FILED

NOT FOR PUBLICATION

JUN 3 2020

UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK

U.S. COURT OF APPEALS

FOR THE NINTH CIRCUIT

MISSION LINEN SUPPLY, a California No. 19-15392 corporation, D.C. No. Plaintiff-Appellee, 1:15-cv-00672-AWI-EPG

v.

MEMORANDUM*

CITY OF VISALIA,

Defendant-Appellant.

Appeal from the United States District Court for the Eastern District of California Anthony W. Ishii, District Judge, Presiding

Submitted March 27, 2020** San Francisco, California

Before: WALLACE, GRABER, and COLLINS, Circuit Judges.

The City of Visalia timely appeals from the district court’s equal allocation

of future recovery costs between the City and Mission Linen Supply ("Mission"),

in this action under the Comprehensive Environmental Response, Compensation,

* This disposition is not appropriate for publication and is not precedent except as provided by Ninth Circuit Rule 36-3. ** The panel unanimously concludes this case is suitable for decision without oral argument. See Fed. R. App. P. 34(a)(2). and Liability Act of 1980, 42 U.S.C. §§ 9601–75. Reviewing for abuse of

discretion the district court’s selection of factors and for clear error in its allocation

of costs according to those factors, Boeing Co. v. Cascade Corp., 207 F.3d 1177,

1187 (9th Cir. 2000), we affirm.

The district court did not abuse its "broad discretion." TDY Holdings, LLC

v. United States, 885 F.3d 1142, 1149 (9th Cir. 2018). The court permissibly

focused on the factor of geographic distribution and attributed most responsibility

for on-site pollution to Mission and most responsibility for off-site pollution to the

City. See Boeing, 207 F.3d at 1187 (holding that district courts have discretion "to

decide what factors ought to be considered"). On appeal, the City does not

challenge the court’s many factual findings concerning the City’s sewers. They

were "installed below general industry standards." The slope of some sewers was

too flat, allowing wastewater to seep into the ground. Some sewers were too

shallow. Other problems included "holes/broken pipes, exposed soil, cracks, sags,

offset/separated joints, missing portions of pipe, root intrusion, debris, and deposits

of material that indicate blockages and surcharge conditions." The City did not

restrict or limit the dumping of PCE into the sewers. But for the defects in the

sewers, the wastewater would have reached the City’s treatment facilities.

2 The cases cited by the City do not support its view that the court here abused

its discretion. As an initial matter, even if one of the cases were factually similar,

the existence of discretion means that one district court could reach a conclusion

different from the conclusion of another district court. In any event, none of the

cited cases involved factually similar circumstances. See Boeing, 207 F.3d at

1180–82 (affirming allocation of 30% of costs to one landowner and 70% of costs

to another landowner because of differing levels of pollution); Waste Mgmt. of

Alameda Cty., Inc. v. E. Bay Reg’l Park Dist., 135 F. Supp. 2d 1071, 1089–1104

(N.D. Cal. 2001) (allocating 5% of the costs to the park district due to many

factors, including the court’s finding that the district had done little to cause the

contamination); United States v. Davis, 31 F. Supp. 2d 45, 65–67 (D.R.I. 1998)

(allocating 35% of costs to transporters of chemicals even though the landfill

operator was at greater fault), aff’d, 261 F.3d 1 (1st Cir. 2001).

The dissent asserts that the district court abused its discretion because, in the

dissent’s view, the court’s two alternative methodologies rested on differing

underlying assumptions. The City has never raised this argument, so it is forfeited.

Smith v. Marsh, 194 F.3d 1045, 1052 (9th Cir. 1999). The dissent’s selective

quotations from the City’s "Statement of the Case," on page 6 of the opening brief,

are not sufficient to preserve the issue. Nowhere in that passage or elsewhere did

3 the City assert that the district court abused its discretion by using alternative

methods that rested on contradictory assumptions. Moreover, even if we read the

opening pages of the City’s brief expansively to encompass the argument, the

City’s bald assertion in passing is insufficient to preserve the issue. See, e.g.,

United States v. Graf, 610 F.3d 1148, 1166 (9th Cir. 2010) ("Arguments made in

passing and not supported by citations to the record or to case authority are

generally deemed waived."); Greenwood v. FAA, 28 F.3d 971, 977 (9th Cir. 1994)

("We review only issues which are argued specifically and distinctly in a party’s

opening brief."); id. ("We will not manufacture arguments for an appellant, and a

bare assertion does not preserve a claim.").

In any event, we are unpersuaded that the court’s use of alternative

methodologies was an abuse of discretion. The court’s primary method took

account of nuance: the court looked to 46 different plume circles, each extending

35 feet from a central point of measurement; the court allocated costs to each

party—the City, Mission, and Mission’s predecessor—depending on whether the

party’s activities contributed to the plume; and the court assigned proportionate

responsibility to the City and Mission for the predecessor’s share. The court’s

alternative method was simpler: it looked solely at the 46 points of measurement

and allocated all costs from on-site or on-the-border measurements to Mission and

4 all costs from purely off-site measurements to the City. We commend the district

court for checking the reasonableness of its nuanced primary methodology by

reference to a simpler alternative methodology.

AFFIRMED.

5

FILED

Mission Linen Supply v. City of Visalia, No. 19-15392

JUN 3 2020

COLLINS, Circuit Judge, dissenting: MOLLY C. DWYER, CLERK

U.S. COURT OF APPEALS

This case involves an action under the Comprehensive Environmental

Response, Compensation, and Liability Act (“CERCLA”) to allocate responsibility

for underground pollution that originated at a dry-cleaning facility in the City of

Visalia, California. Between 1971 and 1983, Mission Linen Supply (“Mission”)

and the previous owner of the main property in question (Star Laundry & Dry

Cleaning (“Star”)) operated dry-cleaning facilities that discharged

perchloroethylene (“PCE”) into the City’s sewers. Due to the sewers’ numerous

defects and poor maintenance, PCE leaked out of the sewers and created a

substantial underground “plume” in the vicinity of the property. After a bench

trial, the district court allocated 50% of the responsibility for future cleanup costs

to Mission and 50% to the City. (Star was no longer in existence and was not a

party to the CERCLA action.) The majority rejects the City’s appeal, concluding

that the district court did not abuse its broad discretion. I respectfully dissent.

CERCLA explicitly gives district courts discretion to “allocate response

costs among liable parties using such equitable factors as the court determines are

appropriate.” 42 U.S.C. § 9613(f)(1). I agree that the district court did not abuse

its discretion in identifying the three principal considerations on which it based its

allocation decision: (1) how to divide up the pollution plume by its “geographic

1 features,” i.e., which portions of the plume counted as being on Mission’s property

and which counted as offsite; (2) how to assign responsibility for offsite portions of

the plume; and (3) how to allocate the “orphan” responsibility of Star. See Boeing

Co. v. Cascade Corp., 207 F.3d 1177, 1187 (9th Cir. 2000) (exercise of “discretion

to select factors” is reviewed only for abuse of discretion).

Having identified those considerations, the district court then needed to

make a judgment as to each of them and determine an appropriate allocation in

light of those judgments. We “review for clear error the allocation according to the

selected factors,” TDY Holdings, LLC v. United States, 885 F.3d 1142, 1147 (9th

Cir. 2018), meaning that “[w]e may not reverse a district court’s exercise of its

discretion unless we have a definite and firm conviction that the district court

committed a clear error of judgment in the conclusion it reached upon weighing the

relevant factors,” SEC v. Coldicutt, 258 F.3d 939, 941 (9th Cir. 2001). Here, the

district court committed a clear error of judgment by relying upon two sets of

internally inconsistent findings. See, e.g., Perez-Arceo v. Lynch, 821 F.3d 1178,

1186 (9th Cir. 2016) (“‘Factual findings that are internally inconsistent . . . are

clearly erroneous.’”) (quoting 9C WRIGHT & MILLER, FEDERAL PRACTICE AND

PROCEDURE § 2579 n.9 (3d ed. 2016)); Caruso Enters., Inc. v. U.S.A. Motel Corp.

(In re U.S.A. Motel Corp.), 450 F.2d 499, 503 (9th Cir. 1971) (findings are clearly

erroneous when they are “significantly contradictory”). Specifically, the district

2 court adopted two alternative sets of findings as to how to resolve the three key

questions it had identified, and these two methods rest on an unexplained mixing

and matching of contradictory answers to these three questions.1

In devising two alternative methods of allocation, the district court began

with the proposed allocation method of Mission’s expert, Keith O’Brien, and then

made various modifications to that method. As the district court explained,

O’Brien devised a color-coded map showing several dozen circles, each of which

was “imposed over the soil vapor contamination” in a given area of the property

and its vicinity. The district court concluded that these circles could “be used as a

form of units in measurement[,] . . . because they are indicative of the underlying

contamination plume.” O’Brien assigned those circles he characterized as being

1 The majority asserts that the City “has never raised this argument,” see Memorandum at 3, but that is wrong. In its opening brief, the City explicitly complained that, by invoking two alternative theories, the district court “did not specify the theory with which it was rendering its decision” and that, as a result, its findings are tainted by the “substantial flaw” that “the parties and this Court cannot tell how the District Court actually arrived at its decision.” The fact that these contentions were made in the City’s highly argumentative “Statement of the Case,” see id., is of no moment; we should not ignore an argument simply because it is made in one section of a brief versus another. See, e.g., United States v. Valdivia, 492 F.2d 199, 203–04 (9th Cir. 1973) (considering sufficiency-of-the-evidence argument made in the “statement of facts” in the brief, even though the appellant had “not devoted a specific section of his brief” to that argument). Moreover, after expressly complaining about the district court’s use of alternative methods, the City then goes on in its brief to explain why the district court’s various judgments under each method were unexplained or flawed. That is enough to place the adequacy of those findings before us, and the findings’ self-contradictory nature is impossible to ignore.

3 on the property and away from the sewers to Mission and Star. The circles he

considered to be offsite he assigned in classes either to the City and Mission, to the

City and Star, or to the City, Mission, and Star, depending upon which companies

had been shown to have used which nearby sewer lines. For each such class of

circles, he divided the number of circles in the class by the number of parties

responsible for that class and then assigned each party an equal share of the circles

in that class. Having thus assigned total fractional shares of circles to each

responsible party, he calculated the ratio of Mission’s share to the City’s share and

then reassigned Star’s share to each in that same proportion. His method yielded

approximately 24 circles for the City and 22 for Mission.

The first of the two alternative methods adopted by the district court

followed O’Brien’s method, except that it applied a more lenient concept of what

counted as an onsite circle attributable to Mission and Star, thereby increasing the

number of such circles from three to nine. That resulted in an allocation of 15

circles to Mission, 16 to Star, and 15 to the City. Because Mission and Star each

had 15 circles, the court then reallocated Star’s 16 circles equally to Mission and

Star, giving each 23 circles—a 50/50 split.

In the second alternative method, the district court applied an even broader

concept of what counted as an onsite circle, increasing the number of such circles

from nine to 23. Having done so, however, it then changed its rule for allocating

4 the offsite circles: instead of allocating those circles to the City and the respective

users of the relevant sewer line, the court instead now allocated all of the offsite

circles to the City. It then likewise changed its rule for allocating Star’s share and

instead allocated it completely to Mission. That yielded 23 onsite circles allocated

entirely to Mission and 23 offsite circles allocated entirely to the City—the same

answer as the first method.

Although both alternative methods yielded the same 50/50 split, the two

methods rested on an arbitrary pairing of contradictory answers to the three key

allocation factors. In what I will call method “A,” the court applied these three

assumptions:

(A)(1) Defining onsite circles: The court narrowly defined what counted as an onsite circle to be attributed to Mission and Star (although not as narrowly as O’Brien).

(A)(2) Allocating offsite circles: For the offsite circles, the court allocated responsibility equally to the City and to the relevant operators.

(A)(3) Allocating Star’s share: For the orphan share of Star, the court allocated Star’s share equally to Mission and to the City based on their respective total allocations from (A)(1) & (A)(2).

But in the alternative method (which I will call method “B”), it applied these

three different assumptions:

(B)(1) Defining onsite circles: The court more broadly defined what counted as an onsite circle to be attributed to Mission and Star. (B)(2) Allocating offsite circles: For the offsite circles, the court allocated responsibility 100% to the City.

5 (B)(3) Allocating Star’s share: For the orphan share of Star, the court allocated 100% to Mission.

The district court committed clear error by arbitrarily pairing collections of

contradictory answers to the three key allocation factors. The district court did not

explain why, for example, expanding the concept of an onsite circle (as in method

B) should lead to changing the methods for allocating offsite circles or Star’s

share. Nor did the court explain why it did not instead combine, say, assumption

(B)(1) with assumptions (A)(2) and (A)(3) (which would have yielded roughly a

37/63 split between the City and Mission) or assumption (B)(1) with assumptions

(A)(2) and (B)(3) (which would have yielded roughly a 22/78 split between the

City and Mission). The decision as to which of the opposite predicate assumptions

to pair with which of the other opposite predicate assumptions seems to me on this

record to have been wholly arbitrary.

The majority claims that these various contrary propositions can be

reconciled, but the logic escapes me. According to the majority, the first method

reflects “nuance” that attempted to allocate shares “depending on whether the

party’s activities contributed to the plume,” while the second method is “simpler”

and “allocated all costs from on-site or on-the-border [circles] to Mission and all

costs from purely off-site [circles] to the City.” Memorandum at 4–5. The

majority’s description of one method as “nuanced” and the other as “simpler”

simply begs the question. The majority never addresses, much less resolves, the

6 fundamental contradictions between the district court’s two methods, which are

that (1) they inexplicably used two significantly different understandings of what

constituted an onsite circle (one of which counted 23 circles as onsite, and the

other of which counted only nine as onsite); and (2) they inexplicably use two

different methods for re-allocating Star’s orphan share. One is not a “simpler”

form of the same, more “nuanced” analysis. Each is a different—and

contradictory—analysis.

We afford broad discretion to district courts in CERCLA cases, but we do so

based on the assumption that they will articulate a coherent and internally

consistent resolution of the major factors supporting any given allocation. That did

not happen here, and the resulting allocation therefore rests on clear error. I

respectfully dissent.

7