ 3.25.) The

“agreement” encompassed the DOSA, wherein the parties agreed to disallow $80.6 million in

costs related to the Burbank site and disallow $29 million in credits paid to Lockheed by the

United States pursuant to the Burbank Consent Decree. (USX1033 ¶¶ 3.1-.3.)

During the tolling period, Lockheed was involved in several related matters. In 1993, a

group of insurers sued Lockheed in state court in California, seeking a declaration that they were

not required to defend or indemnify Lockheed for the costs of environmental remediation at

multiple sites, including the Redlands facility. Procter v. Lockheed Corp., Case No. 731752,

Complaint (Cal. Sup. Ct. May 27, 1993) (PX2070). The trial court ruled for the insurers,

holding, inter alia, that Lockheed’s indirect recovery of response costs through government

contracts triggered the “government reimbursement exclusion” contained in the policies, and

thus, the insurers were not required to defend or indemnify Lockheed. Procter v. Lockheed

Corp., Case No. 731752, Statement of Decision Re: Phase II at 65-79 (Cal. Sup. Ct. Oct. 27,

2003) (PX2073). Prior to the entry of judgment, Lockheed sought declaratory relief in federal

court to enjoin the state court’s preliminary ruling. In that action, the government filed an

application to intervene in support of Lockheed’s position that indirect payments for response

costs through government contracts does not reduce or eliminate insurance coverage under the

government reimbursement exclusion. (PX2069 ¶¶ 25-26 (Porterfield Decl.).) The federal court

dismissed Lockheed’s action for lack of subject matter jurisdiction, and the state court entered

31
final judgment on October 22, 2003. (Id. ¶¶ 27-28.) Lockheed immediately appealed, and on

November 22, 2005, the California Court of Appeal affirmed the trial court’s decision. See

Lockheed Corp. v. Cont’l Ins. Co., 35 Cal. Rptr. 3d 799 (Cal. Ct. App. 2005). The Supreme

Court of California denied Lockheed’s petition for review on March 22, 2006, and the parties

subsequently entered into a confidential settlement agreement whereby the insurers “bought

back” the disputed policies. (Trial Tr. at 479 (Lockheed counsel); PX2069 ¶ 31 (Porterfield

Decl.); PX2075-76.) The proceeds from that agreement were credited to the DiscOps Pool.

(Trial Tr. at 1455-56 (Lockheed counsel).)

In 1995, Lockheed sued Seven W Enterprises, which at the time operated a

manufacturing facility on a portion of the Redlands facility, for recovery of response costs

incurred for the Redlands TCE plume. See Lockheed Martin Corp. v. Seven W Enterprises, 95-

cv-6153, Complaint (C.D. Cal. Sept. 14, 1995). Pursuant to a confidential settlement in late

1996, Lockheed indemnified Seven W from future liability in exchange for a lump sum payment.

Lockheed credited the amount recovered under this confidential agreement to the DiscOps Pool.

(Trial Tr. at 476-77 (Lockheed counsel).)

That same year, a group of plaintiffs filed a toxic tort class action suit against Lockheed

seeking medical monitoring costs and punitive damages based on the contamination of drinking

water from the Redlands TCE plume. See Lockheed Martin Corp. v. Superior Court, 63 P.3d

913, 916 (2003). The California Court of Appeal reversed the trial court’s class certification and,

in 2003, the Supreme Court of California affirmed. Id. at 922. The case settled some ten years

later, while the second bellwether group of plaintiffs was seeking class certification. (Trial Tr. at

372-73 (Lockheed counsel).) Under the DOSA, Lockheed’s payments pursuant to its

32
confidential settlement of the toxic tort action are specifically allowable as indirect costs charged

to the DiscOps Pool. (USX1033 ¶1.8.)

Finally, in 2004, Lockheed – sponsored by The Boeing Company – pursued an appeal

under the Contracts Dispute Act, 41 U.S.C. § 7101 et seq., to the Armed Services Board of

Contract Appeals (“ASBCA”) seeking indemnification from the government for response costs

and toxic tort liabilities at the Sites arising out of LPC’s performance of SRAM production

contracts from 1971 to 1975. See generally In re Boeing Co., ASBCA No. 54853, 06-1 BCA ¶

33,270 (Apr. 12, 2006).28 Lockheed sought indemnification under provisions in its SRAM

production subcontracts that provided for government indemnification for certain claims or

losses resulting from “unusually hazardous” risks. See id. After several rulings on cross-

motions for partial summary judgment, see, e.g., Appeal of the Boeing Co., ASBCA No. 54853,

11-2 B.C.A. ¶ 34,813 (July 28, 2011), the parties filed summary judgment briefs on whether

environmental response costs arising out of LPC’s solid propellant rocket motor production were

“unusually hazardous risks.” (Trial Tr. at 374-75 (government counsel).) On April 15, 2013,

after the toxic tort settlement and before the ASBCA could rule on those motions, Lockheed

(through Boeing) voluntarily dismissed its appeal with prejudice. (USX74.)

B. The government’s partial motion for summary judgment

Meanwhile, after over fourteen years of cleanup actions at the Redlands facility, on July

1, 2008, Lockheed brought this action against the United States under CERCLA § 107(a). The

government filed a counterclaim against Lockheed under CERCLA § 113(f) seeking, to the

extent it was a PRP, equitable allocation of response costs between the parties.

28
Boeing sponsored LPC’s claim because Boeing was the prime contractor for the SRAM
production contracts. (Trial Tr. at 371 (Lockheed counsel).)

33
 Several months later, the government moved for a partial summary judgment on the

ground that Lockheed could not recover under CERCLA § 107(a) those response costs for the

Sites that it had already recovered through indirect costs charged to the DiscOps and passed

through to government contracts. (Mot. for Partial Summ. J., Jan. 9, 2009 [Dkt. No. 25-2] at 10.)

The government contended that Lockheed’s recovery of costs under CERCLA, for which the

government-as-client (primarily the DOD) had already indirectly paid pursuant to the DOSA,

would result in a double recovery – a violation of both the DOSA and CERCLA. See 42 U.S.C.

§ 9614(b).

On September 29, 2009, Judge Robertson rejected the government’s legal arguments on

several grounds. See Lockheed Martin Corp. v. United States, 664 F. Supp. 2d 14 (D.D.C.

2009). Judge Robertson first distinguished the bar on double recovery in CERCLA § 114(b)

from the present case. Section 114(b) states that “[a]ny person who receives compensation for

removal costs or damages or claims pursuant to any other Federal or State law shall be precluded

from receiving compensation for the same removal costs or damages or claims as provided in

[CERCLA].” 42 U.S.C. § 9614(b). Judge Robertson faulted the government’s argument for

failing appreciate the distinction between the “government-as-client” and the “government-as-

PRP.” To Judge Robertson, the government-as-client’s indirect cost payments under

government contracts were not “compensation” for the government’s liability in the same

manner as direct payments from an insurer, private PRP, or government PRP. Lockheed Martin

Corp., 664 F. Supp. 2d at 19. Judge Robertson further noted that the DOSA “explicitly states

that it ‘does not settle claims, if any, arising under CERCLA.’” Id.

Judge Robertson also distinguished the double recovery cases cited by the government.

In those cases, plaintiffs sought to recover response costs which they had already recovered from

34
other PRPs (through settlement or state-law versions of CERCLA) or response costs which their

insurers had paid for directly. Id. The courts in those cases applied CERCLA section 114(b) in

order to prevent plaintiffs from receiving “a windfall from its environmental cleanup.” Id.

Judge Robertson concluded that there would be “no windfall” in this case because Lockheed was

required under both the DOSA and the FAR to credit any CERCLA recovery from the

government to the DiscOps Pool. Id. at 19-20. Thus, “any CERCLA recovery from the

government would lead to a commensurate reduction in the [DiscOps Pool] that Lockheed could

charge as indirect costs on its government contracts. From a monetary standpoint, Lockheed

would be back where it started.” Id. at 19.

Judge Robertson also emphasized the “important implications for both parties” –

Lockheed would “improve its competitive position” and the government-as-client would benefit

from decreased contract prices. Id. at 20. Accordingly, Judge Robertson concluded:

If Lockheed is only partially liable for the response costs it is incurring at the Site,
it should not have to include all its response costs in the [DiscOps] Pool. The
ruling on the instant motions ensures that Lockheed may recover separately under
CERCLA from the government-as-PRP (if the suit ends with a recovery),
burdened in its dealings with the government-as-client only by those costs for
which it is actually liable.

Id.

On October 22, 2009, the government moved for reconsideration, inter alia, on the

grounds that Judge Robertson improperly imported requirements, such as the potential for

“windfall,” into the language of CERCLA § 114(b) and failed to address the argument that

Lockheed’s indirect recovery of response costs through government contracts could be

considered as an equitable factor during allocation of response costs under CERCLA § 113(f)(1).

(Mot. for Reconsideration, Oct. 22, 2009 [Dkt. No. 39] at 2-3.)

35
 Judge Robertson denied the motion on February 18, 2010. (Mem. Order, Feb. 18, 2010

[Dkt. No. 43] at 2-3.) While remaining steadfast in his denial of summary judgment, Judge

Robertson acknowledged that some of the government’s positions, “including those about the

equitable considerations under Section 113(f)(1) and burden to taxpayers, may be relevant to

allocation determinations that might lie ahead in this litigation.” (Id. at 3.)29 Judge Robertson

therefore left open the door to the equitable double recovery and taxpayer burden arguments that

are now at the center of the allocation phase of the case.30

C. Trial on equitable allocation

The parties engaged in extensive factual and expert discovery over the next four years.

On September 26, 2013, the Court bifurcated the pending trial into two phases: (1) the parties’

liability as PRPs under CERCLA § 107(a) and the equitable allocation of response costs under

CERCLA § 113(f); and (2) the “accurate accounting” of response costs. (Order, Sept. 26, 2013

[Dkt. No. 97] at 2.) The Court scheduled a two-week bench trial for Phase I to begin on

February 10, 2014. (Id.)

Leading up to the trial, the parties stipulated that they were both PRPs liable under

CERCLA § 107(a) for the response costs that Lockheed had incurred at the Sites and that the

Court should allocate liability for the response costs between them according to equitable

principles under CERCLA § 113(f)(1). (Stipulation on Liability, Dec. 20, 2013 [Dkt. No. 103] at

2-3.) The parties also stipulated that, as required under CERCLA, Lockheed had incurred at

least one dollar of “necessary costs of response” at the Sites in a manner “consistent with the

29
In Raytheon (cited supra n. 21), the government raised an affirmative defense analogous to that
rejected by Judge Robertson in this case. 2007 WL 4300221, at *2-3. As Judge Robertson did, the judge
in Raytheon concluded that Raytheon’s indirect recovery of response costs through U.S.-government
contracts could be considered as an equitable matter. Id. at *3. However, as noted, the Raytheon case
settled before the issue of allocation was reached.
30
Following Judge Robertson’s retirement, the case was reassigned to this Court on June 3, 2010.

36
National Contingency Plan.” (Stipulation on Response Costs, Dec. 20, 2013 [Dkt. No. 104] at 1-

2 (quoting 42 U.S.C. § 9607(a)(4)(B)).)

Given the stipulations on liability, the Phase I bench trial concerned only the equitable

allocation of liability under CERCLA § 113(f)(1). The Court presided over the Phase I bench

trial over twelve days beginning February 10 and ending on March 14, 2014. Through no fault

of the parties or their counsel, the evidence at the trial was both voluminous and inconclusive in

many significant respects. Arguably, the most useful testimony came from the only two

witnesses who had been on the ground during LPC’s operations at the Sites, but even those

witnesses had limited recollections. (See, e.g., Trial Tr. at 95 (Oppliger); id. at 1374 (Dull).)

Otherwise, the record suffered from the many shortcomings inherent in CERCLA actions:

hundreds of missing contracts31 and significant gaps in the documents; a lack of

contemporaneous accounts of waste disposal operations at the Sites; almost no fact witnesses,

thus requiring extensive reliance on “historic” depositions, themselves taken some twenty to

thirty years after operations at the Sites; and contamination that, at least at the Redlands facility,

had migrated miles from the location of original disposals. In view of these evidentiary

constraints, the parties understandably relied almost exclusively on experts. The experts, who

provided direct testimony by declaration and were subject to cross examination and redirect at

trial, predictably presented diametrically opposed opinions regarding almost every important

subject, often relying on assumptions that were not sufficiently provable for the Court to draw

any reliable conclusions.

31
Out of the estimated hundreds or thousands of contracts relevant to this action, Lockheed’s
contracting expert had reviewed “perhaps three” and could not confirm that any contract he had reviewed
was actually complete. (Trial Tr. at 148-49 (Johnson).) Without contracts themselves, the experts had to
rely on Master Project Orders and documentation of contract revisions to attempt to piece together the
“very limited database” of contracts available in this case. (Id. at 149.)

37
 The testimony and evidence presented at trial addressed five somewhat-overlapping

issues, and as could be expected, on each issue save one, the Court had a battle of the experts.

First, Dr. Peter Roman, Lockheed’s expert historian, provided a mostly undisputed account of

the history of the solid rocket propellant industry and LPC’s government contracts. Second, the

parties presented evidence of the degree of government activity at and control over the Sites with

respect to government specifications, safety manuals, project reviews, and safety inspections.

Each party also called a fact witness to describe the government’s presence at the Sites. Mr.

Gerald Oppliger testified about his experiences working as an engineering analyst for LPC

designing rockets for several government contracts, including SRAM, from 1962 to 1971. (Trial

Tr. at 74-77 (Oppliger).) Mr. Peter Dull testified about his work on SRAM as a propulsion

engineer for the SPO from the program’s inception to Lockheed’s delivery of the rockets. (Id. at

1337-38 (Dull).) Third, the parties put on dueling experts relating to the subject of government

contracting – Mr. Richard Johnson (for Lockheed) and Mr. James Nagel (for the government) –

who provided opinions on whether and to what extent the government, through contracting

regulations, owned the TCE and AP at the Sites.

Fourth, the parties spent an excessively large amount of trial time on the issues of

contamination and, in particular, which of LPC’s activities caused it. Lockheed called Thomas

Blackman to provide foundational testimony regarding the contamination and response actions at

the Redlands facility. Both parties then put on multiple experts. Lockheed’s primary expert, Dr.

Stanley Feenstra, a hydrogeologist, opined on the probable sources of TCE (vapor degreaser) and

perchlorate contamination (burn pits) at the Sites and the level of care with which LPC handled

its solvent and propellant wastes. Dr. Tod Delaney, an environmental engineer, provided a

rebuttal opinion to the government expert opinion from Thomas Cain, a chemical and process

38
engineer, regarding the sources of perchlorate contamination from the southern half of the

Redlands facility. The government’s primary expert, Dr. Robert Sterrett, a hydrogeologist,

opined primarily on probable sources of TCE (employee dumping) and perchlorate

contamination (burn pits and grinder part washing) at the Sites, while Mr. David Bauer, a

chemist, opined on the level of care with which LPC handled its solvent and propellant wastes.

The government also called Ms. Mary Sitton as an aerial photography expert who analyzed

historical aerial photographs of the Sites for evidence of sources of pollution. Of the myriad,

though incomplete, sources available to them, the experts relied primarily on LPC’s process

specifications, engineering diagrams of buildings and equipment, and the historic deposition

testimony of dozens of former LPC workers. The experts also relied on their own personal

experiences and general industrial and military practices at the time. Nonetheless, the ultimate

expert opinions often amounted to no more than exercises in reasoned speculation.

Finally, the parties spent a significant amount of trial on the accounting and economic

issues that make this case unique. Lockheed called Mr. Robert Gatchel, Lockheed’s current vice

president of government finance, to describe the implementation of the DOSA and Lockheed’s

government contracting procedures. Each party also called two experts to address the DOSA,

double recovery, and Lockheed’s economic benefit from a CERCLA recovery in this case. For

the government, Mr. Wiley Wright opined on the general function of DOSA, and Dr. Joan Meyer

presented an economic model demonstrating the benefit to Lockheed from various CERCLA

recovery scenarios. Lockheed’s experts, Mr. Mark Kiefer and Mr. Rodney Mateer, criticized the

assumptions and methodologies of the government experts, but provided no competing economic

model to substantiate their position that Lockheed would not economically benefit from a

CERCLA recovery in this case.

39
 The difficulty of an incomplete record and heavy reliance on expert testimony was

compounded by the lack of analogous CERCLA cases.32 While many CERCLA actions have

been brought by government contractors against the U.S. government, only a few appear to have

reached the allocation stage. And none of those address the key issue in this case: whether the

fact that the government contractor has been indirectly recovering its response costs from the

U.S. government-as-client through U.S.-government contracts should, as an equitable

consideration, reduce its recovery from the U.S. government-as-PRP under CERCLA.

LEGAL FRAMEWORK

Congress enacted CERCLA “in response to the serious environmental and health risks

posed by industrial pollution.” United States v. Bestfoods, 524 U.S. 51, 55 (1998). The statute

“was designed to promote the timely cleanup of hazardous waste sites and to ensure that the

costs of such cleanup efforts were borne by those responsible for the contamination.” Burlington

N. & Santa Fe Ry. Co. v. United States, 556 U.S. 599, 602 (2009) (internal quotation marks

omitted). By requiring responsible parties to pay for cleanup efforts, CERCLA also ensures that

“the taxpayers [are] not required to shoulder the financial burden of a nationwide cleanup.” B.F.

Goodrich Co. v. Murtha, 958 F.2d 1192, 1198 (2d Cir. 1992).

In furtherance of these goals, CERCLA allows private parties to recover the costs of

cleaning up hazardous wastes from several broad categories of PRPs. 42 U.S.C. § 9607(a)(1)-

(4).33 Liability under these provisions is strict and, by default, joint and several. PCS Nitrogen

32
At the Court’s request, the parties filed several helpful memoranda on legal issues, including
operator and arranger liability, equitable indemnification, double recovery, and prejudgment interest.
33
A private-party plaintiff establishes a prima facie case for cost recovery under CERCLA by
establishing that (1) the defendant is a PRP; (2) the site constitutes a “facility”; (3) a “release” or a
threatened release of hazardous substances exists at the “facility”; (4) the plaintiff has incurred costs
responding to the release or threatened release of hazardous substances (“response costs”); and (5) the
response costs conform to the National Contingency Plan. PCS Nitrogen Inc. v. Ashley II of Charleston

40
Inc. v. Ashley II of Charleston LLC, 714 F.3d 161, 168 (4th Cir. 2013). Relevant to this action,

PRPs include any past “owner” or “operator” and any “arranger.” See 42 U.S.C. § 9607(a)(2)-

(3).34

Under CERCLA, a person is liable as a past “owner” or “operator” if he “at the time of

disposal of any hazardous substance owned or operated any facility at which such hazardous

substances were disposed of.” Id. § 9607(a)(2). The phrase “owner or operator” is unhelpfully

defined “only by tautology . . . as ‘any person owning or operating’ a facility.” Bestfoods, 524

U.S. at 56 (quoting 42 U.S.C. § 9601(20)(A)(ii)). Courts consider “owner” and “operator”

liabilities in the disjunctive. Commander Oil Corp. v. Barlo Equip. Corp., 215 F.3d 321, 328 (2d

Cir. 2000); cf. Bestfoods, 524 U.S. at 64. In this regard, past owner liability is fairly simple, and

premised on the ownership, whether de facto or de jure, of a “‘facility’ at the time of the disposal

of a hazardous substance.” PCS Nitrogen, 714 F.3d at 172 (citing 42 U.S.C. § 9607(a)(2)); see

Commander Oil, 215 F.3d at 331-32.35 Notably, the ownership of the hazardous substances

disposed of at a given facility is irrelevant to ownership liability. See 42 U.S.C. § 9607(a)(2).

LLC, 714 F.3d 161, 167-68 (4th Cir. 2013); see also 42 U.S.C. §§ 9601(9), (22), 9607(a). Each of these
requirements is stipulated to or otherwise undisputed in this case. (See Stipulation on Liability, Dec. 20,
2013 [Dkt. No. 103]; Stipulation on Response Costs, Dec. 20, 2013 [Dkt. No. 104].)
34
Because both parties have stipulated to liability under CERCLA § 107(a), the Court need not
determine under which provision(s) each party falls. Nonetheless, the Court does find it useful to
consider the breadth (and limits) of types of PRP liability when balancing the equities under CERCLA §
113(f).
35
Under CERCLA, the term “facility” has two meanings. First, the term “facility” serves as a
catch-all for almost any original source from which a disposal might have occurred. See 42 U.S.C. §
9601(9)(A) (defining “facility” as “any building, structure, installation, equipment, pipe or pipeline
(including any pipe into a sewer or publicly owned treatment works), well, pit, pond, lagoon,
impoundment, ditch, landfill, storage container, motor vehicle, rolling stock, or aircraft”). Second, the
term “facility” includes “any site or area where a hazardous substance has been deposited, stored,
disposed of, or placed, or otherwise come to be located.” Id. § 9601(9)(B). Courts have interpreted this
second meaning to cover, at a minimum, “‘the bounds of the contamination.’” See PCS Nitrogen, 714
F.3d at 178 (quoting United States v. Twp. of Brighton, 153 F.3d 307, 313 (6th Cir. 1998)). Thus, “[t]here
may be several ‘facilities’ at a site for purposes of CERCLA, including separately owned ‘equipment’

41
 Operator liability is more nuanced, but Bestfoods provides some helpful guidance. There,

the Supreme Court clarified that when defining “operator” liability under CERCLA, Congress

“obviously meant something more than mere mechanical activation of pumps and valves” and

intended liability to extend to those who “exercise . . . direction over the facility’s activities.”

524 U.S. at 71. On this basis, the Court concluded that an operator under CERCLA must

“manage, direct, or conduct operations specifically related to pollution, that is, operations having

to do with the leakage or disposal of hazardous waste, or decisions about compliance with

environmental regulations.” Id. at 66-67. The definition of “operator” in Bestfoods “clearly

requires actual participation, not merely the potential to do so.” City of Wichita, Kansas v.

Trustees of APCO Oil Corp. Liquidating Trust, 306 F. Supp. 2d 1040, 1054 (D. Kan. 2003); see

also United States v. Twp. of Brighton, 153 F.3d 307, 314 (6th Cir. 1998) (describing Bestfoods

as “highlight[ing] the importance of establishing some actual control by a putative operator”).

Although “actual participation” in this sense “does not require a finding that the [defendant]

directly participated in the day-to-day activities” at the facility, see United States v. Kayser-Roth

Corp., 103 F. Supp. 2d 74, 82 (D.R.I. 2000), Bestfoods requires that an operator “make the

relevant decisions” regarding the disposal of hazardous wastes “on a frequent, typically day-to-

day basis.” City of Wichita, 306 F. Supp. 2d at 1055 (collecting cases).

Arranger liability generally “prevents owners of hazardous waste from avoiding liability

under CERCLA by transferring ownership of the waste to another party for the purposes of

disposal.” Wilson Rd. Dev. Corp. v. Fronabarger Concreters, Inc., -- F. Supp. 2d. ---, 2013 WL

4875071, at *6 (E.D. Mo. Sept. 11, 2013). CERCLA defines “arranger” as “any person who by

within a larger facility,” Am. Int’l Specialty Lines Ins. Co. v. United States, 2010 WL 2635768, at *21
(C.D. Cal. June 30, 2010), as well as the location where the contamination has migrated over time. In this
regard, the “facilities” in this case include not only any historic sources of contamination at the Redlands,
Potrero Canyon, and LaBorde Canyon properties, but also the area of the Redlands TCE and perchlorate
plumes.

42
contract, agreement, or otherwise arranged for disposal or treatment, or arranged with a

transporter for transport for disposal or treatment, of hazardous substances owned or possessed

by such person, by any other party or entity, at any facility or incineration vessel owned or

operated by another party or entity and containing such hazardous substances.” 42 U.S.C. §

9607(a)(3).

Arranger liability requires ownership or possession of hazardous substances, but “cannot

be imposed on that basis alone.” Morton Int’l, Inc. v. A.E. Staley Mfg. Co., 343 F.3d 669, 679

(3d Cir. 2003).36 Instead, as the Supreme Court clarified in Burlington Northern, “whether an

entity is an arranger requires a fact-intensive inquiry that looks beyond the parties’

characterization of the transaction as a ‘disposal’ or a ‘sale’ and seeks to discern whether the

arrangement was one Congress intended to fall within the scope of CERCLA’s strict-liability

provisions.” 556 U.S. at 610. Under the plain language of the statute, a person is only an

arranger if he “plans for” or “takes intentional steps to dispose of a hazardous substance” he

owned or possessed. Id. at 611. Accordingly, although knowledge that a hazardous substance

will be disposed of “may provide evidence of the entity’s intent to dispose of its hazardous

wastes, knowledge alone is insufficient to prove that an entity ‘planned for’ the disposal,

particularly when the disposal occurs as a peripheral result of a legitimate” transaction, such as

the sale of a useful product. Id. at 612. Courts oftentimes consider – as with operator liability –

a party’s exercise of control over the disposal of hazardous substances as “a crucial element of

the [fact-specific] determination of whether a party is an arranger.” See United States v. Shell

Oil Co., 294 F.3d 1045, 1055 (9th Cir. 2002).

36
But see United States v. Iron Mountain Mines, Inc., 881 F. Supp. 1432, 1451 (E.D. Cal. 1995)
(collecting cases where courts “impose[d] arranger liability on parties who did not literally own or
physically possess hazardous wastes,” but noting that “in each of these cases the party either was the
source of the pollution or managed its disposal by the arranger.”).

43
 A plaintiff who is also a PRP theoretically may avoid CERCLA liability altogether by

imposing joint and several liability on a PRP-defendant under CERCLA § 107(a). However, the

PRP-defendant can “blunt any inequitable distribution of costs by filing a § 113(f) counterclaim”

against the PRP-plaintiff. United States v. Atl. Research Corp., 551 U.S. 128, 140 (2007). In

such instances – as here – a court must determine the allocation of liabilities between the PRPs

pursuant to CERCLA § 113(f)(1).

CERCLA § 113(f)(1) states that “[i]n resolving contribution claims, the court may

allocate response costs among liable parties using such equitable factors as the court determines

are appropriate.” 42 U.S.C. § 9613(f)(1). Courts have universally held that “[t]his plain

language grants a court significant discretion to choose which factors to consider in determining

equitable allocation of liability.” PCS Nitrogen Inc., 714 F.3d at 186; see also Boeing Co. v.

Cascade Corp., 207 F.3d 1177, 1187 (9th Cir. 2000). “[I]n any given case, a court may consider

several factors, a few factors, or only one determining factor . . . depending on the totality of

circumstances presented to the court.” Envtl. Transp. Sys., Inc. v. ENSCO, Inc., 969 F.2d 503,

509 (7th Cir. 1992).

“Courts generally trot out two lists of factors when considering allocation under

CERCLA.” Yankee Gas Servs. Co. v. UGI Utilities, Inc., 852 F. Supp. 2d 229, 247 (D. Conn.

2012). The first, the so-called “Gore Factors,” find their source in the legislative history (and

unsuccessful amendment) of CERCLA by then-Representative Al Gore. See Boeing, 207 F.3d at

1187. The “Gore Factors” include:

[1.] the ability of the parties to demonstrate that their contribution to a discharge, release
or disposal of a hazardous waste can be distinguished;

[2.] the amount of the hazardous waste involved;

[3.] the degree of toxicity of the hazardous waste involved;

44
 [4.] the degree of involvement by the parties in the generation, transportation, treatment,
storage, or disposal of the hazardous waste;

[5.] the degree of care exercised by the parties with respect to the hazardous waste
concerned, taking into account the characteristics of such hazardous waste; and

[6.] the degree of cooperation by the parties with Federal, State or local officials to
prevent any harm to the public health or the environment.

Envtl. Transp. Sys., 969 F.2d at 508 (quoting United States v. A & F Materials Co., 578 F. Supp.

1249, 1256 (S.D. Ill. 1984)).

Courts also often invoke the so-called “Torres Factors,” named after the “critical factors”

enumerated by then-Judge Torres:

1. The extent to which cleanup costs are attributable to wastes for which a party
is responsible.

2. The party’s level of culpability.

3. The degree to which the party benefitted from disposal of the waste.

4. The party’s ability to pay its share of the cost.

United States v. Davis, 31 F.Supp.2d 45, 63 (D.R.I. 1998).37

Given the broad discretion granted in CERCLA § 113(f)(1), courts also look beyond the

Gore and Torres factors when equitably allocating response costs. See, e.g., Am. Int’l Specialty

Lines Ins. Co. v. United States (AISLIC II), 2013 WL 135405, at *9 (C.D. Cal. Jan. 9, 2013). Of

relevance to this case, courts have also considered the following factors under CERCLA §

113(f)(1):

37
Some scholars have argued that while “[t]he Gore factors are most relevant in academic and
theoretical analysis of the way Superfund liabilities should be allocated. . . . in the real world Judge
Torres’ list of four critical factors often provides the basis upon which Superfund allocations are made.”
Robert P. Dahlquist, Making Sense of Superfund Allocation Decisions: The Rough Justice of Negotiated
and Litigated Allocations, 31 ENVTL. L. REP. 11098, 11099 (2001) (emphasis added); see also Yankee
Gas, 852 F. Supp. 2d at 247.

45
 1. The “knowledge and/or acquiescence of the parties in the contaminating
activities.” Weyerhaeuser Co. v. Koppers Co., 771 F.Supp. 1420, 1426 (D. Md.
1991).

2. The value of the contamination-causing activities to furthering the
government’s national defense efforts. Cadillac Fairview/Cal., Inc. v. Dow
Chemical Co., 299 F.3d 1019, 1026 (9th Cir. 2002); Shell Oil, 294 F.3d at 1060.

3. The existence of an indemnification agreement demonstrating “the parties’
parties’ intent to allocate liability among themselves.” Halliburton Energy Servs.,
Inc. v. NL Indus., 648 F. Supp. 2d 840, 863 (S.D. Tex. 2009); see also Beazer E.,
Inc. v. Mead Corp., 412 F.3d 429, 447 (3d Cir. 2005).

4. “The financial benefit that a party may gain from remediation of a site.” Litgo
New Jersey, Inc. v. Martin, 2011 WL 65933, at *9 (D.N.J. Jan. 7, 2011); see also
City of Wichita, 306 F. Supp. 2d at 1101.

5. The potential for windfall “double recoveries” by a plaintiff. See, e.g., Litgo
N.J. Inc. v. Comm’r N.J. Dep’t of Envtl. Prot., 725 F.3d 369, 391 (3d Cir. 2013);
Friedland v. TIC-The Indus. Co., 566 F.3d 1203, 1207 (10th Cir. 2009).

6. The potential that a plaintiff might “make a profit on the contamination” at the
expense of another PRP. See Vine St., LLC v. Keeling ex rel. Estate of Keeling,
460 F. Supp. 2d 728, 765 (E.D. Tex. 2006).

7. CERCLA’s intent that “‘responsible parties, rather than taxpayers, bear the
costs’” of cleanup. Yankee Gas, 852 F. Supp. 2d at 256 (quoting Marsh v.
Rosenbloom, 499 F.3d 165, 182 (2d Cir. 2007) (emphasis added)).

With these factors in mind, the Court now turns to the ultimate question before it: the equitable

allocation between Lockheed and the government of past and future response costs at the Sites.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

To reach an equitable allocation for the Sites, the Court proceeds in three steps. First, the

Court determines, to the extent possible, the sources of contamination at the Sites. From there,

the Court bifurcates its equitable allocation analysis to consider (1) the familiar exercise of

allocating response costs among parties and (2) the novel issue of whether Lockheed’s indirect

46
recovery of those response costs through U.S-government contracts should affect the otherwise-

equitable allocation.

I. SOURCES OF CONTAMINATION AT THE SITES

Although in many cases courts can find that a particular facility, party, or disposal action

was responsible for a specific proportion of contamination at a site, see, e.g., AlliedSignal, Inc. v.

Amcast Int’l Corp., 177 F. Supp. 2d 713, 724 (S.D. Ohio 2001), the Court is unable to do so here

given the gaping holes in the record. Instead, the Court must limit itself to determining the

probable sources of perchlorate and TCE contamination at the three Sites.

A. Redlands facility

1. Perchlorate

The parties agree that a significant proportion of the perchlorate contamination in the

Redlands perchlorate plume originated from the burning of propellant wastes in the burn pits in

the northern portion of the Redlands facility. (Feenstra Decl. ¶¶ 123, 127; Sterrett Decl. ¶¶ 17,

136-41.) As described by Dr. Sterrett, the “burn pits were essentially trenches dug in the bare

ground.” (Sterrett Decl. ¶ 136.) LPC employees burned propellant wastes, containing waste AP

in liquid and solid form, on a daily basis. (USX963.0086 (Speer Decl.).)

The operation of the burn pits contributed perchlorate to the Redlands perchlorate plume

because, unbeknownst to anyone at the time, the AP disposed of in the pits was not completely

destroyed by burning. (Trial Tr. at 730 (Feenstra); see PX1224 at 10.) Any AP residue

remaining in the pits after burning could dissolve in a solvent (including water) and infiltrate

through the porous soil down into the groundwater. (See Feenstra Decl. ¶ 126; PX1224 at 10).

Dissolution of residue AP occurred during rainfall events (Feenstra Decl. ¶ 126) and when LPC

47
operators “thoroughly soaked with water” the burn pits when used multiple times in a day. (See

PX0007 § 1505(d) (1968 DOD Safety Manual).)

The presence of perchlorate in groundwater wells in the southern portion of the Redlands

facility also suggests the existence of perchlorate sources unassociated with the burn pits.

(Feenstra Decl. ¶¶ 122, 148, 153.) The parties vigorously dispute the southern source of the

perchlorate at the facility. The Court finds that there were three probable sources.38

First, beginning as early as November 1958, liquid solvent and AP wastes drained onto

the ground outside of Building 52, where LPC mixed propellants.39 (Sterrett Decl. ¶¶ 146-47.)

At that time, a pipe intended to connect Building 52 to Evaporation Pit 61 only “drained toward”

the pit, even though LPC recognized that it “should be piped right into [the] pit.” (USX43.0002

(emphasis in original).) An October 16, 1959 aerial photograph showed staining and liquid

38
Contrary to Dr. Feenstra’s opinion (Feenstra Decl. ¶¶ 148-49), neither fugitive dust from AP
handling and processing nor leakage from evaporation pits were probable sources of perchlorate in the
Redlands plume. While some fugitive dust inevitably escaped the grinding buildings at the Redlands
facility – particularly in the 1950s (see USX1160.0006, .0012-14) prior to the installation of advanced
baghouse technologies (see USX1117) – prevailing winds would have spread the light, small AP dust
particles far and wide before they settled onto the ground. (Cain Decl. ¶¶ 45-47.) Spread so far, it is
unlikely the AP dust, once settled and dissolved by rainfall, could develop a wetted pathway necessary to
migrate through the soil to the groundwater. Similarly, there is no evidence that the evaporation pits at
the Redlands facility ever leaked (Feenstra Decl. ¶ 150; see also Trial Tr. at 1252-53 (Sterrett)), nor any
evidence that AP-laden wastewater from the “soaking out” of rocket motors in the evaporation pits was
disposed of on the ground. (See Feenstra Decl. ¶ 152.)
Nor, contrary to Dr. Sterrett and Mr. Cain’s opinion (Sterrett Decl. ¶¶ 142-44; Cain Decl. ¶¶ 108-
15), was Building 114’s settling basin a probable source of perchlorate in the Redlands plume. LPC used
Building 114 as a propellant research laboratory from which it discharged approximately 2500 gallons of
wastewater into a dedicated settling basin. (USX781.0001.) Each week during dry weather, LPC
pumped the water out of the settling basin “to an underground drain (storm sewer) catch basin” that
“traverse[d] . . . into an open swale.” (Id.) From that point, the water ran “on open ground” and
infiltrated “due to the great percolation ability of the soil.” (Id.) Although LPC’s research laboratory
used AP (Trial Tr. at 1282-83 (Cain)), the effluent pumped from the settling basin was actually lower in
total dissolved solids (which would include AP) than the average groundwater in the area. (See id. at
1256-58, 1261 (Sterrett).) Based on this conflicting evidence, the Court cannot conclude that Building
114 was a probable source of perchlorate for the Redlands plume.
39
Because LPC used cyclohexane, and not TCE, as a solvent in Building 52 in the late 1950s
(USX43.0002), there is no evidence that this unconnected pipe was a source of the Redlands TCE plume.

48
within a drainage channel consistent with drainage from Building 52 toward rather than into

Evaporation Pit 61. (Sitton Decl. ¶ 34; USX670.) Although it is unclear precisely when LPC

addressed the problem it first recognized in late 1958,40 aerial photography confirms by at least

April 1963 a pipe did connect Building 52 to Evaporation Pit 61. (See Trial Tr. at 1434 (Sitton).)

Second, AP-laden wastewater percolated into the ground outside of several buildings

when LPC employees washed down the interiors of those buildings with hoses and failed to

collect the water or otherwise direct it to evaporation pits. (Sterrett Decl. ¶¶ 148-53; Cain Decl.

¶ 102; see Feenstra Decl. ¶ 151.) LPC employees washed down the interiors of grinding

buildings at the Redlands facility on a weekly basis. (See USX808.0022 (Building 77 Process

Specification); USX873 at 126-27 (Caldwell Dep.); USX973 at 157-58 (Spencer Dep.).) LPC

employees also washed down the interiors of the mixing buildings, although with less frequency.

(USX895 at 103-05 (Eastman Dep.).) For the portions of those buildings that did not have

drains, the AP-laden wastewater “ran out of the building” (id. at 101) and onto the ground

(USX994 at 115-16 (White Dep.), where it percolated into the soil and ultimately into the

groundwater. (Sterrett Decl. ¶ 153.)

The final – and most significant – source of perchlorate from the southern portion of the

Redlands facility was the washing of AP off of grinder parts and dust collection bags into the

south sump outside of Building 77. LPC operated Building 77 as an AP grinding facility. In

1962, LPC issued a manufacturing process standard – Standard 00COO5 – addressing cleanup

procedures for oxidizer grinding operations at Building 77. (USX32.) The Process Standard

instructed LPC employees after each grind to “flush[] with water” certain parts of the grinder “to

remove contamination.” (Id. at .0003.) The standard further specified that “[t]he flushing will

40
A July 11, 1960 letter from LPC to the Los Angeles Ordnance District suggests that the
unconnected pipe probably was addressed by that point. (See PX1046 at 2.)

49
be performed at the faucet and sump outside the lower level of Bldg. 77. Do not use water inside

the building.” (Id. (emphasis in original).) The same process was mandated for the washing of

the AP dust collection bags. (Id.; see also USX807.0031-33.)

Based on this process standard, both parties agree that after 1962, LPC washed AP dust

off of the grinder parts and bags outside of Building 77. However, the parties dispute into which

of the sumps on the west side of Building 77 the AP-contaminated washwater flowed. Mr. Cain

opined that LPC employees washed the grinder parts and bags on the concrete slab outside of

Building 77 and the washwater flowed naturally into the south sump, where it then percolated

into the ground through a drain or was pumped to the south to drain naturally on the bare ground

downhill and away from Building 77. (Cain Decl. ¶¶ 50-59, 79; see also Sterrett Decl. ¶¶ 125-

35.) In rebuttal, Mr. Delaney, Lockheed’s expert, opined that LPC employees washed the parts

directly over and into the north sump, which pumped the washwater to an evaporation pit.

(Delaney Decl. ¶¶ 38-50; see also Feenstra Decl. ¶¶ 109-21.)

The Court credits Mr. Cain’s opinion that the washwater from cleaning the grinder parts

and bags flowed to the south sump and ultimately percolated into the soil and migrated to the

groundwater. LPC’s process standards provided no clear instructions on precisely how to wash

the grinder parts and bags at the “at the faucet and sump” outside of Building 77 (see USX32 §

1.3),41 and the record lacks any evidence on how the workers actually performed the task. Mr.

Delaney opined that LPC employees must have washed the parts and bags directly over the 3-

foot by 3-foot north sump because they would have known that washing the parts over concrete

would have presented a fire and explosion hazard if the AP recrystallized before reaching the

south sump. (Delaney Decl. ¶¶ 28, 43-44.) Although Mr. Delaney’s concern has a scientific

41
Generally, wastewater containing AP was to be treated the same as waste propellant. (See, e.g.,
PX1023 § 6.31.1; PX1043 §100.6.5.)

50
basis (cf. Cain Decl. ¶ 49), he vastly overstates the safety consciousness of LPC’s employees.

Former employee Donald Eastman stated in a deposition that, when the workers washed AP-

laden wastewater out of grinding buildings, there was no effort to contain the water because “it

wasn’t considered a problem.” (USX895 at 101 (Eastman Dep.).) Likewise, the numerous other

instances of employees ignoring LPC-established safety protocols (see infra I.A.2) undermine

Mr. Delaney’s suggestion that those same employees would develop at Building 77 their own,

more burdensome, cleaning protocols in the name of safety. Ultimately, there is little evidence

to support the conclusion that the workers would on a near-daily basis spend the time and effort

to remove the heavy lid over the north sump and wash the parts and bags directly over that sump

when they had available the much easier process of washing the parts and bags over the concrete

pad and allowing the wastewater to flow naturally into the south sump. (See Trial Tr. at 1298-99

(Cain).)42

Moreover, two pieces of evidence support Mr. Cain’s opinion that washwater from

cleaning the grinder parts and bags flowed into the south sump. First, George Nelson White, a

maintenance mechanic at LPC from 1961 to 1975, stated that when the south sump reached

capacity, “it pumped the water up to the top of the dike out into the rocks. They’re all oxidizers

there.” (USX994 at 116 (White Dep.).) Second, and perhaps most persuasively, an April 16,

1966 aerial photograph shows an area of staining or standing liquid at the end of the drainage

channel down-gradient from the south sump fallout. (See Trial Tr. at 1412-13 (Sitton); Sitton

42
When Building 77 and its sumps were constructed, the washing of grinder parts occurred inside
the building. Because the top floor had an internal drain to the north sump and, ultimately, to an
evaporation pit, releases of AP-laden washwater onto the ground were minimal in the early years. (See
Trial Tr. at 1301 (Cain).) It was not until 1962, when, due to humidity concerns, LPC changed its process
specification to disallow washing inside Building 77. (Trial Tr. at 1300 (Cain); see also USX32 § 1.3).)

51
Decl. ¶ 41; USX672.)43 Because there had been no precipitation in the fifteen days prior to the

photograph (Sitton Decl. ¶ 42), the south sump would have only filled up and pumped water to

the outfall if LPC employees were washing grinder parts and bags into the south sump.

Based on this evidence, the Court finds that AP-laden washwater from LPC employees’

regular washing of grinder parts and bags outside of a Building 77 flowed into the south sump

and contributed significantly to the Redlands perchlorate plume.44

2. Trichloroethylene

The parties disagree vehemently as to the identity of the source of the Redlands TCE

plume. The government argues that the TCE originated primarily from LPC employees’

intentional dumping of solvents onto the bare ground at various locations at the Redlands

facility. (Trial Tr. at 897-98 (gov’t opening); Sterrett Decl. ¶¶ 54-75.) In contrast, Lockheed

asserts that the TCE primarily originated from discharges from a solvent-water separator

attached to the vapor degreaser in Building 91. (Trial Tr. at 42-43 (Lockheed opening); Feenstra

Decl. ¶¶ 63-95.)45 Having considered the evidence on these sources, the Court finds that both

probably contributed substantially to the Redlands TCE plume.

43
Lockheed suggested, based on an engineering drawing of Building 77 (PX837), that the water
from the south sump flowed northeast, rather than southwest, and thus the staining or standing liquid Ms.
Sitton identified could not have originated in the south sump. (See Trial Tr. at 1410 (Sitton).) However,
close inspection of the engineering drawing reveals that the “existing drainage” flowing to the northeast
had been “close[d] off” by the construction of a dyke. (PX837.) With the “existing drainage” to the
northeast closed off, the “sump drains over [the] dyke” (id.) to the southwest, consistent with Ms. Sitton’s
analysis of the aerial photography and the general topography at the Redlands facility.
44
Mr. Cain estimated that up to 20,000 pounds of the between 40,000 and 60,000 pounds of
perchlorate in the Redlands plume may have originated from the washing of AP dust collection bags at
Building 77. (Cain Decl. ¶ 92; see Sterrett Decl. ¶ 51.) The Court finds it unnecessary to decide on the
validity of his estimations; it suffices to say that the washing of grinder parts and bags outside of Building
77 contributed significantly to the Redlands perchlorate plume and thus cleanup costs at the Redlands
facility.
45
Although in his declaration Dr. Sterrett opined that TCE could have leaked into the ground
through cracks in the evaporation pits (Sterrett Decl. ¶ 78), he admitted on cross-examination that he had

52
 First, the Court finds ample evidence that LPC employees poured TCE on the bare

ground and that those actions were a probable source of the Redlands TCE plume. (See Sterrett

Decl. ¶ 54.) Repeated pouring of small amounts of a liquid onto a porous and permeable soil can

create a “wetted pathway” by which a liquid’s chance of evaporation in the soil strata decreases

and newly dumped liquids move quickly and easily to groundwater. (Trial Tr. at 829 (Feenstra);

Sterrett Decl. ¶ 60.) The record is replete with credible historic depositions from LPC employees

admitting to frequently pouring TCE and other solvents on the bare ground in contravention of

company policy. (See PX961.) For example, Earl Wessman, a maintenance mechanic at the

Redlands facility from 1963 to 1975, testified that he dumped a quart to two gallons of TCE

daily at the same “very porous” bed of gravel north of Building 119, where the TCE would “sink

right in.” (USX989 at 188-90 (Wessman Dep.); USX990 at 419-420 (Wessman Dep.); see also

USX987 at 15-17 (Wessman Test.).) Likewise, Christian Mulder, a night shift operator at the

Redlands facility from 1956 to 1965, testified that he and his colleagues dumped one to two

gallons of TCE on the ground outside Buildings 8 and 12 “just about every night.” (USX941 at

34-35, 46-47 (Mulder Dep.).) Although Mulder and his colleagues knew of the availability of a

nearby evaporation pit for solvent disposal, they refused to walk there at night for fear of the

“rattlesnakes and tarantulas.” Instead, they poured the waste TCE on the ground. (Id. at 86-87.)

no evidence that cracks existed in the evaporation pits at the Redlands facility. (See Trial Tr. at 1252-53
(Sterrett).)
The parties’ experts also dispute whether burn pits were a minor source for the Redlands TCE
plume. (Compare Feenstra Decl. ¶¶ 40-46, with Sterrett Decl. ¶¶ 70-73.) Although LPC employees
poured small amounts of TCE directly into the burn pit at the Redlands facility (see USX944A.0002
(Nunes Decl.)), there is no evidence of TCE in the soil gas surrounding the burn pits. (See USX818.)
Thus, the Court credits Dr. Feenstra’s conclusion that the burn pits at the Redlands facility were not a
probable source of any sizable amount of TCE in the Redlands plume.

53
Several other LPC employees also admitted to intermittent pouring of TCE on the bare ground at

Redlands. (See Sterrett Decl. ¶¶ 68-72; USX944A.0002 (Nunes Decl.).)46

However, recognizing that some of the TCE poured onto the ground would have

evaporated before it could percolate into the soil (Trial Tr. at 820 (Feenstra)) and that LPC

shifted from using TCE to TCA as its preferred solvent in the mid-1960s (USX847 at 246-49

(Borgelt Dep.)), the Court concludes that it is improbable that LPC employee’s pouring of TCE

on the ground could account for the entirety of the Redlands TCE plume.

Instead, some portion of the TCE in the Redlands plume probably originated in the vapor

degreaser in Building 91. Vapor degreasers are used to clean grease and oil off of metal

instruments. Heating elements in the bottom of a vapor degreaser boil a “solvent liquid to

produce a zone of hot solvent vapor.” (Feenstra Decl. ¶ 64.) When degreaser operators place

cold metal parts into the vapor zone, either in a basket or on a chain, the vaporized solvent

condenses on and cleans the part. The “soiled” liquid solvent drips back down into the degreaser

sump, where it is reheated into a vapor. (Id. ¶ 65.) Solvent vapors that do not condense on the

metal parts are condensed by cooling coils on the perimeter of the degreaser tank. This

condensed liquid collects and is directed to a solvent-water separator. When functioning

properly, the solvent-water separator separates water that had condensed with the solvent from

the solvent and directs the solvent back into the main degreaser tank to revaporize. (Id. ¶¶ 64,

81.) Any condensed water will float and, once it reaches a certain level, it is released via a drain

onto the floor. (Trial Tr. at 833 (Feenstra).)

46
Dr. Feenstra questioned whether the employees could recall which solvent they had poured on
the ground decades before, or if, at the time of disposal, they could have distinguished between TCE and
TCA. (See Feenstra Decl. ¶ 32.) The Court finds Dr. Feenstra’s opinion in tension with Mr. Oppliger’s
testimony that LPC was a “very unique company” with a “lot of highly educated” and “attentive [and]
detail-oriented type people” (Trial Tr. at 85-86 (Oppliger)), and therefore, it credits the LPC employees’
identification of the solvents they dumped on the bare ground. (See Sterrett ¶¶ 58, 66.)

54
 LPC operated the vapor degreaser in Building 91 to degrease rocket motor casings and

other large components. (See Feenstra Decl. ¶ 23.) Dr. Feenstra opined that the solvent-water

separator in the degreaser functioned improperly and released one-to-two gallons of watered-

downed or even pure TCE per day onto the floor of Building 91, which then flowed through a

drain directly into the subsurface soils. (Id. ¶¶ 76, 85-87; Trial Tr. at 817 (Feenstra); PX841

(diagram showing drain below vapor degreaser).).

The Court credits Dr. Feenstra’s theory. Vapor degreasers (and their solvent-water

separators) are common sources of solvent contamination at industrial and dry cleaning sites and

can release pure solvent into the environment. (Feenstra Decl. ¶ 84.) That there is no evidence

that any LPC employees noticed releases of small amounts of TCE from the solvent-water

separator is not surprising, considering the substantial TCE odor associated with the operation of

the degreaser, which was capable of boiling 120 gallons of TCE at a time, and the fact that

solvent-water separators release water onto the floor when functioning properly. (Trial Tr. at

685 (Feenstra).)

However, there is little evidence to support Dr. Feenstra’s opinion that the vapor

degreaser was the sole source of TCE in the Redlands plume. First, Dr. Feenstra’s opinion that

the solvent-water separator released exclusively pure TCE throughout LPC’s operations at the

Redlands facility lacks convincing support in the record. (Trial Tr. at 1185 (Sterrett); Sterrett

Decl. ¶ 109.) Second, the evidence indicates that LPC did not operate the vapor degreaser every

day or “with any great frequency.” (USX951 at 61 (Rodgers Dep.); USX995 at 117 (White

Dep.).) And finally, contrary to Dr. Feenstra’s opinion and Lockheed’s arguments, the Court

finds, based on Lockheed’s sworn statements in prior legal and administrative actions, that LPC

55
switched from using TCE to TCA in the vapor degreaser “about late 1966 or early 1967.”47

(USX22.0012-13 (Lockheed interrogatory responses in Seven W litigation); see also

USX120.0007-09 (response to request for information from California Department of Toxic

Substances Control).) Thus, while the Court credits Dr. Feenstra’s opinion that the solvent-water

separator on the vapor degreaser in Building 91 is a probable source of a significant portion of

the Redlands TCE plume, it is improbable that, as Dr. Feenstra opines, the vapor degreaser was

the only source. (See Feenstra Decl. ¶ 92.)48

B. Potrero Canyon facility

As described above, the sources of perchlorate contamination in the soil and groundwater

at the Potrero Canyon facility are undisputed. (See Trial Tr. at 1238-39 (Lockheed and

government counsel).) A primary source of contamination for both soil and groundwater is the

Burn Pit Area. (Sterrett ¶¶ 220-21, 226-27.) Some of the perchlorate contamination at the

Potrero Canyon Burn Pit Area, like at the Redlands facility, resulted from the fact that a burn

event would not completely destroy all of the AP, and some of the remaining AP residue

47
Lockheed attempted to disassociate itself from its prior sworn representations by arguing that
the representations were made “very early in [its] fact development.” (Trial Tr. at 1731-32 (Lockheed
closing).) The Court rejects Lockheed’s attempt to play fast-and-loose with its prior and at-the-time-
advantageous sworn representations regarding the switch from TCE to TCA in the vapor degreaser,
especially since Lockheed’s past position is corroborated by Larry Borgelt, who was a safety engineer for
LPC from 1966 to 1974. (See USX841 at 251-52 (Borgelt Dep.).)
48
The Court also concludes that the contribution, if any, of TCE from the Norton Air Force Base
TCE plume into the Redlands TCE plume is of no consequence. The map relied on by Dr. Feenstra that
showed the Norton and Redlands TCE plumes overlapping (see PX1972) demarcated the plumes using
TCE concentrations of 0.5 µg/L, ten times lower than California’s current maximum contaminant level
for TCE of 5 µg/L. (Sterrett Decl. ¶¶ 165, 209; see also Trial Tr. at 648-50 (Feenstra).) The two plumes
do not currently overlap at TCE concentration levels requiring cleanup, and to the extent that they
overlap, the commingling will not cause the TCE concentration in the Redlands TCE plume to increase
above the maximum contaminant level. (Trial Tr. at 653-54 (Feenstra).) Moreover, Lockheed failed to
demonstrate that contributions from the Norton Air Force Base TCE plume has caused in the past or
would cause in the future increased response costs for the Redlands TCE plume. (Id. at 655-56; Sterrett
Decl. ¶¶ 20-21, 209-10.)

56
dissolved in the rain and percolated into the soil and eventually the groundwater. (Cf. Trial Tr. at

730 (Feenstra); Sterrett Decl. ¶ 141; PX1224 at 10.)

However, the Court also finds that some of the perchlorate contamination at the Burn Pit

Area probably originated pre-burn. LPC operated the Burn Pit Area at Potrero Canyon much

differently than its burn pits at the Redlands facility. Rather than burning propellant wastes on a

daily basis, mere minutes after putting the wastes into the pit (see USX910 at 183-84 (Heeseman

Dep.); USX963 at 86 (Speer Dep.)), at the Potrero Canyon facility LPC accumulated wastes

(propellant, solvent, etc.) over months and burned them only a few times per year. (See

USX49.0074-75; USX53.)49 The pits were deeper than those at Redlands and the burns much

larger – each containing up to 250,000 pounds of wastes. (USX49.0074-76.)

Larry Borgelt, a safety engineer at LPC from 1966 to 1974, explicitly contrasted the

Redlands and Potrero Canyon burn pits (USX842 at 432-33 (Borgelt Dep.)), referring to the

latter as a “disposal area.” (USX841 at 318 (Borgelt Dep.).) In anticipation of a burn event, LPC

stored some wastes in drums on a concrete pad (USX49.0077), but also accumulated wastes

directly in the burn pits. (See Sterrett Decl. ¶ 228.) Some of the barrels accumulated in the pits

were leaking and some were emptied directly onto the ground. (USX841 at 318-19 (Borgelt

Dep.); USX842 at 432 (Borgelt Dep.).) These pre-burn event practices contributed to the

perchlorate contamination associated with the Burn Pit Area. Liquid wastes containing AP,

either poured directly into the burn pit or leaking out of barrels, percolated into the soil and

groundwater. Likewise, solid AP dumped into the burn pit weeks or months before a burn would

dissolve in the rain and infiltrate into the soil and groundwater. (Sterrett Decl. ¶ 230.)

49
This was only the case following the termination of the Large Solid Propellant Motor Program
in 1966. Prior to 1966, LPC burned materials on a daily or weekly basis. (Feenstra Decl. ¶ 156;
USX49.0076.)

57
 LPC’s “hog out” procedures caused the perchlorate soil and groundwater contamination

associated with the Large Motor Washout Area and the Rocket Motor Production Area. (Sterrett

Decl. ¶¶ 220-21.) As described above, LPC washed propellant out of defective rocket motors

using high-pressure water jets to allow the reuse of the expensive metal rocket casings. (Feenstra

Decl. ¶ 164; USX49.0068; USX12.0050, .0055-56.) The propellant was hogged out onto the

bare ground where the AP leached out of the propellant, dissolved in the water, and percolated

into the soil and groundwater. (Sterrett ¶ 221; USX49.0068; cf. PX1067 at 4.) Today, hog out

procedures are a recognized source of perchlorate groundwater and soil contamination. (See

PX1224 at 10.) As Mr. Bauer testified, “if they hogged out on the ground, then it’s a source.”

(Trial Tr. at 1160 (Bauer).)

C. LaBorde Canyon facility

As with the Potrero Canyon facility, the parties do not dispute the sources of perchlorate

contamination at the LaBorde Canyon facility. (See Trial Tr. at 1238-39 (Lockheed and

government counsel).) Perchlorate contamination associated with the Test Bay Canyons

originated, as the name suggests, from rocket motor testing and firing operations. (Feenstra

Decl. ¶ 169; Sterrett Decl. ¶ 240.) Although it is unclear precisely by what mechanism AP from

the test-fired rocket motors ended up on the ground, the levels of perchlorate contamination

associated with the Test Bay Canyon indicate that the ground surface was “flood[ed]” with

“perchlorate-bearing water.” (Feenstra Decl. ¶ 169; see also Sterrett Decl. ¶ 240.) Dr. Feenstra

hypothesized that LPC may have “hogged-out” defective motors at the Test Bay Canyon.

(Feenstra Decl. ¶ 169.)

Perchlorate contamination associated with the Waste Discharge Area was caused, at least

in part, by LPC’s disposal of propellant wastes from its burn pits. (Feenstra Decl. ¶ 170; Sterrett

58
Decl. ¶ 241.) In 1962, the Santa Ana Regional Water Pollution Control Board issued a

Resolution to LPC to direct its operation of the Waste Discharge Area. (See PX1116.) Under

that Resolution, LPC was allowed to dispose of approximately 5000 gallons per year of

propellant waste that remained after burning in a burn pit. (PX1116 at 1-2.) As described above,

burning did not completely destroy all of the AP in propellant wastes. Thus, some of the AP

residue remaining in the wastes disposed of at the Waste Discharge Area dissolved in rainwater

and percolated into the soil and groundwater. (See Trial Tr. at 730 (Feenstra); Feenstra Decl. ¶

171; PX1224 at 10.)

II. TRADITIONAL EQUITABLE ALLOCATION

Because the parties have stipulated to liability, and substantially all of LPC’s operations

at the Sites were in performance of government contracts or subcontracts, the Court begins its

analysis by invoking the per capita approach: a fifty-fifty split between Lockheed and the

government. Cf. Davis, 31 F. Supp. 2d at 63-64 (describing the proper and improper times to

begin with a per capita approach). (See also Trial Tr. at 29-30 (Lockheed counsel advocating a

per capita approach); id. at 1931, 1941 (Lockheed closing).) From there, the Court adjusts the

equitable allocation based on its findings, keeping in mind that “[m]athematical precision in this

process is not realistically achievable or desirable.” United States v. Consolidation Coal Co.,

184 F. Supp. 2d 723, 744 (S.D. Ohio 2002).

A. Limited value of certain equitable factors

1. Waste attributable to each party

In many cases, “the dominant factor in determining each party’s equitable share of

liability is the extent to which the response costs are attributable to waste for which that party is

directly responsible.” Davis, 31 F. Supp. 2d at 64. This consideration, reflected in the first three

59
Gore Factors and the first Torres factor, is inapplicable here, for all of the perchlorate and TCE

contamination originated from LPC’s operation of the Sites in the performance of its government

contracts. It is therefore impossible to attribute even rough quantities of the perchlorate and TCE

contamination between the parties without delving into issues regarding the extent of the

government’s involvement in and control over waste disposal practices at the Sites. (See infra

Section II.B.-C.)

2. Parties’ relative benefits from waste disposal activities

“Fairness suggests that parties deriving greater benefit from disposal of hazardous waste

should bear a greater portion of the responsibility for mitigating its adverse effects.” Davis, 31

F. Supp. 2d at 66; see also Weyerhaeuser, 771 F. Supp. at 1426. The economic benefits to the

parties from LPC’s operations at the Sites were “roughly equal.” See AISLIC II, 2013 WL

135405, at *9. The government benefitted from LPC’s research and development and received

rockets supporting eight major Cold War-era programs, and LPC received payments for these

services and products. See id. Although LPC may not have reaped large profits from its

contracts with the government, it helped its parent corporation Lockheed Aircraft Corporation

establish a foothold in the rocket propulsion field, a position that Lockheed retains to this day.50

3. Degree of cooperation.

The sixth and final Gore factor considers “the degree of cooperation by the parties with

Federal, State or local officials to prevent any harm to the public health or the environment.”

Envtl. Transp. Sys., 969 F.2d at 508. Because non-cooperating parties can undermine

CERCLA’s goal of promoting quick and efficient cleanups, see Consolidation Coal, 184 F.

Supp. 2d at 751, “[t]he degree of cooperation with government officials to prevent any harm to

50
See Lockheed’s Memo. on the Benefit Received by Lockheed Propulsion Company, Feb. 20,
2014 [Dkt. No. 124] at 3-6; U.S. Memo. on the Benefit Received by Lockheed Propulsion Company, Feb.
23, 2014 [Dkt. No. 134] at 1-2.

60
the public health or the environment is very important in the contribution analysis.” Cent. Maine

Power, Co. v. F.J. O’Connor Co., 838 F. Supp. 641, 646 (D. Me. 1993). Lockheed argues that

because it has cooperated with California cleanup authorities since before remediation efforts

began, and the government did not admit liability until the eve of this trial, the “cooperation”

factor should favor Lockheed and disfavor the government. (Trial Tr. at 1939 (Lockheed

closing).)

However, Lockheed proceeds on a somewhat distorted view of the facts. Although

Lockheed has cooperated with the state authorities regarding cleanup at the Sites, it cooperation

can hardly be considered voluntarily. Lockheed denied liability for the Redlands TCE plume for

years (PX1677 ¶¶ 12, 36; Trial Tr. at 332 (Blackman)) and initiated cleanup efforts only after

ordered to do so by the Santa Ana Regional Water Quality Control Board. (Trial Tr. at 331-32

(Blackman).) For the government’s part, there is no evidence that any California state agency

ever asked the government to undertake response efforts, and neither Lockheed nor any other

party sought reimbursement from the government under CERCLA until this suit was filed in

2008. And while the government has not directly paid for any response costs yet, see

Consolidation Coal, 184 F. Supp. 2d at 751 (finding that a failure to pay for response costs or

actions prior to CERCLA suit relevant to whether the PRP meaningfully cooperated), its protests

against doing so are grounded primarily on the fact that it has already indirectly paid for the

majority of Lockheed’s response costs at the Sites. (See Meyer Decl. ¶ 29 fig. 5.).

In short, Lockheed was forced by cleanup and abatement orders to undertake its

remediation efforts, and the government has (albeit indirectly) footed most of the bill.51 On these

51
The fact that Lockheed has indirectly recovered through government contracts over 72% of its
past response costs for the Sites distinguishes this case from AISLIC II, in which that court concluded the
plaintiff’s cooperation with state regulatory authorities over a twenty-year period, without assistance from

61
facts, the Court cannot conclude that the government has been meaningfully uncooperative such

that it should receive an upward adjustment in its equitable allocation for the Sites.

4. The government’s ownership of waste

Pursuant to the government property clauses in its contracts with LPC, the government

owned an unknown, but undoubtedly sizable, amount of the AP and TCE that was disposed of at

the Sites. Under government contracting provisions, title will vest in the government for certain

property – including chemicals and other raw materials – purchased and used by a government

contractor. When title vests in the government generally depends on the type of contract and

whether contractor charges the property as a direct or indirect cost.

Under LPC’s cost-reimbursement contracts, such as the Apollo subcontract (PX104), title

vested in the government for property charged as a direct cost as soon as the property was

delivered to LPC, and title vested for property charged as indirect costs at the earliest of three

events, including LPC’s use of property in performance of the contract or the government’s

reimbursement of the cost to LPC. (Johnson Decl. ¶ 8.) For LPC’s fixed-price contracts with

progress payments, such as the SRAM production subcontracts, title vested in the government

for property charged as both direct and indirect costs upon the issuance of a progress payment to

LPC. However, title reverted back to LPC at the termination of the contract. (Id. ¶ 9; Nagle

Decl. ¶¶ 98, 103-105.)52

the federal government, necessitated a 5% increase in the government’s equitable share. See AISLIC II,
2013 WL 135405, at *5-6.
52
Government regulations did not require fixed-price contracts without a progress payment
clause (e.g., USX188) to include a government property clause. (Nagle Decl. ¶ 100-01.) Thus, LPC
owned and retained ownership of all AP and TCE used in its performance of those contracts. (See id. ¶
102.) Due to the limited record in this case (see supra n. 31), it is impossible to determine what
percentage of LPC’s contracts were fixed-price without a progress payment clause or what percentage of
AP and TCE wastes those contracts represented. (See Trial Tr. at 151-52, 156-57 (Johnson).)

62
 LPC most likely charged most, if not all, AP as a direct cost and TCE as an indirect cost

to government contracts. (Johnson Decl. ¶ 20; Nagle Decl. ¶ 135.) Thus, title vested in the

government for both materials during some point in the manufacturing processes and, depending

on the contract type, remained vested in the government indefinitely (cost-reimbursement) or

until the end of the contract (fixed-price with progress payments). In either event, the

government retained title over at least some of the AP and TCE at the time of their disposal at

the Sites.53 Lockheed argues that the government’s ownership of the AP and TCE ultimately

disposed of at the Site should result in a higher equitable allocation for the government. (E.g.,

Trial Tr. at 33 (Lockheed opening).)

Merely owning the hazardous substances that caused response costs is, however,

insufficient for liability under CERCLA. See Morton Int’l, 343 F.3d at 679. Instead, a non-

disposing owner of a hazardous substance is generally liable only if he arranges for its disposal

or treatment by a third party. 42 U.S.C. § 9607(a)(3); see also Burlington N., 556 U.S. at 610-

11. The Court considers this distinction to be an important signal that, in passing CERCLA,

Congress was more interested in holding liable, where distinguishable, those who disposed of

and controlled the disposal of hazardous substances rather than those who owned them.

In its exercise of equitable discretion, the Court applies this distinction here. It is unclear

how much of the AP or TCE used (and disposed of) at the Sites the government actually owned.

While the government ownership was likely significant in absolute terms (e.g., pounds of AP or

53
Mr. Nagle, the government’s contracting expert, opined that “any title to materials acquired by
the government pursuant to a title vesting clause[] would not survive the consumption of the material in
the manufacturing process and, therefore, the government did not ‘retain’ title to waste generated by
[LPC] during contract performance.” (Nagle Decl. ¶ 143.) The Court is sympathetic to Mr. Nagle’s
pragmatic opinion that the title vesting clauses were not meant to extend to waste and shares his
reluctance to engage in a debate over the metaphysical question of who owns waste at the molecular level.
(See Trial Tr. at 1058 (Nagle).) However, from a textual standpoint, the contract clauses do not provide
automatic title reversion or automatic government abandonment for “waste” property. Thus, the Court
concludes that the government’s title in AP and TCE survives even after each becomes “waste.”

63
gallons of TCE), Lockheed’s government contracts expert could not conclude with a reasonable

degree of certainty that the government owned more than fifty percent of either AP or TCE used

at the Sites. (Trial Tr. at 157 (Johnson).) Moreover, regardless of the government’s ownership,

Lockheed possessed the waste at all times up to and including the time of disposal. And, for

fixed price contracts with progress payments (like SRAM), the waste – even if already disposed

of – reverted back to Lockheed upon the termination of the contracts. See AISLIC II, 2013 WL

135405, at *8.

Thus, while ownership of the TCE and AP may be an important factor in establishing

arranger liability under CERCLA § 107(a)(3), this is not the issue before the Court. Rather, as to

the allocation analysis, the Court finds the ownership of the substances to be of limited value.

Cf. AISLIC II, 2013 WL 135405, at *8. The critical issue is not ownership in the first instance,

but rather the parties’ respective control over the disposal of TCE and AP at the Sites. (See infra

Section II.B.-C.)

5. The government’s ownership of facilities

Owning a facility at the time hazardous substances were disposed of at that facility results

in liability under CERCLA. See 42 U.S.C. § 9607(a)(2). From this statutory hook, courts

frequently allocate an “owner’s share” of liability to parties “simply by virtue of being the

landowner,” United States v. R.W. Meyer, Inc., 932 F.2d 568, 571 (6th Cir. 1991), or owning the

leaking or faulty equipment at a site. E.g., AISLIC II, 2013 WL 135405, at *4-5.

Both parties owned equipment at the Sites. (See, e.g., USX26.0018-28; USX27;

USX28.0025-27.) Indeed, the government stipulated to its liability under CERCLA because it

owned and furnished some of the major pieces of equipment for LPC’s operations. (See Trial Tr.

at 899 (government counsel).) Lockheed argues that government ownership of this equipment –

64
including the vapor degreaser in Building 91 and the hogged-out rocket casings at the Potrero

and LaBorde Canyon facilities – should weigh in Lockheed’s equitable favor because those

pieces of equipment were sources of TCE and perchlorate contamination (i.e., “facilities”) at the

Sites. (E.g., Trial Tr. at 33 (Lockheed opening); id. at 1946 (Lockheed closing.)

Lockheed is correct that both the government-owned vapor degreaser and hogged-out

rocket casings are “facilities” under CERCLA § 101(9). See Am. Int’l Specialty Lines Ins. Co. v.

United States (AISLIC I), 2010 WL 2635768, at *23-24 (C.D. Cal. June 30, 2010). The Court,

however, does not consider the government’s mere ownership of certain pieces of equipment to

have much importance in determining an equitable allocation. Instead, just as with the

ownership of the waste, the Court considers “[f]actors additional to the simple fact of ownership”

of facilities – in particular the parties’ respective control over the disposal operations – to be the

paramount equitable consideration. See Yankee Gas, 852 F. Supp. 2d at 248. Thus, the Court

does not adjust the per capita equitable allocation on account of either party’s ownership of

equipment or facilities at the Sites.

6. Knowledge of risk of pollution from AP and TCE

The fifth Gore Factor considers “the degree of care exercised by the parties with respect

to the hazardous waste concerned.” Envtl. Transp. Sys., 969 F.2d at 508. As the Court

concluded above, the contamination at the Sites originated, inter alia, from pouring TCE and AP

wastewaters, and burning solid propellant wastes, on the bare ground. Mr. Bauer, a government

expert, opined these disposal practices violated the standard of care in existence at the time of

LPC’s operations. (Trial Tr. at 1086-87 (Bauer); Bauer Decl. ¶¶ 11(b), 11(e), 138(b).) The

Court does not credit Mr. Bauer’s testimony for two reasons.

65
 First, although LPC knew that the Sites were located in areas with sensitive groundwater

supplies (see, e.g.,USX15.0002-03; USX43.0002; USX50.0002; USX970 at 140-41 (Speer

Dep.); USX977 at 35-36, 43 (Stickney Dep.)), neither TCE nor perchlorate were known

groundwater contaminants at the time LPC operated the Sites. “[T]here was generally no

recognition that TCE posed a danger to the environment until the mid 1970’s.” Bolinder Real

Estate, L.L.C. v. United States, 2002 WL 732155, at *8 (D. Utah Apr. 24, 2002). Indeed, TCE

groundwater contamination was first discovered in California only in 1979 (PX1537 at 12), and

the EPA did not regulate TCE as a drinking water contaminant until 1989. (PX2060 at 18.)

Similarly, according to the U.S. Government Accountability Office, perchlorate “was initially

identified by EPA as a potential [groundwater] contaminant in 1995” (PX1229 at 3), and,

according to the DOD, did not “emerge[] as an environmental issue of national interest [until]

1997,” when new testing processes decreased the detection threshold from 400 to 4 µg/L.

(PX1223 at 5; see also PX1229 at 3-4.)54 Although LPC knew when it operated the Sites that

AP was a dangerous material, their safety concerns were limited to fires and explosions and did

not extend to environmental contamination. (See, e.g., Trial Tr. at 87, 94-95 (Oppliger).)

Second, both pouring TCE and AP wastewaters and burning solid propellant wastes on

bare ground were entirely consistent with the general standards of care in existence at that time.

For TCE, the Court need look no further than the government’s position in a prior case:

54
Accordingly, the Court does not credit Mr. Bauer’s testimony that members of scientific
community knew that pouring TCE or AP wastewaters or burning propellants on the bare ground would
cause groundwater contamination. (Trial Tr. at 1081-87 (Bauer).) Mr. Bauer’s reliance on “personal
experience and knowledge and chemical research” (id. at 1083), as well as a single, ambiguous scholarly
article from 1953 that concluded “[s]olvents and soluble chemicals contained in industrial wastes . . .
remain in solution in the liquid phase . . . and percolate downward to the underlying groundwater”
(USX810.0006) is countered by the vast weight of the evidence that neither TCE nor perchlorate were
considered groundwater pollutants at the time LPC operated the Sites. (See Feenstra Decl. ¶¶ 96-102
(TCE), 175-91 (AP/perchlorate); PX2020-2024 (summary tables of historic published literature on
groundwater contamination re: TCE); PX2025-2028 (summary tables of historic published literature on
groundwater contamination re: AP/perchlorate).)

66
 It is undisputed that the plaintiff’s allegations of negligence, in particular,
‘pouring the solvents onto the ground . . . ,’ even if true, could not have violated
any standard of care in existence prior to, or during . . . the early 1970s.

Snyder et al. v. United States, 04-cv-627, Gov’t Memo. in Support of Mot. for Summ. J. at 3-4,

(Feb. 22, 2006 S.D. Miss.) (PX2060). In Snyder, the government adopted its expert’s opinion

that the disposal of TCE on the bare ground and “‘letting it evaporate’ was common practice

throughout the period from 1940 until 1973, followed manufacturer’s recommendations, and was

not at variance with state or federal regulations.” (Id. at 19; see also PX1948 (summary table of

historic manufacturer disposal recommendations for TCE).) This was so because the prevailing

understanding at the time was that TCE poured on bare ground would evaporate or be absorbed

in the surface or subsurface soils and “would not contaminate or alter the intended use of

underground water supplies.” (PX2060 at 19.)

For perchlorate, the Court does not credit Mr. Bauer’s opinion that LPC should have

burned AP and propellant wastes at the Redlands and Potrero Canyon facilities in burn pans

rather than on the bare ground in burn pits. (Bauer Decl. ¶¶ 11(e), 133-36, 138(c).) LPC’s

burning of propellant wastes on the bare ground was consistent with both military and industry

practice during LPC’s operations at the Sites. Military manuals applicable to the LPC’s

operations explicitly recommended the burning of explosive wastes on bare ground. (See, e.g.,

PX0002 at 316; PX0003 § 125(b)(1), (c); PX0005 §§ 3-6, 4-11, 4-12; PX0007 § 1505; PX0009 §

7-2(c).) And the United States military itself burned propellant wastes on the bare ground and in

burn pits at its facilities (Trial Tr. at 1380 (Dull)), including at Camp Irwin, where the

government occasionally disposed of wastes from the Redlands facility. (See USX852 at 168

(Borgelt Dep.).) Indeed, the government failed to identify any government facilities where burn

pans were used for propellant wastes prior to the 1980s. It is no surprise, then, that the rampant

67
perchlorate groundwater contamination at historic DOD and NASA facilities was caused by the

same activities for which the government now seeks to fault LPC: “disposing of perchlorate

wastes in open pits, open burning and detonation of perchlorate, and using water to remove

perchlorate residue from rocket engines.” (See PX1229 at 26.)55 Indeed, in 2010, the GAO

identified, in California alone, seven government facilities with perchlorate-contaminated

groundwater that were listed on the Superfund National Priorities List and eleven DOD facilities

with perchlorate groundwater levels exceeding 15 µg/L. (Id. at 42, 44.)

On these facts, the Court cannot agree with the government’s position that LPC breached

any generally recognized standard of care by pouring TCE and AP wastewaters and burning

solid propellant wastes on the bare ground at the Sites.56

7. Violation of California water quality laws

A party’s violation of an environmental law in force at the time of its disposal of

hazardous wastes can be an important equitable factor relevant to a PRP’s degree of care,

55
The Court also concludes that LPC’s hog out procedures did not violate any a standard of care
in existence at the time. In the 1970s, “hogging out” was widely used by solid propellant rocket motor
manufacturers as a cost-saving tactic that allowed the reuse of expensive metal rocket casings. (See
PX0457 at 5-10 to 5-11; see also PX1229 at 26.) Indeed, as late as 1975 DOD contracted with companies
to hog out rocket motors. See AISLIC I, 2010 WL 2635768, at *11-12.
56
Mr. Bauer appeared to adopt the so-called “precautionary principle” as the relevant standard of
care in a CERCLA allocation action. (See, e.g., Trial Tr. at 1087, 1141 (Bauer).) An amorphous pillar of
contemporary environmental theory, the precautionary principle “requires that in the light of scientific
uncertainty, when credible evidence is put forth that a risk exists, action should be taken to minimize that
risk or eliminate it even though absolute proof has not been obtained which quantifies the risk.” New
Mexico v. Gen. Elec. Co., 335 F. Supp. 2d 1185, 1221 (D.N.M. 2004); see also OXFORD DICTIONARY OF
ENVIRONMENT AND CONSERVATION 353 (2008); Thomas O. McGarity, MTBE: A Precautionary Tale, 28
HARV. ENVTL. L. REV. 281, 334-35 (2004). While the precautionary principle may be equitably
applicable in other CERCLA cases, it is of little value here. For applying the precautionary principle as a
standard of care requires, at a minimum, a demonstration that some factual basis for the exercise of
precaution – i.e., knowledge of the potential environmental risk – existed at the time of the relevant
disposals. There is no such factual basis here, for scant evidence exists that anyone (the government
included) was exercising precaution, or had any inkling to do so, with regard to the disposal of TCE or
AP between 1954 and 1975. Hindsight, however perfect, is an insufficient basis for imposing a standard
of care as demanding as the precautionary principle.

68
culpability, and cooperation with government authorities. Environmental violations are most

relevant to the extent that (1) the law is not coterminous with CERCLA liability (i.e., strict

liability for disposal of hazardous substances); (2) the party’s actions clearly violated the law as

established at the time of the disposals; (3) those violations were either knowing or reckless; and

(4) compliance with the law would have reduced the total contamination or response costs.

The government argues that LPC’s equitable share should be increased because it failed

to submit reports of various waste discharges at the Sites as required under the Dickey Act, 1949

Cal. Stat. ch. 1549. (Trial Tr. at 874, 901-02 (gov’t opening).)57 Passed in 1949, California’s

Dickey Act established regional water quality control boards and a state water quality control

board “to co-ordinate state activities with regard to water pollution and nuisance by sewage and

industrial waste.” See Note, California’s Water Pollution Problem, 3 STAN. L. REV. 649, 652

(1951). Dickey Act § 13054 required “[a]ny person proposing to discharge sewage or industrial

waste within any region, other than into a community sewer system, [to] file with the regional

board of that region a report of such proposed discharge.” (USX122.0007.)58 Once a discharger

filed a report, the regional board was to “prescribe requirements as to the nature of such

discharge with relation to the conditions existing from time to time in the disposal area or

receiving waters upon or into which the discharge is proposed and notify the person proposing

the discharge of its action.” (Id.) The discharger was required to “provide adequate facilities to

meet any such requirements” prescribed by the regional board. (Id.)

57
The government did not dispute Lockheed’s assertion that LPC complied with California’s air
pollution laws by obtaining the necessary permits when operating the evaporation and burn pits at the
Sites. (Trial Tr. at 1751 (Lockheed closing); Feenstra Decl. ¶¶ 40, 49-50.)
58
Dickey Act § 13005 defined “industrial waste” broadly as “any and all liquid or solid waste
substance, not sewage, from any producing, manufacturing or processing operation of whatever nature.”
(USX122.0002.)

69
 According to the California Attorney General’s interpretation of the Dickey Act, the

disposal of industrial waste “on land with a possible subsequent movement by evaporation or

percolation into . . . underground waters” was considered an “indirect” discharge. 48 Op. Cal.

Att’y Gen. 85, 86 (1966) (USX1175). The government asserts that LPC repeatedly violated the

Dickey Act by disposing liquid industrial wastes – e.g., TCE and AP wastewaters – on the bare

ground at various locations at the Sites without filing notices of proposed indirect discharges

with the Santa Ana Regional Water Quality Control Board. (E.g., Trial Tr. at 900-03 (gov’t

opening); Bauer Decl. at ¶ 168.)59 Mr. Bauer opined that LPC was aware of the Dickey Act

because it submitted a notice of proposed waste discharge in 1962 for the Waste Discharge Area

at the LaBorde Canyon facility and subsequently received a Resolution from the Board

approving the proposed discharge. (Bauer Decl. ¶ 168.)60 Mr. Bauer further opined that had

LPC submitted notices of proposed waste discharges for its various indirect discharges across the

Sites, the Board would have investigated and likely placed requirements on the discharges to

limit groundwater pollution. (Id. ¶ 166.)

Lockheed’s expert, Mr. Delaney, testified in response that LPC had no responsibility

under the Dickey Act to file notices of proposed waste discharges with the Board for the

59
In particular, Mr. Bauer opined that LPC violated the Dickey Act by failing to notify the Board
of most of its disposal activities, including, but not limited to, LPC’s use of burn pits at the Redlands and
Potrero Canyon facilities, the washout from grinding and mixing buildings and the discharges from
Building 77’s south sump at the Redlands facility, the hog out procedures at the Potrero Canyon and
LaBorde Canyon facilities, and the waste disposal areas at the Potrero Canyon and LaBorde Canyon
facilities. (See, e.g., Bauer Decl. ¶¶ 104, 117, 122, 137, 168.)
60
Mr. Bauer also opined that LPC violated its Dickey Act Resolution for the Waste Discharge
Area at the LaBorde Canyon facility – Resolution 62-24 – by constructing too many ponds and by
discharging perchlorate and other compounds not listed in the resolution. (Bauer Decl. ¶¶ 205-06.)
However, under that Resolution, the Board was allowed to and did “[p]eriodically” inspect the Waste
Discharge Area to ensure compliance with the Resolution. (PX1116 at 3; PX1121; PX 1130.) Further, as
noted above, no one at the time knew that the burn pit residues disposed of at the Waste Discharge Area
would still contain perchlorates. (See Trial Tr. at 730 (Feenstra); Feenstra Decl. ¶ 171; PX1224 at 10.)
Based on these facts, the Court cannot agree that LPC violated Resolution 62-24 when operating the
Waste Discharge Area at the LaBorde Canyon facility.

70
disposals of TCE and propellants on the bare ground at the Sites. (Delaney Decl. ¶ 13.) In

particular, Mr. Delaney opined that neither the Dickey Act nor its successor, the Porter Cologne

Water Quality Act of 1969, 1969 Cal. Stat. ch. 482, prohibited “intermittent or accidental

wastewater [indirect] discharges” such as those that occurred at the Sites. (Delaney Decl. ¶ 13;

see also Trial Tr. at 419-20 (Delaney).) Instead, according to Mr. Delaney, the notification

provisions of the Dickey Act only covered “large-scale, purposeful” disposals of industrial

wastes directly or indirectly into surface or groundwaters. (Trial Tr. at 419 (Delaney); Delaney

Decl. ¶ 13.)

Although the parties dedicated a significant amount of time to the issue of whether LPC

violated the Dickey Act, the Court is unable to resolve this debate for purposes of determining

Lockheed’s equitable share. First, it is unclear whether LPC had the duty to report to the Board

its disposal of TCE and propellant wastes on the bare ground at the Sites. Contrary to Mr.

Delaney’s opinion, the Dickey Act does not provide an exception for de minimis or intermittent

discharges. Instead, the plain language of the statute required a notice of proposed waste

discharge for any discharge of industrial waste. (USX122.0007.) Further, that TCE and AP

were not known groundwater pollutants at the time (supra Section II.A.6) seems irrelevant under

the Dickey Act’s broad definition of “industrial waste.” (USX122.0002.) On the other hand, the

Board periodically inspected each of the Sites (see PX1121; PX1130), and no evidence exists

that the Board ever instructed LPC to file a notice of proposed waste discharge for any of its

disposal activities. (Cf. PX1130 (“No written instructions or regulations from [the Board] has

been received.”).) Indeed, in 1967, the Board considered “the amount of industrial waste

discharged by LPC into the water drainage basin . . . to be negligible.” (Id.) This arguable

conflict between the statute and the Board’s on-the-ground enforcement of the Dickey Act

71
suggests that the Board and LPC’s contemporaneous understanding of the Act was that it did not

apply to LPC’s disposal activities.61

Second, even assuming arguendo that the Dickey Act applied to LPC’s disposal

activities, the Court cannot agree with Mr. Bauer’s speculation had LPC complied with the law,

the Board would have imposed requirements that would have reduced groundwater

contamination or response costs at the Sites. Under the Dickey Act, the Board lacked the power

to prohibit indirect discharges. 48 Op. Cal. Att’y Gen. at 86.62 Further, although the Board

could set limits on the concentrations of contaminants in discharged industrial waste (see Trial

Tr. at 1129-31), the statute “prohibited [the Board] from specifying the design, location, type of

construction or particular manner in which compliance may be had.” Ronald R. Robie, Water

Pollution: An Affirmative Response by the California Legislature, 1 PAC. L. J. 2, 19-20 (1970).

Considering the Board’s seeming lack of concern regarding LPC’s industrial waste disposal

practices (see PX1130), the fact that neither TCE nor perchlorate were recognized groundwater

contaminants at the time,63 and that LPC generally operated consistent with the recognized

standards of care for TCE and AP waste disposal (see supra Section II.A.6), the Court is unable

to conclude that if LPC had reported its waste discharges, the Board would have required LPC to

61
The government provided no evidence that other solid rocket motor manufacturers filed notices
of proposed waste discharges for the common disposal methods, i.e., burn pits and hogging out, that are at
issue here. Indeed, with the exception of the Aerojet General Corporation’s notice of proposed waste
discharge (which, as discussed infra n. 64, is not comparable to Lockheed’s activities at the Sites), the
government provided no evidence that any manufacturers at the time filed notices of proposed waste
discharges related to TCE or perchlorate.
62
Although the Porter-Cologne Water Quality Control Act was passed in 1969, it was not until
1975 that the California Attorney General interpreted the statute as providing regional boards with the
authority to prohibit indirect discharges. 58 Op. Cal. Att’y Gen. 531, 531 (1975) (USX1176).
63
Indeed, as the Los Angeles Regional Water Quality Control Board stated in 1980, “TCE and
other volatile organics now listed on EPA priority pollutant lists were not known or believed to have been
posing [a] pollution threat to groundwaters because of their high volatility and loss to evaporation at the
time of use” in the 1950s through the 1970s. (PX1532 at 128.)

72
undertake preventative actions to reduce the TCE or perchlorate concentrations in its industrial

wastes before disposal.64 The government’s suggestions that the Board could have set discharge

limits that required burn pans or other preventative measures (Trial Tr. at 881, 903-04 (gov’t

opening); id. at 1140 (Bauer); Bauer Decl. ¶ 35) are also speculative and lack support in the

record.

In sum, the Court cannot conclude that LPC violated the Dickey Act as it was understood

or applied at that time or that LPC’s “compliance” with the Act would have led to Board

regulation of its discharges and thus reduced contamination and response costs at the Sites today.

Accordingly, the Court gives no equitable weight to the issue of LPC’s compliance (or lack

thereof) with the Dickey Act.

64
To argue otherwise, the government points to the regulation, under the Dickey Act, of the
Aerojet General Corporation’s solid rocket motor production site in Rancho Cordova, California. (See,
e.g., Trial Tr. at 883-84, 900-01 (gov’t opening); Bauer Decl. ¶ 34.) Aerojet submitted a notice of
proposed waste discharge to the Central Valley Regional Water Pollution Control Board in the early
1950s and the Board adopted a Resolution prohibiting the discharge of wastes containing, inter alia, TCE
and AP “in a manner which will permit their entry into either the groundwater or the waters of the
American River.” (USX128.0002.) The Board modified the Resolution in 1962 to impose stricter
discharge limits. (USX132.) The government argues that Aerojet’s filing of a notice of proposed waste
discharge and the Board’s issuance of Resolutions regarding those discharges demonstrates both that LPC
violated the Dickey Act and that, had LPC filed a notice of proposed waste discharge, its regional board
would have regulated its disposals as indirect discharges.
However, Aerojet’s operations are too distinct from LPC’s to credit the government’s argument.
First, Aerojet’s discharge methods were specifically designed to allow its industrial wastes to percolate
into the ground. (See Trial Tr. at 413-14 (Delaney); id. 802-03 (Feenstra).) LPC’s only directly
comparable disposal processes at the Sites were the “percolation pits” in the Waste Discharge Area at the
LaBorde Canyon facility, for which LPC did file a notice of proposed waste discharge. (See id. at 802-03
(Feenstra).) Second, Aerojet’s disposals were voluminous – up to 1000 gallons per day – and contained a
much higher quantity of AP than at LPC’s Sites – up to 270 pounds per day. (PX1112 at 1; Trial Tr. at
1154 (Bauer).) Even so, the Board’s concern with Aerojet’s discharges was toxicity to plant life.
(PX1112 at 3.) And finally, the Aerojet and LPC operations fell under the jurisdiction of different
regional water quality control boards – Central Valley and Santa Ana, respectfully – such that the Court
cannot automatically infer that the Santa Ana Board would have taken the same actions for LPC that the
Central Valley Board took at the Aerojet facility. Accordingly, the Court finds that Aerojet’s regulation
under the Dickey Act is of limited relevance to this case.

73
 8. Ability to Pay

“[T]he principal reason for considering ability to pay is to ensure that the party seeking

contribution will not bear sole responsibility for any portion of the joint liability otherwise

attributable to defendants from whom recovery is unlikely.” Davis, 31 F. Supp. 2d at 66. This

factor is not an open invitation for courts to increase or decrease a party’s equitable share based

solely on net worth, but is instead meant to recognize “that a PRP’s share of liability should not

be established at a level that exceeds its resources,” lest the plaintiff be left to shoulder that

PRP’s equitable share. Id.65 In this case, the parties’ abilities to pay have no real importance

because both Lockheed and the federal government are capable of shouldering the entirety of the

cleanup costs for the Sites. The Court notes, however, that in the recent years of increased

government austerity and serious fiscal shortfalls, Lockheed has – primarily through U.S.-

government contracts – seen “record . . . earnings, and profit margin as well as strong cash

generation.” (See Press Release, supra n.22.)66

9. Indemnification provisions

Both parties claim that certain indemnification clauses in the contracts argue in favor of

lowering their share. Indemnification merely shifts, but does not negate, CERCLA liability. See

42 U.S.C. § 9607(e)(1). To determine whether pre-CERCLA indemnification clauses cover

65
In this regard, the “ability to pay” consideration is an ex ante protection against so-called
“orphaned” shares – “those shares of ‘liability attributable to a party who is insolvent, cannot be located,
or cannot be identified.’” PCS Nitrogen, 714 F.3d at 168 (quoting Lyondell Chem. Co. v. Occidental
Chem. Corp., 608 F.3d 284, 303 (5th Cir. 2010)).
66
Since 1994, Lockheed’s stock price has soared from around $20 per share to around $160 per
share today, more than doubling the performance of the S&P 500 and the Dow Jones Industrial Average
during that time period. See Lockheed Martin Corporation, YAHOO! FINANCE, http://finance.yahoo.com/
echarts?s=LMT+Interactive#symbol=LMT;range= (last visited April 22, 2014) (compare with “DJI” and
“GSPC” from January 3, 1994 to April 1, 2014). Although Lockheed’s net sales were down in the first
quarter of 2014, its profits were up by over 22% ($172 million) compared to the first quarter of 2013.
Press Release, Lockheed Martin Corp., Lockheed Martin Reports First Quarter 2014 Earnings Results
(April 22, 2014), available at http://www.lockheedmartin.com/us/news/press-releases/2014/april/0422hq-
earnings.html.

74
CERCLA liabilities, “courts have uniformly held that the clause must be either ‘[1] specific

enough to include CERCLA liability or [2] general enough to include any and all environmental

liability which would, naturally, include subsequent CERCLA claims.’” Elf Atochem N. Am. v.

United States, 866 F. Supp. 868, 870 (E.D. Pa. 1994) (quoting Beazer E., Inc. v. Mead Corp., 34

F.3d 206, 211 (3d Cir. 1994)). Neither the indemnification provisions in the government-owned

facilities contracts (which favor the government) nor the ultra-hazardous activities

indemnification provisions in the SRAM production subcontracts (which favor Lockheed) satisfy

this test.67

The government asserts that indemnification provisions in its facilities contracts support

imposing a higher equitable allocation on Lockheed. During LPC’s operations at the Sites, the

government and Lockheed entered into facilities contracts that provided (often rent-free)

government-furnished equipment – e.g., the vapor degreaser in Building 91 – to LPC for its use

in performance of its contracts. (USX228.0004 (28 C.F.R. §13.101-8 (1955).) Facilities

contracts were required to have government indemnification clauses. (Trial Tr. at 952-53

(Nagle); Nagle Decl. ¶ 109.) Under one example of a government indemnification provision,

LPC agreed to

indemnify and hold the Government harmless against claims for injury to persons
or damage to property of the contractor or others arising from the Contractor’s
possession or use of the Facilities. However, the provisions of the Contractor’s
related procurement contracts shall govern the Government’s assumption of
liability for such claims arising out of or related to the performance of each such
related procurement contract and involving the possession or use of the Facilities.

(USX84.0029; USX 113.0020.)

67
At least two abandonment contracts, through which the United States abandoned government-
owned equipment or raw materials, contained indemnification provisions that clearly shifted liability
arising from the disposal of those equipment and raw materials (including AP) to LPC. (E.g., PX461;
PX1073.) Because those contracts are so few, and limited to relatively small amounts of hazardous
substances, they are insufficient to shift the equities in this case.

75
 The language in this provision is typical “of the type of language used to indemnify a

transferor against a tort, nuisance or trespass claim” – environmental liabilities are nowhere

mentioned. See Mobay Corp. v. Allied-Signal, Inc., 761 F. Supp. 345, 358 (D.N.J. 1991).

Without more, the Court may not infer any intent to cover CERCLA response costs. Id.

Moreover, the provision is expressly limited by any contrary liability-shifting provisions in the

related procurement contract. (USX84.0029.) Thus, even assuming the facilities

indemnification provision did cover CERCLA response costs, it is impossible to determine

whether the provision applies to any given instance of disposal without reference to the

procurement contract – the vast majority of which are not available. In any event, the Court’s

conclusion that the ownership of facilities has little probative value in determining allocation as

compared to the operation of those facilities (see supra Section II.B.5) means that the facilities

indemnification provisions are of little importance. Accordingly, the Court gives no weight to

the government indemnification provisions in the facilities contracts.

For its part, Lockheed asserts that indemnification provisions in the 1971-75 SRAM

production subcontracts – LPC’s largest contracts at the Sites – support an equitable adjustment

in its favor. (Trial Tr. at 1848-51 (Lockheed closing).) Through these indemnification

provisions, the government agreed to “hold harmless and indemnify” LPC against, inter alia,

[c]laims (including reasonable expenses of litigation or settlement) by third
persons . . . for death, bodily injury (including sickness or disease), or loss of,
damage to, or loss of use of property; [l]oss of or damage to property of [LPC],
and loss of use of such property, but excluding loss of profit; . . . to the extent that
such a claim, loss or damage arises out of the direct performance of this contract;
is not compensated by insurance or otherwise; and results from a risk defined in
this contract to be unusually hazardous.

(PX0560 § 5.4.2.1 (emphasis added).) The contracts defined as “unusually hazardous”:

all risks resulting from or in connection with (i) the explosion, detonation,
combustion, or surface impact of a missile, simulated missile or component

76
 thereof utilizing the material delivered or services rendered under this contract;
(ii) the use of materials containing radioactive, toxic, explosive or other hazardous
properties of chemicals or energy sources . . . regardless of whether the harm
occurs before or after delivery to the Government of equipment or materials under
this contract, or within or outside the United States.

(Id. § 5.4.1.1.)

The broad definition of “unusually hazardous” risks arguably covers the use and disposal

of AP and TCE, as “explosive” and “toxic” chemicals, respectfully. However, assuming

arguendo that the SRAM indemnification provisions are broad enough to cover CERCLA

response costs,68 the Court finds that they are not useful in making an allocation determination

for two reasons. First, Lockheed voluntarily dismissed with prejudice its contract action before

the Armed Services Board of Contract Appeals, in which it sought from the United States

indemnification under these precise provisions in its SRAM production subcontracts.

(USX83.)69

68
The Court questions whether the provision’s relevant limitation to “damage to property” is too
narrow to unambiguously cover CERCLA response costs. Because CERCLA is not a tort statute, it “does
not provide compensation to a private party for damages resulting from contamination.” Gussack Realty
Co. v. Xerox Corp., 224 F.3d 85, 91 (2d Cir. 2000). And, although a majority of courts hold that, in the
insurance context, response costs incurred under CERCLA are considered “property damages,” see, e.g.,
Indep. Petrochemical Corp. v. Aetna Cas. & Sur. Co., 944 F.2d 940, 945-46 (D.C. Cir. 1991) (applying
Missouri law); AIU Ins. Co. v. Superior Court, 799 P.2d 1253, 1270-71 (Cal. 1990); Carol A. Crocca,
Annotation, Liability Insurance Coverage for Violations of Antipollution Laws, 87 A.L.R. 444 (1991),
“[v]iewed outside the insurance context, the term ‘damages’ is ambiguous: it is reasonably open to
different constructions.” Cont’l Ins. Companies v. Ne. Pharm. & Chem. Co., 842 F.2d 977, 985 (8th Cir.
1988); see also New Castle Cnty. v. Hartford Acc. & Indem. Co., 933 F.2d 1162, 1188 (3d Cir. 1991).
The case law distinguishes between “ordinary” meaning of “damages,” which would encompass
traditionally equitable recoveries, and the “legal” meaning, which differentiates legal damages (like in a
tort action) and equitable remedies (like a recovery under CERCLA). See New Castle Cnty., 933 F.2d at
1187-88. Considering this ambiguity in the term “damages”, it is at least questionable whether the plain
language of the SRAM indemnification provisions encompassed environmental cleanup costs, or was
instead simply a run-of-the-mill tort provision narrowly targeted to certain “unusually hazardous” risks.
Cf. Mobay Corp., 761 F. Supp. at 358.
69
Even had Lockheed dismissed its claims before the ASBCA without prejudice, under the
Tucker Act this Court would lack subject-matter jurisdiction over any legal claims for indemnification
under the SRAM production subcontracts because those claims would be based entirely in contract and

77
 Second, and more importantly, the predominant concern in equity is the intent of the

parties, not whether the plain language of the indemnification provision would bind either party

in law. See Beazer E., Inc. v. Mead Corp., 412 F.3d 429, 447 (3d Cir. 2005) (taking into account

intent of parties even though the court had previously concluded that, as a matter of law, the

agreement did not provide indemnification for CERCLA response costs); Halliburton, 648 F.

Supp. 2d at 880-81, 884 (“[T]he issue is not whether the indemnity provisions apply to this case

or are enforceable, but what the parties intended as to allocation.”) Beyond the arguably

ambiguous text of the indemnification provisions, there is no evidence that the parties intended

the government to indemnify LPC for environmental contamination caused by routine disposals

of TCE and propellant wastes. The SRAM indemnification provisions were added to the SRAM

subcontracts pursuant to Public Law 85-804. (Trial Tr. at 217-18 (Johnson).) According to

Lockheed’s contracting expert, Mr. Johnson, Public Law 85-804 was a “special statute” enacted

“to handle catastrophic risks that insurance cannot possibly cover” (id. at 217) – namely a

“missile going astray and causing civilian casualties.” (Id. at 254.) Indeed, as described by the

Senate Report accompanying the law:

The need for indemnity clauses in most cases arises from the advent of nuclear
power and the use of highly volatile fuels in the missile program. The magnitude
of the risk involved under procurement contracts in these areas have rendered
commercial insurance either unavailable or limited in coverage.

S. REP. NO. 2281, 85th Cong., 2d Sess., at 3, reprinted in 1958 U.S.C.C.A.N. 4043, 4045.70 The

pollution here, while tragic, is by no means “catastrophic” in the sense envisioned by drafters of

seek relief greater than $10,000. 28 U.S.C. §§ 1346(a)(2), 1491(a)(1); see also Franklin-Mason v.
Mabus, 742 F.3d 1051, 1054-55 (D.C. Cir. 2014).
70
Further, Boeing first requested indemnification under Public Law 85-804 for its SRAM
contracts in 1966 “on the basis that performance under the contract will involve unusually hazardous risks
in the event of an incident of catastrophic proportion resulting from explosion, malfunction or
ground/flight accident precipitating a nuclear incident.” (PX506 at 1.)

78
Public Law 85-804 and could have been insured against at the time.71 Accordingly, the Court

gives no equitable weight to the indemnification provisions in LPC’s SRAM production

subcontracts.

B. LPC exercised significantly more control than the government over the day-
to-day hazardous waste disposal operations at the Sites.

As noted, the paramount equitable factor in this case is the comparative level of control

the parties exercised over disposal practices at the Sites, a variation on the fourth Gore factor’s

focus on “the degree of involvement by the parties in the generation . . . [and] disposal of the

hazardous wastes.” Envtl. Transp. Sys., 969 F.2d at 508. The Court analyzes this equitable

factor through the operator liability framework set forth in Bestfoods and its progeny.

Of course, because the parties have stipulated to liability, the Court is not required to

determine whether either party was an operator at the Sites. However, the Supreme Court’s

definition of operator liability in Bestfoods is helpful in delineating the types of control over

which CERCLA extends and thus which party should be more responsible as an equitable

matter.

Bestfoods limited operator liability under CERCLA to those parties who “manage, direct,

or conduct operations specifically related to pollution, that is, operations having to do with the

leakage or disposal of hazardous waste, or decisions about compliance with environmental

regulations.” 524 U.S. at 66-67. LPC clearly meets the Bestfoods standard. LPC employees

planned, managed, and performed the day-to-day operations that resulted in disposals of AP and

TCE at the Sites.

71
In addition, Lockheed’s argument that the SRAM indemnification provisions were intended to
cover routine environmental spills flies in the face of one of Lockheed’s central theses of the cases – that
no one knew that releases of TCE and propellant wastes onto the ground would cause environmental
harms. Having accepted Lockheed’s position (see supra Section II.A.6), the Court is not inclined to view
TCE and perchlorate pollution as posing “unusually hazardous” risks.

79
 Lockheed argues that the government is also an operator under Bestfoods because (1)

LPC’s contracts incorporated government safety manuals, which included disposal practices; (2)

DCAS approved process specifications, which included disposal procedures; (3) DCAS

inspectors observed production processes and occasionally undertook facility-wide safety

inspections; and (4) the government provided technical advice to LPC. (Trial Tr. at 31-32

(Lockheed opening); see also Lockheed’s Response to the U.S. Memo. on Operator Liability,

Feb. 25, 2014 [Dkt. No. 136] at 2-4.)

Even considering the height of government presence and influence at the Sites during the

SRAM production years, the Court disagrees. Although the government had a significant

presence and role at the Sites, there is no evidence that the government used its influence to

manage or control the day-to-day disposal of hazardous wastes there. See City of Wichita, 306 F.

Supp. 2d at 1055.

First, “courts have consistently held that contract provisions, specifications, and even

mandates similar to those expressed in the [manuals at issue] are insufficient to show ‘direction’

or ‘control’ over waste disposal for purposes of establishing operator liability.” Steadfast Ins.

Co. v. United States, 2009 WL 3785565, at *7 (C.D. Cal. Nov. 10, 2009) (collecting cases); see

also City of Moses Lake v. United States, 458 F. Supp. 2d 1198, 1227 (E.D. Wash. 2006) (where

Lockheed, as a defendant, successfully argued that “specifications that govern the operation of

missile maintenance facilities are relevant only insofar as they show that Lockheed managed or

directed not just any ‘operations,’ but operations having to do with the leakage and disposal of

hazardous waste.”). The many manuals and specifications at issue in this case are no exception.

The manuals permitted – but did not mandate – some of the common disposal operations LPC

used at the Sites, including the use of burn pits for propellant wastes. (See, e.g., PX0007 §§

80
1504-07 (allowing disposal of propellant waste by burning on bare ground, dumping at sea, or

destruction by detonation).72 And even assuming arguendo that the manuals guided LPC’s

decision to, for instance, use burn pits, this does rise to the level of daily control over waste

disposal operations at the Sites by the government. Indeed, Lockheed was free to seek waivers

from even mandatory provisions of manuals and such waivers were commonly granted if they

would have no effect on design or performance. (Trial Tr. at 941 (Nagle); cf. USX221.0002.)73

Likewise, the government is not an operator at the Sites simply because DCAS

representatives approved process standards that included disposal processes, observed certain

production processes, and conducted periodic safety inspections. DCAS’ approval, without

more, of process standards does not constitute the degree of “direct[ion]” necessary to establish

operator liability. Cf. United States v. Dart Indus., Inc., 847 F.2d 144, 145-46 (4th Cir.1988) (a

state regulatory body was not an operator for merely approving and inspecting disposal practices

at private site). No manufacturing process standard existed for the ultimate disposal of wastes,

either at the burn pits at the Redlands or Potrero Canyon facilities or at the waste disposal area at

the LaBorde Canyon facility. Thus, the record does not support Lockheed’s argument that the

United States “‘made, approved or ratified all significant operating decisions’ at the sites.”

(Lockheed Operator Br. at 5 (quoting Cadillac Fairview, 299 F.3d at 1022). Indeed, even the

process standards that did exist, although generally detailed, were vague as to what specific

72
However, the manuals did not direct or recommend the other disposal methods LPC used,
including the pouring of TCE or release of AP-contaminated wastewaters on the bare ground at Redlands.
73
Lockheed urges that the distinction between “recommendation” and “requirement” in this case
is a false one, citing Nu-W. Min. Inc. v. United States, 768 F. Supp. 2d 1082, 1090-91 (D. Idaho 2011).
(Lockheed Operator Liability Br. at 4-5.) Nu-West is distinguishable because in that case the government
was “actively involved in the design and location of the waste dumps, and in ensuring that the waste
dumps complied with the mining plans and environmental rules.” 786 F. Supp. 2d at 1091. Here, there is
no such evidence of specific, let alone pervasive, government direction as to the placement, design, or
operation of the burn pits, evaporation pits, sumps, or other waste disposal facilities or processes.

81
disposal practices should be followed. (See, e.g., USX32 § 1.3 (stating that the washing of

grinder parts “will be performed at the faucet and sump outside the lower level of Bldg. 77”

without specifying into which sump the water was to flow.) DCAS’ approval of such vague

process standards cannot signal any meaningful degree of government direction or control over

LPC’s waste disposal practices.

Further, DCAS’ day-to-day role at the Sites was limited to ensuring LPC’s compliance

with contract specifications for quality assurance purposes. (Nagle Decl. ¶¶ 40-41.) DCAS was

not contractually obligated to perform inspections at the Sites. The government had the right,

but not the duty, to inspect. (Id. ¶¶ 42-46; see also USX242.0003.) At least until 1970, the

record indicates that DCAS’ quality assurance inspections were insufficient both in quantity and

quality. (See PX577 at 117; see also USX221.0004.) In fact, the AFRPL criticized DCAS –

along with Boeing and LPC – for inadequate quality assurance procedures. (See PX577 at 117.).

Most importantly, there is no evidence that any of the DCAS quality assurance inspection points,

even at the height of DCAS oversight during the later SRAM years, included disposal – rather

than production – processes. (See generally USX264 (listing “mandatory production

certification inspection characteristics,” none of which concerned disposal of wastes).) See

Miami-Dade Cnty. v. United States, 345 F. Supp. 2d 1319, 1343 (S.D. Fla. 2004) (government

was not liable as an operator where Air Force inspectors had “no objective, duty, or

responsibility other than to enforce the . . . contract provisions by ensuring the delivery of quality

products.”).

DCAS’s sporadic (prior to 1970) and periodic (after 1970) safety inspections74 present a

closer issue, yet still do not support a finding that the government was an operator. Wastes and

74
There is evidence of one site-wide safety inspection by DCAS or its predecessor prior to 1969.
(See, e.g., PX321; PX471-472; PX1046.) Under SRAM, safety surveys were more common, but by no

82
waste (particularly propellant) disposal implicated worker safety, rather than environmental

contamination. (See Trial Tr. at 87 (Oppliger); see also PX0009 § 7-1.1.) As a result, some

DCAS safety inspections briefly touched on issues of waste. (See, e.g., PX471 at 1 (1960 –

questioning the efficacy of a drain from Building 52 to Evaporation Pit 61); PX0372 at 2 (1970 –

out of over fifty safety-related observations, noting broken grating over sump pump at Building

77); PX0484 at 3 (1972 – out of nineteen items of concern, noting improper handling and storage

of propellant trimmings)). However, these safety inspections – to the limited extent they

concerned wastes at all – did not make recommendations regarding the ultimate disposal of the

waste; rather, they were limited to the safe handling, storage, or transportation of waste prior to

the disposal processes that LPC chose to use. Moreover, LPC could – and on occasion did –

reject DCAS’s waste-related recommendations arising out of its safety inspections. (See PX484

at 3 (rejecting DCAS’s recommendation to use metal containers, rather than hamburger cartons

and water, when handling propellant trimmings); see also Trial Tr. at 1023-24 (Nagle).)

Ultimately, the sporadic nature of DCAS’s inspections, combined with their limited focus on

waste disposal activities per se, fails to demonstrate the level of frequent control over hazardous

waste disposal activities required for operator liability under Bestfoods

The same can be said for the government’s technical involvement at the Sites. During the

SRAM period, government representatives at the Sites from the SPO were primarily “observers”

pursuant to the TSPR initiative. (See, e.g., Trial Tr. at 1338-40 (Dull).) To the extent SPO or

AFRPL representatives gave technical direction to Boeing or LPC, the guidance related solely to

product development and performance, not issues such as safety or, more specifically, waste

disposal. (See, e.g., id. at 1353.) Inspections and guidance unrelated to waste disposal, no matter

means frequent. (See, e.g., PX474 (1969); PX372-373 (April 1970); PX476-477 (June 1970); PX398,
484-485 (November 1972).)

83
how pervasive, are not indicative of operator liability under CERCLA. See, e.g., Miami-Dade

Cnty., 345 F. Supp. 2d at 1343; State of Wash. v. United States, 930 F. Supp. 474, 485 (W.D.

Wash. 1996) (finding no operator liability when “[g]overnment inspectors and accountants had

no responsibility for directing activities that led to the deposit of the wastes. The primary

concern of the inspectors and accountants was efficiency and cost control.”).75

75
The evidence of actual “direction” regarding the disposal of any wastes is limited to a few
instances where the government abandoned property and instructed LPC – with LPC’s consent – to
destroy it by burning in the Potrero Canyon burn pits. (See PX0461; PX1073 at 1 (“Dispose [AP] by pit
burning at your Portereo [sic] plant.”).) These “occasional” instances of “direction,” involving a
miniscule percentage of the total wastes burned at the Sites, are insufficient to demonstrate the level of
“frequent” direction or control required for operator liability. See City of Wichita, 306 F. Supp. 2d at
1055.
These documents do, however, demonstrate the government’s liability as an arranger under
CERCLA § 107(a)(3) for the Potrero Canyon facility. Lockheed attempts to go a step further by arguing
that the government is liable as an arranger for all of the Sites based on its ownership of some of the TCE
and AP wastes and its benefits under contracts with LPC, pursuant to which the wastes were disposed.
(Lockheed’s Memo. on Arranger Liabilty, Feb. 12, 2014 [Dkt. No. 121] at 3-5.) AISLIC I’s contrary
holding notwithstanding, see 2010 WL 2635768, at *30, the Court has serious doubts that arranger
liability – with its focus on whether the party “planned for” the disposal – attaches to the government
based solely on the existence of output contracts and government title over wastes due to idiosyncratic
federal procurement regulations. See Burlington N., 556 U.S. at 612. A party “may not be held liable as
an arranger under CERCLA unless the plaintiff proves that the [party] entered into the relevant
transaction with the specific purpose of disposing of a hazardous substance.” Team Enterprises, LLC v.
W. Inv. Real Estate Trust, 647 F.3d 901, 909 (9th Cir. 2011). “Disposal of hazardous wastes must be a
purpose of the transaction, not merely a foreseeable byproduct of the transaction.” Pakootas v. Teck
Cominco Metals, Ltd., 832 F. Supp. 2d 1268, 1274 (E.D. Wash. 2011) (holding that Washington state was
not liable as an arranger for contamination that occurred as a foreseeable result of allowing mining
companies to operate mines on state lands.) Thus, where, as here, disposals of hazardous wastes occur as
a foreseeable but incidental result of a production process, arranger liability does not presumptively
attach. See Shell Oil, 294 F.3d at 1059; Pakootas, 832 F. Supp. 2d at 1274. Instead, arranger liability
attaches only if the government exercised direction and control over waste disposal activities related to its
contracts with LPC. See Shell Oil, 294 F.3d at 1055-56 (considering government’s control over waste
disposal at the site); Gen. Elec. Co. v. AAMCO Transmissions, Inc., 962 F.2d 281, 286 (2d Cir. 1992)
(same); Pakootas, 832 F. Supp. 2d at 1275 (same); AISLIC I, 2010 WL 2635768, at *30 (same). For this
reason, the Court’s analysis regarding control and direction for the purposes of operator liability
necessarily leads the Court to decline to adjust the government’s allocation as an equitable matter for
being an arranger at any facility other than Potrero Canyon.
The Court will, however, minimally adjust its equitable allocation for the government based on its
status as an arranger at the Potrero Canyon facility. The indemnification provisions that favor the
government for those arranged disposals (see PX0461 at 2; PX1073 at 1) and the fact that the arranged
disposals were limited to relatively small amounts of hazardous substances (see supra n.67), renders the
government’s arranger liability for the Potrero Canyon facility of limited importance.

84
 Perhaps recognizing the insufficiency of the government’s safety manuals and inspector

presence at the Sites, Lockheed argues that the government exercised such significant control

over the Sites by virtue of its monopsony over the solid rocket propellant industry as to establish

operator liability. (Trial Tr. at 31 (Lockheed opening); Lockheed Operator Br. at 1-2, 5-8.) In so

arguing, Lockheed clings to a pre-Bestfoods Third Circuit decision where the government was

found to be an operator for a contractor’s facility based on its pervasive authority over both the

site and the industry. (Lockheed Operator Br. at 1-2, 5-8 (citing FMC Corp. v. U.S. Dep’t of

Commerce, 29 F.3d 833 (3d Cir. 1994) (en banc).) In FMC, the en banc majority affirmed the

district court’s application of the “substantial control” test to hold the government was an

operator of a high tenacity rayon production facility owned and operated by American Viscose

during World War II. 29 F.3d at 843-45.76

Even assuming that FMC remains good law in the wake of Bestfoods,77 this case does not

present the pervasive levels of control exhibited in FMC and other World War II cases. See, e.g.,

76
The FMC en banc majority concluded that the government exercised significant day-to-day
control over the American Viscose facility because
American Viscose would not have been making high tenacity rayon if not at the
government’s direction. To obtain the commercial product it needed, the government
diverted American Viscose from its previous commercial endeavors. Thus, every day
American Viscose did what the government ordered it to do. Second, although the
government officials and employees personally did not take over the plant, the
government maintained a significant degree of control over the production process
through regulations, on-site inspectors, and the possibility of seizure. Third, the
government built or had built plants supplying raw materials to American Viscose,
controlled these plants, arranged for an increased labor force, and supervised employee
conduct, at least to the extent of helping American Viscose deal with labor disputes and
worker absenteeism. Fourth, the government supplied machinery and equipment for use
in the manufacturing process. Fifth, the government controlled product marketing and
price.
29 F.3d at 844.
77
The government argues that Bestfoods effectively abrogated FMC’s “substantial control” test.
(Trial Tr. at 1809 (gov’t closing).) Although the “substantial control” test is in tension with Bestfoods’s
focus on a party’s particularized control over hazardous waste disposal processes, see Miami-Dade Cnty.,
345 F. Supp. 2d at 1342, the Court need not decide if FMC’s remains good law.

85
Cadillac Fairview, 299 F.3d at 1022; Shell Oil, 294 F.3d at 1049-50.78 The government’s

monopsony over the solid propellant rocket industry does not alone make the government an

operator of the Sites. See E. Bay Mun. Util. Dist. v. U.S. Dep't of Commerce, 142 F.3d 479, 486

(D.C. Cir. 1998) (an output contract that “reflected the monopsonistic wartime market” did not

make the government an operator). Indeed, important under FMC was not only the fact that the

government created and held a monopsony over the high tenacity rayon market but also that it

forced American Viscose to participate in that market under the threat of takeover. FMC, 29

F.3d at 844. There is no comparable evidence here. LPC voluntarily – and repeatedly – bid on,

won, and completed significant government contracts and subcontracts.79 See E. Bay Mun. Util.

Dist., 142 F.3d at 486-87 (concluding that the record lacked evidence that the government

coerced plaintiff to operate mine).

78
In Cadillac Fairview, “[t]he government owned the land; the government owned the plant; the
government owned the raw materials; the government owned the byproducts and wastes; and the
government owned the [product –] rubber.” 299 F.3d at 1022. The government also had unrestricted
control over its contractor’s operations of the site, required monthly reports regarding hazardous waste
disposal, and agreed to indemnify the contractor for all costs. Id. at 1022, 1026.
79
Lockheed also cites the government’s provision of government-owned equipment and raw
materials, as well as putative control over personnel at the Sites, as evidence of the government’s
“pervasive” control. (Lockheed Operator Br. at 6-7.) The provision of government-owned equipment is
of limited importance to operator liability when, as in this case, the government does not operate, no less
control, that equipment. The same can be said for raw materials. Although the government’s provision or
aid in procuring the raw materials, such as AP, may be instructive for arranger liability, it is of limited
importance in determining who directed waste disposal operations at the Site
Similarly, Lockheed’s evidence that the government “exerted control over LPC’s personnel”
(Lockheed Operator Br. at 7) is unconvincing. In FMC, the government obtained draft deferrals for
personnel, directed workers from other industries to the American Viscose plants, provided housing for
the additional workers, resolved labor disputes, and had a full-time worker at the plant dealing with labor
issues. 29 F.3d at 837. In this case, there is evidence that on two occasions over a span of twenty years
the government made recommendations to LPC regarding personnel issues. (PX388 (recommending
removal of an LPC employee who reworked a nozzle design without informing superiors); PX577 at 118
(recommending SPO negotiation of LPC’s engineering workforce down by at least fifty percent upon
commencement of SRAM production).) And there is no evidence that either recommendation was ever
followed. (See Trial Tr. at 1377 (Dull) (noting the negotiations in PX0577 never took place).) Based on
this limited evidence, the Court cannot conclude that the government exerted any, let alone pervasive,
control over LPC’s personnel.

86
 Relatedly, while Lockheed’s argument that the government determined “‘what product

the facility would produce, the level of production, the price of the product, and to whom the

product would be sold’” (Lockheed Operator Br. at 8 (quoting FMC, 29 F.3d at 843)), is

technically correct, it is ultimately irrelevant without evidence of government coercion. For

LPC’s argument, taken to its logical conclusion, would render the government an operator for

practically any military output contract. Such a conclusion is inconsistent with Bestfoods’s

requirement that operator liability is concerned first and foremost with control over “operations

having to do with the leakage or disposal of hazardous waste,” 524 U.S. at 66-67, and is

expressly foreclosed by Circuit precedent. See E. Bay Mun. Util. Dist., 142 F.3d at 486

(“[E]ntering into an output contract does not make the government an operator.”). Thus, without

evidence that the government coerced LPC to enter into its solid propellant rocket contracts, the

general terms – e.g., product, quantity, and price – of those contracts cannot form the basis of a

finding of operator liability. See Rospatch Jessco Corp. v. Chrysler Corp., 962 F. Supp. 998,

1005-06 (W.D. Mich. 1995) (the government was not an operator where it did not “twist [the]

arm” of contractor to produce wartime materials, and the government’s control over site was

limited to process specifications and inspections).

Accordingly, considering the totality of the circumstances, the Court concludes that the

government was not an operator of the Sites. To be sure, all of LPC’s operations at the Sites

were in performance of government contracts or subcontracts and the government had a

pervasive influence over general activities at the Sites, whether through process specifications,

safety manuals, inspections, or technical direction. However, the government did not manage,

direct, or otherwise control on a frequent basis the day-to-day hazardous waste disposal activities

at the Sites. See Coeur D’Alene Tribe v. Asarco Inc., 280 F. Supp. 2d 1094, 1127-30 (D. Idaho

87
2003) (government is not liable as an operator in case with “pervasive involvement of federal

government” including knowledge of “how the waste material was disposed of and that it was

done in accordance with the customary and usual practices of the time” because the “federal

government did not make the day-to-day decisions regarding operations of [disposal]”); United

States v. Iron Mountain Mines, Inc., 987 F. Supp. 1277, 1287-88 (E.D. Cal. 1997) (same).

Because LPC was the sole operator of the Sites under Bestfoods, the Court concludes that

Lockheed should shoulder a large proportion of the liability for response costs at the Sites.

C. The government acquiesced in many of LPC’s disposal operations at the
Sites.

That the government was not an operator of the Sites under Bestfoods does not, however,

shift the entire equitable allocation to Lockheed. Although evidence that a party knew of

another’s disposal practices is insufficient to impose either operator or arranger liability, see,

e.g., Burlington N., 556 U.S. at 610 (arranger); Coeur D’Alene Tribe, 280 F. Supp. 2d at 1127-30

(operator), courts often consider the “acquiescence of the parties in the contaminating activities”

as a factor in equitable allocation. See, e.g., Weyerhaeuser, 771 F. Supp. at 1426; see also

Cadillac Fairview, 299 F.3d at 1025.

Even though the government did not direct or control LPC’s day-to-day hazardous waste

disposal activities at the Sites, it was aware of and acquiesced in many of them. The government

contracted with LPC (and Boeing) with the knowledge that “[d]isposal . . . should be regarded as

an integral part of solid propellant rocket operations” (PX0009 § 7-1.1 (1973 Air Force Manual);

see also PX431 (“[LPC] will generate under normal operations approximately 10,000 pounds of

waste materials per month.”), and wrote the manuals that provided general recommendations for

waste propellant disposal procedures. (See, e.g., PX0007; PX0009.) It would be inequitable for

the Court to allocate to Lockheed full responsibility for the response costs at the Sites when the

88
government could have anticipated, and in some instances knew, how LPC disposed of the TCE

and propellant wastes created during the performance of its government contracts. See

Weyerhaeuser, 771 F. Supp. at 1424-26. Thus, the Court will allocate the government an

equitable share based on its acquiescence.

That said, the level of government acquiescence varied among the Sites. Government

presence and acquiescence was at its greatest at the Redlands facility. The full-time DCAS

representatives were located there, the majority of DCAS inspection points were located there,

and the inspections (safety and otherwise) focused on operations there. (See Trial Tr. at 1357;

e.g., PX476 at 109.) Even though the government’s technical observers and DCAS inspectors

were not focused on the ultimate acts of disposal at Redlands – e.g., evaporation pits and burn

pits – it is improbable that the government representatives were unaware of how Lockheed

managed its wastes. Indeed, there is direct evidence that government representatives at least

knew of both LPC’s use of evaporation pits (e.g., PX471 at 1) and burn pits at Redlands. (E.g.,

Trial Tr. at 1380 (Dull); PX453.)

Government presence was lower, but still significant, at the Potrero Canyon facility.

While DCAS inspections focused on the Redlands facility, they also covered some of the

production operations at the Potrero Canyon facility. (See, e.g., PX474; PX476 at 109; PX0479

at 304.) The government also knew of LPC’s burn pit operations at the Potrero Canyon facility

and on several occasions it instructed LPC to burn abandoned government property there. (See

PX0461; PX1073 at 1; see also supra n. 75.) Further, although there is no evidence that the

government directed LPC to hog out defective rocket motors, the government was aware that

LPC was using this procedure. (See Trial Tr. at 1352 (Dull); PX326 at 3; PX550 at 46.).)

89
 The government had the least involvement with the LaBorde Canyon facility. The

government knew of and probably observed a significant number of rocket motor tests in the

Test Bay Canyons. The government also knew, to the extent it occurred at the LaBorde Canyon

facility, that LPC hogged out defective rocket motors to reuse the casings. (See Trial Tr. at 1352

(Dull); PX326 at 3; PX550 at 46.).) However, there is little evidence of government inspections

– safety or otherwise – at LaBorde Canyon facility. There is also no evidence that the

government had any oversight over the Waste Disposal Area.

D. Some of LPC’s disposals at the Sites violated internal LPC rules or
government requirements

Of course, the government did not know of and acquiesce in all of LPC’s disposal

practices, including many that are sources of the contamination at the Sites. Indeed, in several

instances LPC violated its own internal rules or a rare government requirement with regard to the

handling and disposal of waste solvents and propellants. The Court focuses only on three

violations that best demonstrate LPC’s lack of due care at the Sites. Whether a result of

inadequate training, poor oversight, or sloppy practices in general, these instances favor an

upward adjustment to Lockheed’s equitable allocations for the Redlands and Potrero Canyon

facilities.80

The disposal of TCE on the bare ground at the Redlands facility (see supra Section I.A.2)

is the most prominent example of LPC employees violating LPC’s rules. As the Court has

80
This consideration falls within the fifth Gore factor – “the degree of care exercised by the
parties with respect to the hazardous waste concerned.” Envtl. Transp. Sys., 969 F.2d at 508. Under that
same factor, the Court has previously concluded that LPC did not violate any generally recognized
standard of care by pouring liquid TCE or propellant wastes, or burning propellant wastes, on the bare
ground. (See supra Section II.A.6.) Here, the Court asks the different question of whether LPC failed to
live up to its own rules or shirked any government requirements. Just as the government should have
been able to rely on LPC to operate according to generally recognized standards of care, the government
should have been able to rely on LPC to follow its own rules and any relevant government requirements.

90
concluded, pouring TCE and other solvents on the bare ground did not violate any standard of

care generally recognized in the 1950s through mid-1970s. (See supra Section II.A.6.)

Nonetheless, recognizing the safety risks of pouring AP-contaminated solvents on the bare

ground, LPC constructed evaporation pits for the disposal of solvent wastes. (Trial Tr. at 95, 99

(Oppliger); USX977 at 48-49 (Stickney Dep.).)81 LPC memorialized its policy for the collection

of solvents and disposal in evaporation pits in its Standard Operating Procedure 11 (“SOP-11”).

(See PX961.) SOP-11 stated, inter alia, that “[a]t no time are solvents to be poured on the

ground or in the water [settling] basin south of Bldg. 114.” (Id. at 2.) Several witnesses testified

that SOP-11 was more than hortatory. Mr. Oppliger testified that, even though he did not

manage waste disposal processes, he would have “definitely” stopped LPC employees from

dumping solvent on the bare ground because such disposals were “dangerous” and “would not be

allowed.” (Trial Tr. at 95, 99, 108-10 (Oppliger).) Similarly, Mr. Donald Eastman, who worked

as a process operator and later as foreman at LPC from 1956 to 1974, testified that he was

“indoctrinated” to “not dump chemicals in the ground.” (USX895 at 5-9, 106-07 (Eastman

Dep.).) Nonetheless, whether as a result of inadequate training (E.g., USX987 at 21 (Wessman

Dep.), poor oversight (e.g., id. at 26), or sloppy practices, several LPC employees testified to

routinely violating SOP-11. Had LPC better enforced SOP-11, some portion of the TCE

contamination in the Redlands plume probably could have been prevented.

Second, LPC’s washing of grinder parts and bags into Building 77’s south sump at the

Redlands facility (see supra Section I.A.1) also violated the company’s internal protocols. As

Mr. Delaney opined, LPC’s process specifications treated “AP-contaminated wastewater . . . no

differently than dry AP.” (Delaney Decl. ¶ 24.) Although the process specifications for washing

81
After the mid-1960s, LPC no longer used evaporation pits for solvent wastes, but instead it sent
contaminated solvents directly to burn pits. (See Trial Tr. at 808-09 (Feenstra).)

91
grinder parts and bags did not specify how to treat AP-contaminated wastewater from grinder

part and bag washing (see USX32 § 1.3), manufacturing process standards pertaining to Building

77 specifically stated that AP-contaminated “water is to be discarded as waste propellant.”

(PX1023 § 6.31.1; see also PX1043 §100.6.5 (“Label and treat the drum of [contaminated] water

as waste propellant.”).) Under LPC’s safety standard for propellant wastes, waste propellant was

to be collected in drums and transported to the burn pits. (See PX1061 at 885-87.) As Mr.

Oppliger testified, disposal of AP-contaminated wastewaters onto the bare ground “would not be

allowed.” (Trial Tr. at 95 (Oppliger).) Had this admonition been followed, a substantial portion

of the perchlorate contamination in the Redlands plume probably could have been prevented.82

Third, LPC’s burial of propellant wastes constituted a clear violation of a government

requirement.83 While the government did not mandate any specific method for LPC’s disposal

of propellant wastes (see supra Section II.B), it did explicitly proscribe the disposal of solid

propellant wastes by burial. (See USX47 § 2704 (1951 Army Safety Ordnance Manual)

(“Collected explosive wastes must not be disposed of by being buried . . . .” (emphasis added)).)

At some point prior to 1963, LPC violated this government proscription by burying propellant

wastes at the Potrero Canyon facility. (See USX55.)84 The record provides no indication where

82
The foregoing conclusion also applies to the LPC employees’ release of AP-contaminated
wastewaters onto the bare ground after washing down the interiors of buildings. (See supra Section
Section I.A.1.) However, it does not apply to LPC’s hog out operations. Although hogging out defective
rocket motors onto the bare ground did violate LPC’s general protocol regarding the disposal of AP-
contaminated wastewaters, the Court does not consider this violation to be of significance because the
government was aware of the process. (See supra Section II.C.)
83
Additional violations of LPC protocols and government recommendations regarding AP wastes
included pouring of propellant scraps and wastewater directly into burn pits rather than burning them in
drums (compare USX1002A.0001 (Wright Decl.), with PX0009 § 7-2, 7-3 (recommending propellant
cuttings in water to be burned in drums); PX961 (SOP-11)), and the failure for several years to connect
the pipe between Building 52 to Evaporation Pit 61. (See supra Section I.A.1.)
84
Lockheed claims that LPC buried “inert propellants” at the Potrero Canyon facility. However,
the fact that in 1963 LPC was concerned about the presence of the buried propellants “in the watershed”

92
LPC buried these wastes, how much was buried, whether the wastes were ever removed, or

whether the wastes contributed to the perchlorate contamination at the facility. (See Sterrett

Decl. ¶ 223.) Nonetheless, LPC’s flagrant violation of the Safety Ordnance Manual necessitates

a slight upward adjustment in Lockheed’s equitable allocation for the Potrero Canyon facility.

E. Conclusion under traditional equitable allocation

Based on these above findings of facts and conclusions of law, were the Court to end its

equitable analysis here, it would allocate liability for response costs (pasts and future) at the Sites

as follows: at the Redlands facility, a 30% share of liability for the government and a 70% share

of liability for Lockheed; at the Potrero Canyon facility, a 25% share of liability for the

government and a 75% share of liability for Lockheed; and at the LaBorde Canyon facility, a

20% share of liability for the government and an 80% share of liability for Lockheed.

III. EFFECT OF INDIRECT RECOVERIES ON EQUITABLE ALLOCATION

However, no analysis can be complete without consideration of the novel issue of what

effect, if any, Lockheed’s indirect recovery of significant percentages of its response costs for the

Sites through U.S.-government contracts should have on the Court’s equitable allocation of those

same response costs between the Lockheed and the government. Lockheed urges that its indirect

recoveries from the government should have no effect on its ability to recover directly from the

government under CERCLA. In contrast, the government argues that allocating it a CERCLA

share would amount to impermissible “double recovery.” Because of the significant economic

and legal distinctions between past and future response costs, the Court considers their

allocations separately.

and considered it necessary to “mark[] with some type of pole or flag” the location of the propellants in
preparation for construction leads the Court to conclude that not all of the buried propellants were inert.
(USX55.)

93
 A. Lockheed’s recovery of past response costs would unfairly burden the
taxpayer.

Lockheed has indirectly recovered from the government through higher contract prices

over 72% of its past response costs for the Sites. (Meyer Decl. ¶ 29 & fig. 5.) Thus, the

government’s “effective share” is already well over two times higher than its equitable share for

the Sites as determined in Section II.E. Moreover, based on the most recent data on the U.S.-

government share of Lockheed’s business, U.S.-government contracts would receive the benefit

of only 87% of any CERCLA payment made by the government for past response costs.85

(Wright Decl. ¶ 74; see Trial Tr. at 1677-78 (Gatchel).) Thus, any allocation payment made by

the United States would cause their effective share to rise even further beyond their equitable

share. This, the government argues, is double recovery. (E.g., Trial Tr. at 1891 (government

closing).)

“CERCLA expressly prohibits double recovery for response costs.” Boeing Co. v.

Cascade Corp., 920 F. Supp. 1121, 1133 (D. Or. 1996). However, this prohibition is fairly

narrow, in that it only applies to bar CERCLA recovery for costs already compensated “under

any other Federal or State law.” 42 U.S.C. § 9614(b). Because of the narrowness of the

statutory bar on double recovery, courts have developed a broader equitable double recovery

theory based on the principle that “permitting a CERCLA contribution-action plaintiff to recoup

more than the response costs he paid out of pocket flies in the face of CERCLA’s mandate to

apportion those costs equitably among liable parties.” Friedland, 566 F.3d at 1207. The theory

85
Under the DiscOps Pool, the government benefits from the same percentage of credits in a
given year as it incurs costs – a feature Lockheed refers to as the “mirror image” principle. (See Trial Tr.
at 1679 (Gatchel).) Thus, if Lockheed recovered $69 million of its past response costs for the Redlands
facility from the government pursuant to this action (30% of Lockheed’s $231 million in total past
response costs for the facility), under the current Lockheed contract mix, only 87% ($60 million) would
pass through as credits to benefit U.S.-government contracts. The remaining 13% ($9 million) would
pass through as credits to the benefit non-U.S. government contracts. (Wright Decl. ¶ 74; see Trial Tr. at
1678 (Gatchel).)

94
also comports with the concept that “the environment is the injured party, not the plaintiff.”

United Alloys, Inc. v. Baker, 797 F. Supp. 2d 974, 1002 (C.D. Cal. 2011). “In other words,

Plaintiffs have not been damaged and are not ‘entitled’ to money as a damaged party; but rather,

Plaintiffs can only receive reimbursement for the costs they expended beyond their share of

actual responsibility for the environmental damage.” Basic Mgmt. Inc. v. United States, 569 F.

Supp. 2d 1106, 1124 (D. Nev. 2008) (emphasis added). Plaintiffs “cannot make a profit on the

contamination.” Vine St., 460 F. Supp. 2d at 765. The effect of an equitable bar on double

recovery is simple: if a party has recovered from a collateral source, the amount of that recovery

is subtracted from the total pool of allocable costs.

Courts have applied equitable principles to bar double recovery in circumstances where

plaintiff-PRPs have received payouts from insurers, see, e.g., Yankee Gas, 852 F. Supp. 2d at

255-56; Basic Mgmt., 569 F. Supp. 2d at 1125; formal settlements with other PRPs, see, e.g.,

K.C.1986 Ltd. P’ship v. Reade Mfg., 472 F.3d 1009, 1017 (8th Cir. 2007); and informal

payments from other PRPs. See, e.g., Vine St., 460 F. Supp. 2d at 766 (so holding even though,

under the agreement, plaintiff had promised to pay back the PRPs for all reimbursements

received). However, as Judge Robertson correctly observed, those cases are distinguishable

from this case because they all considered direct payments to the plaintiff that, if not taken into

account, could amount to a windfall for the plaintiff. Lockheed Martin Corp., 664 F. Supp. 2d at

19.

In that vein, Lockheed argues that this case is more analogous to the utility rate recovery

cases where courts have concluded that double recovery is not an equitable concern. See Yankee

Gas, 852 F. Supp. 2d at 256; N.Y. State Elec. & Gas Corp. v. FirstEnergy Corp., 808 F. Supp. 2d

417, 528-29 (N.D.N.Y. 2011). In those cases, plaintiff-utilities formally recovered their response

95
costs through increased rates charged to ratepayers. The defendants in both cases argued that the

recovery of response costs from ratepayers should equitably bar plaintiffs’ recovery of those

response costs under CERCLA, lest the plaintiffs receive a windfall. See Yankee Gas, 852 F.

Supp. 2d at 256; FirstEnergy, 808 F. Supp. 2d at 528. Both courts disagreed, concluding that the

plaintiffs’ recoveries from defendants posed no large risk of double recovery or windfall.

Yankee Gas, 852 F. Supp. 2d at 256; FirstEnergy, 808 F. Supp. 2d at 529. As the Yankee Gas

court explained:

Money recovered from [defendant] will allow DPUC to reduce the amount it
allows Yankee Gas to collect from its utility customers during future rate cycles.
Thus, money contributed by UGI is money that will not have to be paid by utility
customers in Connecticut. The “windfall,” if that is the word, goes to the rate
payer, not Yankee Gas. This is in keeping with “CERCLA’s goals of cleaning up
environmental contamination and making sure that responsible parties, rather
than taxpayers, bear the costs.”

852 F. Supp. 2d at 256 (quoting Marsh, 499 F.3d at 182) (emphasis in original).

This case, however, is distinguishable from Yankee Gas and FirstEnergy. Here, the

principal “ratepayer” and the defendant-PRP are one and the same –the U.S. government. In

such circumstances, careful consideration of the recovery crediting scheme is necessary to ensure

that the plaintiff does not benefit from double recovery at the expense of the taxpayer. See R.W.

Meyer, 932 F.2d at 572 (“‘The hallmark of a court of equity is its ability to frame its decree to

effect a balancing of all the equities and to protect the interest of all affected by it, including the

public.’” (quoting Kay v. Mills, 490 F. Supp. 844, 855 (E.D. Ky. 1980)).

The DiscOps Pool’s crediting mechanism attempts to prevent “double recovery,” at least

as traditionally understood. Lockheed must allocate its CERCLA response cost recovery to the

DiscOps Pool (see USX1033 ¶ 4.7), and credits in the DiscOps Pool are passed through

Lockheed’s contracts in the same way as costs. (Trial Tr. at 1678 (Gatchel).) As in the utility

rate recovery cases, one hundred percent of any CERCLA recovery ultimately flows to the

96
ratepayers (i.e., the U.S.-government and Lockheed’s other clients) and not Lockheed. Thus,

there is no “double recovery” in the traditional sense because Lockheed cannot recover more in

response costs than it initially paid, and there is little potential for a windfall to the plaintiff from

the crediting system. (Cf. id. at 594-95 (Wright).)

To be sure, under any scenario where the government is allocated an equitable share of

past response costs, it will be worse off and shoulder a larger “effective share” of Lockheed’s

response costs than it does now.86 (Meyer Decl. ¶ 33.) But an increase in the government’s

“effective share” does not alone amount to a “double recovery.” For “double recovery” focuses

on the projected post-recovery economic position of the plaintiff, not the defendant. See

Friedland, 566 F.3d at 1207 (framing the issue as whether the plaintiff “recoup[s] more than the

response costs he paid out of pocket”).

Moreover, the government has been complicit in designing the very system about which

it so bitterly complains. The FAR allows indirect costs to be charged to government contracts

and the DCAA has taken the position that environmental cleanup costs at discontinued sites

generally constitute indirect costs. (See PX1862 ¶ 7-2120.7.) More importantly, in 2000, the

government negotiated with Lockheed and signed the DOSA, which blessed the DiscOps Pool

and its cost allocation and crediting scheme and explicitly stated that it did not settle any claims

arising under CERCLA. (USX1033 ¶ 4.18.) The DOSA also recognized the coexistence of

indirect contract and direct CERCLA recoveries by disallowing certain costs and credits from –

rather than nullifying wholesale – the Burbank Consent Decree. (Id. ¶¶ 3.1-.3.) All the while, the

government agreed, both pre- and post-DOSA, to toll the CERCLA statute of limitations for the

86
In nominal terms, and without accounting for taxes levied on Lockheed profits, the detriment to
the government of a CERCLA allocation for past costs can be expressed as: Detriment = (1 – recovery
rate from U.S. government) x Equitable Allocation x Past Response Costs.

97
Redlands facility, while allowing Lockheed to indirectly recover response costs through U.S.-

government contracts.

Under these facts, the government cannot fairly assert, as it seems to here, that it was

blindsided by Lockheed’s decision to file a CERCLA claim for the Sites or that DOJ should not

be bound by DOD’s decisions regarding government procurement contracting.87 Nor will the

Court in equity save the government from the natural and probable consequences of its own

conduct. Thus, the Court finds that there is no “double recovery” in this case and that – all other

things being equal – it would not be inequitable for the government’s effective share of past

response costs to increase as a result of Lockheed’s recovery of response costs from the

government under CERCLA.

Double recovery aside, the Court is nonetheless concerned about the economic benefit to

Lockheed and the economic detriment to the taxpayers from any CERCLA recovery of past costs

in this case. Framing the issue as one of economic benefit rather than double recovery serves the

important purpose of preventing Lockheed from profiting from CERCLA – beyond the mere

recovery of response costs – at the expense of the taxpayer.88

87
Importantly, Burbank Consent Decree – signed by DOJ –explicitly recognized the possibility of
an agreement like the DOSA, which was signed only months later by Lockheed and DCAS. (See PX1844
¶ 3.25; USX1033.) The DOJ also signed the CERCLA tolling agreements applicable to the Redlands
facility, including one agreement that was executed less than two months before the DOSA was signed.
(See PX1849.)
88
In its closing rebuttal argument, Lockheed cited TRW, Inc. v. United States, 28 Fed. Cl. 155
(Fed. Cl. 1993), in support of its position that the government cannot avoid its obligation to make direct
payments under the law on the basis that it had already made indirect payments through government
contracts regarding the same contractor costs. (Trial Tr. at 1966-68 (Lockheed closing.) In TRW, a
contractor sought to recover from the government $2.9 million in bid and proposal (“B & P”) costs
associated with preparing a bid proposal for an IRS request for proposals on the ground that the IRS did
not consider the proposal “fairly and honestly.” 28 Fed. Cl. at 156-57. The government argued that the
contractor had indirectly recovered $2.1 million of those B & P costs through advance agreements with
the DOD. Id. at 157. The court denied the government’s motion for summary judgment because, under
the advance agreement accounting system, the contractor would have received the same total in payments
whether or not it had bid on the IRS proposal. Id. at 160. In essence, the government failed to

98
 In this regard, the Court credits Dr. Meyer’s opinion that economic benefit should be

analyzed from the status quo – i.e., the parties current, pre-allocation positions based on the

response costs incurred and indirect recoveries made over the past twenty years. (Meyer Decl.

¶¶ 54-57, 138-39.)89

Aside from having already indirectly recovered far more of its response costs through

U.S.-government contracts than the government’s equitable allocation for the Sites, Lockheed

has also benefitted significantly by charging the government a profit factor on those response

demonstrate that it had indirectly paid the contractor $2.1 million dollars for the relevant B & P costs. Id.
at 162. This case is inapposite here. First, it is undisputed that Lockheed has indirectly recovered from
the government over $200 million in response costs associated with the Sites, thus undermining the
accounting basis for the TRW decision. It is not clear what would have happened in TRW had the
government proven that the contractor had indirectly recovered $2.1 million of the relevant B & P costs
through advance agreements with the DOD. Second, and more importantly, the TRW court was not
considering how to equitably allocate liability between two responsible parties.
89
Mr. Kiefer criticized Dr. Meyer’s definition of the status quo as the baseline for economic
benefit analysis and seemed to suggest that the proper analysis would compare Lockheed’s comparative
economic benefit under a CERCLA recovery with the hypothetical baseline assuming the government had
directly reimbursed Lockheed for its equitable share as the costs were incurred. (Cf. Trial Tr. at 1580
(Kiefer); Kiefer Decl. ¶¶ 3-4.) The Court rejects Mr. Kiefer’s proposed baseline for several reasons.
First, it is common practice in an economic benefit analysis to use a status quo baseline. This is so
because economic benefit analyses addressing hypothetical futures are most useful when compared to
real, present-day conditions rather than hypothetical past occurrences. (See Trial Tr. at 1512 (Meyer);
Meyer Decl. ¶ 138.) Indeed, as a practical matter, Mr. Kiefer’s baseline would have required, as early as
1994, the government to have known and directly reimbursed Lockheed for its equitable share of costs –
the very issue before this Court some twenty years later.
Second, Mr. Kiefer provides no quantitative analysis to defend his “conceptual” opinion that,
when compared to his proposed baseline, Lockheed would not receive an economic benefit from a
CERCLA recovery. (Trial Tr. at 1573-75 (Kiefer).) Although Mr. Kiefer testified that he “believe[d] it
would be possible” to perform a quantitative analysis supporting his conceptual opinion, Lockheed did
not ask him “to make an affirmative calculation of economic impact.” (Id. at 1574.) Finally, Dr. Meyer’s
analysis, even though using a status quo baseline, does take into account what has happened in the past:
that Lockheed incurred costs and recovered them (plus a profit) on an amortized basis. (Trial Tr. at 1512-
13 (Meyer).) Because it neither ignores the past nor relies on unrealistic hypotheticals, the Court
concludes that Dr. Meyer’s status quo baseline provides the most equitable framework for determining
how much a CERCLA recovery benefits Lockheed.

99
costs.90 Mr. Wright estimated this pre-tax profit factor to be six percent. (Wright Decl. ¶ 85; see

also Meyer Decl. ¶ 101 n.52.) Lockheed’s public filings, however, reveal significantly higher

pre-tax profit margins over the years. (See USX383.0030 (9.0% for 1998, 8.2% for 1996, 8.5%

for 1996); USX397.0030 (9.4% for 2012, 8.6% for 2011, 9.0% for 2010, 10.2% for 2009, 11.7%

for 2008).) Even assuming Mr. Wright’s conservative six-percent profit markup, Lockheed

effectively recovered at the expense of the taxpayers six cents on every dollar of its response

costs passed on to U.S.-government contracts. (See Wright Decl. ¶ 84.) All told, these pre-tax

profits amounted to approximately $11.8 million in nominal dollars (see USX433A.0007-08), or

adjusting for the time value of money, approximately $17.2 million in net present value terms.

(Id. at 0012-13.)

Mr. Kiefer opined, without conducting any quantitative analysis, that any recovery from

the government in this case would cancel out past profits because credits reduce profits in the

same way that costs create profits. (See Kiefer Decl. ¶¶ 32-33; cf. Meyer Decl. ¶¶ 42, 160

(“[T]he credit associated with the CERCLA payment results in lower profits for [Lockheed]

(from lower costs passed through to contracts with the United States).”). However, as Lockheed

admits (Trial Tr. at 1966 (Lockheed closing)), this argument ignores an important component –

the significant time value of money benefits (over $5 million pre-tax) that accrued to Lockheed

from its past profits on indirect recoveries from the government.

Mr. Kiefer faulted the government for not directly paying its share of response costs all

along and opined that the government’s unwillingness to pay directly for response costs in the

past harmed Lockheed. (E.g., Kiefer Decl. ¶¶ 3-4; Trial Tr. at 1579-83 (Kiefer).) The facts do

not support Mr. Kiefer’s conclusion. Lockheed’s historical rate of indirect recoveries from the

90
Lockheed identifies as its “underlying tenet in pricing [its] contracts with the U.S. government”
as its “ability to recover [its] costs plus profit, regardless of the type of contract.” (USX407 at 4-5
(emphasis added).)

100
government – which again is many times higher than the government’s equitable allocations for

the Sites – and the time value of the substantial profits that Lockheed has realized on these

indirect recoveries, make it difficult, if not impossible, to conclude that Lockheed has been

injured by the government’s lack of direct payments over the last twenty years. Quite the

opposite, in fact. Lockheed has benefitted greatly by recovering more money related to response

costs (due to its profit factor) than it has spent.91

Of course, profiting off of the cleanup of hazardous wastes is not per se undesirable or

improper. CERCLA’s bona fide prospective purchaser exemption, 42 U.S.C. § 9607(r)(1),

makes clear that, in some instances, cleanup of hazardous wastes should be a profitable

enterprise, lest no one voluntarily undertake the important task. See Small Business Liability

Relief and Brownfields Revitalization Act, Pub. L. No. 107–118, 115 Stat. 2356 (2002)

(providing an array of CERCLA amendments intended “to promote the cleanup and reuse of

brownfields”). However, courts have consistently considered as an equitable factor the

“economic benefits realized by a party as a result of remediation efforts.” City of Wichita, 306 F.

Supp. 2d at 1101 (collecting cases); see also FirstEnergy, 808 F. Supp. 2d at 533; Litgo, 2011

WL 65933, at * 9. And CERCLA provides no indication that responsible parties should profit

from the cleanup process at the expense of other responsible parties. Cf. Vine St., 460 F. Supp.

2d at 765.

91
Lockheed attempts to rebut this conclusion by suggesting, through the testimony of Mr.
Gatchel, that it lost contracts due to “overall cost,” including environmental response costs. (Trial Tr. at
1674 (Gatchel).) However, Mr. Gatchel did not identify any specific contract that Lockheed lost because
of environmental response costs at the Sites. Without such evidence, the Court will not credit Lockheed’s
argument that response costs for the Sites – even totaling tens of millions of dollars annually – caused
Lockheed to lose any government contracts, especially given Lockheed’s annual net sales of over $45
billion in each of the past five years. (See Press Release, supra n.22 (2013); USX397.0002 (2012);
USX396.0002 (2011); USX395.0002 (2010); USX394.0002 (2009).)

101
 Were the economic benefit to Lockheed limited to the time value of the profits it has

already earned on indirect recoveries through its U.S.-government contracts, the Court might not

be inclined to exempt the government from paying an equitable share of the past response costs

at the Sites. For as Lockheed contends, it merely seeks to recover as much of its past response

costs as possible under CERCLA to reduce costs for its clients and improve its own competitive

position. (See Trial Tr. at 1667-68 (Gatchel).) However, even accepting these motives,

Lockheed will receive three significant windfalls – all at the expense of the taxpayers – if the

Court allocates the government an equitable share of past response costs.

First, as the parties agree, CERCLA § 107(a) mandates the award of prejudgment

interest.92 42 U.S.C. 9607(a)(4)(D); see K.C.1986 Ltd. P’ship, 472 F.3d at 1018. This statutory

requirement applies with equal force when – as here – a plaintiff has sued under § 107(a) and the

defendant counterclaims under § 113(f). See Litgo, 725 F.3d at 392; Bancamerica Commercial

Corp. v. Mosher Steel of Kansas, Inc., 100 F.3d 792, 799-801 (10th Cir. 1996). The purpose for

awarding prejudgment interest was clearly explained by the Tenth Circuit:

Failure to grant prejudgment interest on contribution awards may . . . result in
inequitable apportionment, because parties awarded contribution will still have
lost the time value of the money they spent on behalf of other liable persons, and
those persons will have gained an equal amount. Further, refusal to grant
prejudgment interest is a disincentive for private parties to voluntarily undertake
cleanup actions because they will lose the time value of the money they spend on
behalf of other persons. Indeed, it would create a perverse incentive for
responsible parties to delay involvement in cleanups, because as they delay, they
gain the time value of the funds they should be investing in the cleanup.

Bancamerica Commercial Corp., 100 F.3d at 801.

This case, however, implicates none of these policy concerns. There is no loss based on

the time value of money because Lockheed indirectly recovered from the government much

92
See Lockheed Memo. on the Availability of Prejudgment Interest, Feb. 20, 2014 [Dkt. No. 128]
at 1-3; Trial Tr. at 1454 (government counsel).

102
more than the government’s equitable share of the response costs for the Sites through U.S.-

government contracts as it incurred the costs.93 Nor would the lack of prejudgment interest in

this case have disincentivized cleanup efforts. Lockheed was ordered to clean up the Sites by the

California environmental authorities and, in any event, it was able to indirectly recover its

response costs (plus a healthy profit) from the U.S. government and its other clients as it incurred

them. Indeed, as the Court has already explained, Lockheed actually benefitted from its cleanup

efforts through the profits it gained on the response costs that flowed down to its U.S.-

government contracts. For these reasons, the purposes of prejudgment interest are inapplicable

to this equitable allocation case.

A rough calculation of the amount of prejudgment interest potentially available in this

case only further underscores the Court’s conclusion. Assuming a demand date in 2000,94 and

that all of Lockheed’s past responses costs qualify for recovery under CERCLA, the potentially

available prejudgment interest in this case totals over $61 million dollars – over 20% of

Lockheed’s total past response costs for the Sites.95 Applying the Court’s above-determined

93
Of course, Lockheed’s amortization of response costs over five years has time value of money
implications. Both parties lament, in their own way, the “interest free loans” they provided each other
during the respective cost and credit amortization periods in the DiscOps Pool. (Kiefer Decl. ¶¶ 25-26;
Meyer Decl. ¶ 60.) However, because both parties benefit from amortization – through cost smoothing,
increased predictability in multi-year contracts, and increased contractual uniformity and comparability –
the Court considers the issue of amortization to be a wash. (See Mateer Decl. ¶ 9; PX1859 at 3-4.)
94
Prejudgment interest accrues under CERCLA “from the later of (i) the date payment of a
specified amount is demanded in writing, or (ii) the date of the expenditure concerned.” 42 U.S.C. §
9607(a)(4). Lockheed represented in its memorandum on prejudgment interest that it sent a CERCLA
demand letter to the government in December 2000. (Lockheed Memo. on Prejudgment Interest at 2 &
n.1; cf. PX2063 (draft of demand letter).)
95
The Court calculated the potentially available prejudgment interest (assuming a 100%
allocation to the government) for a given year using that year’s response costs for the Sites (see Meyer
Decl. ¶ 50 fig. 29) and compounding interest from that year through 2013 using the variable Superfund
interest rates. (See USX427.) (Costs prior to 2000 were aggregated and treated as all having been
incurred during 2000.) The Court then summed the potentially available prejudgment interest for each
year’s response costs to reach an estimated total potentially available prejudgment interest of over $61

103
equitable allocations for the Sites (supra Section II.E), the government would owe Lockheed

over $18 million in prejudgment interest.96 Critically, there is no evidence – and Lockheed has

expressly declined to provide any assurance – that this $18 million in prejudgment interest would

be allocated to the DiscOps Pool and thus partially (87%) credited back to U.S.-government

contracts. (Trial Tr. at 1455 (Lockheed counsel); id. at 1878 (Lockheed closing).) Instead, the

$18 million in prejudgment interest would amount to a bonanza for Lockheed.

Lockheed would additionally benefit from a recovery of past response costs because

between 40 and 50% of Lockheed’s existing government contracts are fixed-price. (Trial Tr. at

1661 (Gatchel).) These pre-judgment fixed-price contracts are of varying terms and were

negotiated without the expectation of a large lump sum CERCLA recovery for past costs. (Id. at

1661, 1680.) Although Lockheed’s business units will flow credits from any recovery down to

existing fixed-cost contracts, these credits will reduce Lockheed’s indirect costs on the contracts

but will not reduce the price Lockheed realizes from the U.S. government-as-client. (Id. at

1679.) Thus, the credit does not functionally accrue to the government, but instead amounts to

additional profit for Lockheed. (Id. at 601-04 (Wright); id. at 1679-80 (Gatchel).)

Of course, fixed-price contracts entered into after an allocation of past costs to the

government in this case presumably would account for upcoming amortized credits from the

DiscOps Pool, and many pre-judgment fixed price contracts would terminate during the five-year

million. This methodology is consistent with Dr. Meyer’s methodology that calculated a total potentially
available prejudgment interest of just over $66 million, except that she used a demand date of 1994,
which is inconsistent with CERCLA § 107(a)(4)(D). (See Trial Tr. at 1518 (Meyer); Meyer Decl. ¶ 144.)
96
The Court conservatively calculated the government’s exposure by calculating prejudgment
interest available by facility (using facility-specific response costs up through 2011) according to the
procedure described above (supra n. 95) and multiplying each facility’s potentially available prejudgment
interest by the government’s equitable allocation for that facility’s response costs. (See supra Section
II.E.) The Court then summed the results for each facility to estimate the government’s total prejudgment
interest exposure under the Court’s traditional equitable allocation.

104
amortization period following a CERCLA recovery of past costs. However, Mr. Gatchel

estimated that over ten percent of pre-judgment fixed-price contracts would remain in effect over

the entire five-year amortization period. (Id. at 1680-81 (Gatchel).) The record before the Court

is insufficient to establish with any precision how much Lockheed would benefit from a

CERCLA recovery of past costs due to this incompatibility between the DiscOps Pool crediting

scheme, unanticipated credits, and fixed-price contracts. Nonetheless, the available evidence

makes clear that the benefits for Lockheed would be substantial and at the expense of the

taxpayer.

Third, the Court is swayed by the fact that the taxpayers have already underwritten a

substantial portion of Lockheed’s suit by indirectly paying for over 85% of Lockheed’s more

than $10 million in expert and legal fees and other costs. (See Meyer Decl. ¶ 172.) This result

flies in the face of CERCLA’s prohibition against “the award of private litigants’ attorney’s fees

associated with bringing a cost recovery action.” See Key Tronic Corp. v. United States, 511

U.S. 809, 819 (1994). While FAR § 31.205-47 may allow Lockheed to recover its legal fees and

costs through government contracts, that outcome is contrary to both CERCLA and the interests

of the taxpayer. Thus, although it is beyond both the Court’s jurisdiction and the scope of this

case to disallow Lockheed’s legal fees and costs associated with bringing this action, the Court

considers it equitably important that the taxpayers are on the hook for over 85% of Lockheed’s

costs incurred in this action which, as concluded above, would result in further substantial costs

to the taxpayers and accrue to the benefit of Lockheed.

Accordingly, considering the totality of the circumstances, the Court concludes that it

would be inequitable to allocate any liability for past response costs for the Sites to the

government under CERCLA § 113(f)(1). Lockheed indirectly recovered through U.S.-

105
government contracts the lion’s share of its past response costs at the Sites, plus a profit. And

Lockheed has indirectly recovered through the same U.S.-government contracts almost all of its

extraordinarily high attorneys’ fees and costs that it has incurred to sue the government. From

this baseline, it would be inequitable for Lockheed to then receive the additional economic

benefits – at the taxpayer’s expense – of substantial prejudgment interest and windfall profits

from fixed-price contracts that would accompany any CERCLA recovery of past response costs.

Accordingly, the Court equitably reduces the government’s share for past response costs at each

of the Sites to 0%.

B. Lockheed’s recovery of future response costs would not unfairly burden the
taxpayer.

The Court does not come to the same conclusion with regard to future response costs.

CERCLA allocation and the DOSA pose no bigger threat of “double recovery” for future costs

than for past costs.97 And it bears emphasis, once more, that the DOSA clearly anticipated (if not

intended) the coexistence of Lockheed’s indirect recoveries from the government through

government contracts and direct recoveries from the government under CERCLA.

Further, most of the equitable considerations that motivated the Court to eliminate any

further recovery from the government for past response costs do not apply to future response

costs. For, pursuant to a declaratory judgment in this case, the government should reimburse

Lockheed for its future response as those costs are incurred. Indeed, in nominal terms Lockheed

is worse off following a direct CERCLA recovery from the government because it loses profits

97
Indeed, Lockheed would probably credit any direct payments from the government pursuant to
a CERLCA allocation to the DOSA before the associated indirect response costs were flowed down to
government contracts. Because the CERCLA allocation payment for a given year’s response costs would
predate Lockheed’s indirect recovery for those costs through government contracts, the DOSA – and not
the CERCLA allocation for future costs – is the source of the government’s rub.

106
that it would otherwise earn if those indirect costs were allocated to contracts (U.S.-government

and otherwise) through the DOSA.98

Nonetheless, the Court must make a small equitable adjustment to Lockheed’s recovery

of future response costs at the Sites to account for the – albeit temporary – issue of pre-judgment

fixed-price contracts. As described above, fixed-price contracts currently make up over forty

percent of Lockheed’s contract base, and Lockheed – rather than the government – will benefit

from all credits passed on down to pre-judgment fixed-price contracts. (Trial Tr. at 601-04

(Wright); id. at 1679-80 (Gatchel).) Further, over ten percent of these contracts will be in

existence in five years. (Id. at 1680-81 (Gatchel).) However, Lockheed will continue incurring

response costs for the Sites far into the future and all post-judgment fixed-price contracts will

price in the predictable government CERCLA allocation (and credits) pursuant to this action.

With these counterbalancing factors in mind, the Court considers it equitable to decrease

modestly the government’s equitable allocation for future costs at each facility by 1%.

CONCLUSION

For these foregoing reasons, the Court will allocate a 0% share of liability to the United

States for past response costs at the Sites. However, Lockheed is entitled to and will be granted a

declaratory judgment that:

1. Twenty-nine percent of its future necessary response costs at or for the Redlands

facility that are consistent with the National Contingency Plan will be allocated to the United

States and shall be paid by the United States;

98
Of course, when taken to net present value, an up-front and direct CERCLA payment by the
government is likely more valuable to Lockheed than the costs recovered and profits earned through
government and non-government contracts over five years.

107
 2. Twenty-four percent of its future necessary response costs at or for the Potrero

Canyon facility that are consistent with the National Contingency Plan will be allocated to the

United States and shall be paid by the United States; and

3. Nineteen percent of its future necessary response costs at or for the LaBorde Canyon

facility that are consistent with the National Contingency Plan will be allocated to the United

States and shall be paid by the United States.

An Order consistent with this Memorandum Opinion will also be entered on this date.

/s/
ELLEN SEGAL HUVELLE
United States District Judge

Date: April 22, 2014 

108