SLOVITER, Circuit Judge.

OPINION OF THE COURT

Appellant Dean Dairy Products, Inc., d/b/a Fairmont Products, Inc., appeals the district court's imposition of a $4,031,000 civil penalty against it for Clean Water Act violations. Dean Dairy contends that the district court erred when it assessed the economic benefit Dean Dairy gained during the period of the Clean Water Act violations on the basis of Dean Dairy's "wrongful profits." Dean Dairy also contends that the district court improperly looked to the financial condition of Dean Dairy's parent company in evaluating whether it could afford the substantial penalty imposed. The American Frozen Food Institute, the Chemical Manufacturers Association, the American Automobile Manufacturers Association, the International Dairy Foods Association, and the Grocery Manufacturers of America support Dean Dairy's appeal as amicus curiae. We have jurisdiction under 28 U.S.C. § 1291.

I.

The underlying facts of this case are undisputed and are comprehensively set forth in the district court's published opinion, United States v. Municipal Auth. of Union Township, 929 F. Supp. 800 (M.D.Pa. 1996). Briefly stated, Dean Dairy, operating in Union Township, Belleville, Pennsylvania, is a wholly-owned subsidiary of Dean Foods, Inc., the country's largest milk processor. Since 1974 Dean Dairy's wastewater, a result of the production of sour cream, cottage cheese, yogurt and ice cream, has been discharged and treated by Union Township's Publicly Owned Treatment Works (POTW). Union Township collected a user fee based upon the volume of wastewater and the amount of conventional non-toxic pollutants treated at the plant, including Total Suspended Solids (or TSS) and Biological Oxygen Demand (or BOD). In June 1989, pursuant to requirements of the United States Environmental Protection Agency, Union Township issued to Dean Dairy an Industrial User Wastewater Discharge Permit ("the IU permit") which established monthly average limits and daily maximum limits for TSS and BOD and for flow volume.

Beginning in July 1989, Dean Dairy exceeded the limits set forth in its IU permit. Its wastewater, containing the impermissibly high levels of BOD and TSS, flowed from Union Township's POTW into the nearby Kishacoquillas Creek, which was damaged as a result. There is no dispute that because Dean Dairy issued monitoring reports to Union Township on a monthly basis, it had been aware of its violations since July 1989.

In 1994, the United States filed a civil enforcement action against Dean Dairy under the Clean Water Act, 33 U.S.C. § 1251 et seq., for close to 1800 violations of the IU permit and for numerous interferences with the POTW. Following discovery, the United States moved for and was granted summary judgment on the issue of Dean Dairy's liability for the CWA violations. The action against the Municipal Authority of Union Township was settled and therefore the Authority is not a party to this appeal. Dean Dairy does not contest its liability for the violations. Its appeal is limited to the amount of the civil penalty imposed.

The district court held a three-day bench trial to determine the appropriate penalty under the Clean Water Act. Under 33 U.S.C. § 1319(d), a violator of a permit issued pursuant to the Act shall be subject to a civil penalty not to exceed $25,000 per day for each violation. This section further provides that in establishing the penalty the court shall consider the following six factors: "[T]he seriousness of the violation or violations, the economic benefit (if any) resulting from the violation, any history of such violations, any good-faith efforts to comply with the applicable requirements, the economic impact of the penalty on the violator, and such other matters as justice may require." Id.

The district court found Dean Dairy liable for 1,754 violations of its IU permit and 79 instances of interference with Union Township's POTW between July 1989 and April 1994. It also found that Dean Dairy continued to violate its IU permit even after the United States filed suit. Although Dean Dairy took certain steps to address the violations of its permit between 1991 and 1994, the district court found these efforts were belated and ineffective. It was only the construction of a $865,000 pretreatment system, which became operational in April 1995, that succeeded in reducing Dean Dairy's pollutants to permissible levels.

Important to the issue before us is that Dean Dairy considered various options to meet its permit obligations but, as the district court found, "it continued to produce at a volume which it recognized was very likely to generate levels of BOD and TSS beyond that allowed by its IU permit. [Dean Dairy] chose not to reduce production volume because it viewed the concomitant reduction in earnings as too high a price to pay for compliance with the Clean Water Act." 929 F. Supp. at 805.

Although the district court applied the six statutory factors a court must consider in assessing the appropriate penalty for a CWA violation, the appellant presents the case as if the court concentrated almost exclusively on the "economic benefit" factor. In fact, the district court made extensive findings of fact and issued conclusions of law on each of the six factors. See id. at 802-09. The court noted that the history of Dean Diary's violations dated back to 1989, that the excessive discharges required the Pennsylvania Fish and Boat Commission to cease stocking fish in areas of the Kishacoquillas Creek, and that its two-year delay to take meaningful action to remedy the violations did "not speak highly of its good faith in this matter." Id. at 803-08.

In connection with its evaluation of the economic benefit factor, which is the primary basis of the appeal, the district court acknowledged that the parties had previously stipulated that Dean Dairy did not realize any economic benefit from delaying the capital investments necessary to achieve compliance with its IU permit. This was due to the unusual fact that, by delaying the construction of the pretreatment plant, Dean Dairy was actually losing money because it was paying higher usage fees to the POTW for its increased volume. Thus, Dean Dairy did not reap an economic benefit by delaying the construction of the pretreatment plant. The court nevertheless found that Dean Dairy did realize an economic benefit during the period of the violations because it produced "at a volume above that which would have allowed it to operate within its IU permit." Id. at 805.

In making the finding of economic benefit, the district court relied upon a document produced by Dean Dairy during discovery and introduced at trial as Joint Exhibit 18 that outlined various options by which Dean Dairy could comply with its permit. The district court noted that Option #4 of that document indicated that Dean Dairy could drop PennMaid as a customer and thereby reduce the amount of wastewater generated. Dean Dairy recognized, however, in Exhibit 18 that losing the revenues from PennMaid would result in a loss of earnings in the amount of $417,000 in fiscal year 1994.

During the damages trial, the government questioned Fairmont plant manager Dean Koontz about that document in the following exchange:

Counsel for the United States: Let me turn your attention to Joint Exhibit 18, please. It is entitled Recap of Wastewater Treatment Options . . . .

* * *

Mr. Koontz: Okay. Yes. This is the information that I fed to Ron Crock based on the costs that Union Township gave us. And he ran some type of analysis with it.

Q: Ron Crock is the comptroller for Dean Dairy Products, is that correct?

A: Yes, he is.

* * *

Q: And Mr. Crock ran numbers using four options, is that right?

A: Yes it is.

Q: And one of those options includes Fairmont not building a treatment plant and discharging to the Authority but cutting back on plant volume to reduce flow rate which does not allow for growth — future growth of the plant. Do you see that?

A: Yes, option four.

Q: Option number four. . . . [I]t has an estimate of the impact on earnings that this option would involve. Is that correct?

A: That is the way I read it, yes.

Q: These are numbers that Mr. Crock arrived at. Is that correct, sir?

A: Yes.

Q: Isn't it true that if you reduce volume, you would lose an account by the name of PennMaid according to the scenario?

A: Yes. This was when we did look at possibilities of reducing volume. The only customer that we could come up with that would have an impact would have been PennMaid because we make only one product for them, cottage cheese. And we could eliminate them as a possible customer, which would have made all of us very happy.

Q: It would have made you all happy, and it would also have set you back $417,000 for fiscal year `94. Is that correct?

A: Yes. There would have been a considerable amount of overhead absorption loss by losing that.

App. at 275-77.

In its opinion, the district court commented on Exhibit 18 as follows: "Production volume at Fairmont was higher in each year from 1989 to 1993 than it was in 1994, and, therefore, it is reasonable to believe that Fairmont gained at least $417,000 in earnings annually during the period of its violations. On this basis, the court concludes that between July 1989 and April 1994, Fairmont gained approximately $2,015,500 by violating its IU permit." 929 F. Supp. at 805. The district court also determined that the figure should be doubled in order to provide a proper deterrent and punishment, and accordingly imposed a total penalty of $4,031,000.

II.

Dean Dairy challenges the district court's analysis of two of the six factors to be considered in imposing a CWA civil penalty—the economic benefit to the violator and the economic impact of a penalty. We consider first its contention that the district erred as a matter of law in using a "wrongful profits" approach to ascertain whether Dean Dairy received any "economic benefit" from its Clean Water Act violations.

A. The Use of "Wrongful Profits" to Measure Economic Benefit

Section 1319(d) of the Clean Water Act provides in pertinent part:

Any person who violates . . . this title, or any permit condition or limitation . . . shall be subject to a civil penalty not to exceed $25,000 per day for each violation. In determining the amount of a civil penalty the court shall consider the seriousness of the violation or violations, the economic benefit (if any) resulting from the violation, any history of such violations, any good-faith efforts to comply with the applicable requirements, the economic impact of the penalty on the violator, and such other matters as justice may require.

33 U.S.C. § 1319(d).

The statute does not define the term "economic benefit" used in this section. It is apparent, however, that the goal of the economic benefit analysis is to prevent a violator from profiting from its wrongdoing. In United States v. Smithfield Foods, Inc., 972 F. Supp. 338, 348 (E.D.Va. 1997), the district court explained that "[c]courts use economic benefit analysis to level the economic playing field and prevent violators from gaining an unfair competitive advantage."

A similar rationale was also given by the Environmental Protection Agency, which emphasized that the reason for considering economic benefit to a violator in assessing a CWA penalty is to remove or neutralize the economic incentive to violate environmental regulations. In a 1990 Manual to its BEN computer program, established to assist in the calculation of civil CWA penalties, the EPA explained:

An organization's decision to comply with environmental regulations usually implies a commitment of financial resources; both initially, in the form of a capital investment or one-time expenditure, and over time, in the form of annual, continuing expenses. These expenditures might result in better protection of public health or environmental quality; however, they are unlikely to yield any direct economic benefit (i.e., net gain) to the organization. If these financial resources were not used for compliance, they presumably would be invested in projects with an expected direct economic benefit to the organization. This concept of alternative investment; that is, the amount the violator would normally expect to make by not investing in pollution control, is the basis for calculating the economic benefit of noncompliance. As part of the Civil Penalty Policy, EPA uses the Agency's penalty authority to remove or neutralize the economic incentive to violate environmental regulations. In the absence of enforcement and appropriate penalties, it is usually in the organization's best economic interest to delay the commitment of funds for compliance with environmental regulations and to avoid certain other associated costs, such as operating and maintenance expenses.

EPA BEN User's Manual I-6 (July 1990), quoted in Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 890 F. Supp. 470 (D.S.C. 1995) ("Laidlaw I").

Few published cases discuss the "economic benefit" factor of the Clean Water Act in any detail, and those that do are, in large part, district court opinions. In Laidlaw, the court described economic benefit as: "the after-tax present value of avoided or delayed expenditures on necessary pollution control measures." 890 F. Supp. at 481. The theory is that economic benefit "represents the opportunity a polluter had to earn a return on funds that should have been spent to purchase, operate, and maintain appropriate pollution control devices." Id.

This court has previously recognized that a violator's economic benefit under the Clean Water Act may not be capable of ready determination. In Public Interest Research Group of New Jersey, Inc. v. Powell Duffryn Terminals, Inc., 913 F.2d 64 (3d Cir. 1990), we stated:

Precise economic benefit to a polluter may be difficult to prove. The Senate Report accompanying the 1987 amendment that added the economic benefit factor to section 309(d) recognized that a reasonable approximation of economic benefit is sufficient to meet plaintiff 's burden for this factor. . . . The determination of economic benefit or other factors will not require an elaborate or burdensome evidentiary showing. Reasonable approximations of economic benefit will suffice.

Id. at 80 (citation omitted).

Because of the difficulty of determining the appropriate penalty under the Clean Water Act, the court will accord the district court's award of a penalty wide discretion, even though it represents an approximation. This was emphasized by the Supreme Court when it said, more than a decade ago, "Congress [made the] assignment of the determination of the amount of civil penalties to trial judges . . . . Since Congress itself may fix the civil penalties, it may delegate that determination to trial judges. In this case, highly discretionary calculations that take into account multiple factors are necessary in order to set civil penalties under the Clean Water Act." Tull v. United States, 481 U.S. 412, 426-27 (1987).

This has been the approach generally followed by the courts. In Smithfield Foods, the court recognized that it is difficult to prove the precise economic benefit to a polluter, but stated that the economic benefit analysis provides an approximation of "the amount of money a company has gained over its competitors by failing to comply with the law." 972 F. Supp. at 348. See Sierra Club v. Cedar Point Oil Co., 73 F.3d 546, 576 (5th Cir. 1996) ("a court need only make a `reasonable approximation' of economic benefit when calculating a penalty under the CWA") (citation omitted); Public Interest Research Group of New Jersey, Inc. v. Hercules, Inc., 970 F. Supp. 363, 364 (D.N.J. 1997) ("In assessing a penalty under the Clean Water Act, a district court has a great amount of discretion."); United States v. Sheyenne Tooling Mfg. Co., 952 F. Supp. 1420, 1422-23 (D.N.D. 1996) ("It must be understood, however, that despite the directional aid and guidance that the six enumerated factors in § 1319(d) provide, the calculation of a final penalty may often be imprecise and approximate at best. Indeed, the accuracy of the final calculations, and the figure of penalty that they produce, is as dependant, or even more so, upon the provision of complete and accurate evidence, as introduced, developed, and explained at trial, as it is upon a good evaluation of this information by the court.").

Courts have applied different methods in determining the appropriate penalty for a Clean Water Act violation. Some courts have employed a "top down" approach in which the maximum possible penalty is first established, then reduced following an examination of the six "mitigating" factors. See, e.g., Sierra Club, 73 F.3d at 574-76; Powell Duffryn, 913 F.2d at 79; Atlantic States Legal Found., Inc. v. Tyson Foods, Inc., 897 F.2d 1128, 1142 (11th Cir. 1990); Hawaii's Thousand Friends v. City and County of Honolulu, 821 F. Supp. 1368, 1395 (D. Haw. 1993); Atlantic States Legal Found., Inc. v. Universal Tool Stamping Co., 786 F. Supp. 743, 746-47 (N.D. Ind. 1992).

Other courts have used a "bottom up" approach whereby the economic benefit a violator gained by noncompliance is established and adjusted upward or downward using the remaining five factors in § 1319(d). See , e.g., Smithfield Foods, 972 F. Supp. at 354; Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 956 F. Supp. 588, 603 (D.S.C. 1997) ("Laidlaw II"). Because the statute does not prescribe either method, it appears that a court is free to use its discretion in choosing the appropriate method. See Smithfield Foods, 972 F. Supp. at 354.

Had the district court in this case taken a "top-down" approach, it would have begun at the maximum penalty, which was approximately $45,825,000, based on the statutory penalty of $25,000 a day. Instead, the court applied the "bottom up" approach, 929 F. Supp. at 806, by determining Dean Dairy's economic benefit acquired through the Fairmont plant's production at a volume that resulted in more wastewater than permissible under its permit. These were knowing violations, as its own document demonstrated it was aware that if it had reduced its wastewater volume by reducing its production, it would have been in compliance with its IU permit. As Koontz specifically testified, if Dean Dairy had reduced volume, it believed it would have lost PennMaid as a customer. Dean Dairy's own document prepared by Ron Crock, its controller, demonstrated that this loss of PennMaid would have had a negative impact of $417,000 per year. This was the basis on which the district court calculated Dean Dairy's economic benefit as $2,015,500, which when doubled, resulted in the penalty of $4,031,000. That penalty was barely 9% of the maximum statutory penalty to which Dean Dairy was subject.

It is not surprising that no published case has used this method of ascertaining a violator's economic benefit because it is the rare violator who actually loses money by delaying compliance with the law. Typically, a violator benefits economically by avoiding or delaying the construction of anti-pollution equipment that would have placed it in compliance with its permit. See, e.g., Sierra Club, 73 F.3d at 574; Hawaii's Thousand Friends, 821 F. Supp. at 1396. In Smithfield Foods, the court explained that, "[w]hen a company delays or avoids certain costs of capital and operations and maintenance necessary for compliance, the company is able to use those funds for other income-producing activities, such as investing that money in their own company." Id. at 349. Therefore, it concluded that in that case, "the avoided and/or delayed cost of compliance, and the weighted average cost of capital (WACC) as a discount/interest rate in the economic benefit calculation, to be both the best and the appropriate method to determine how much money defendants made on the funds they did not spend for compliance." Id. at 349 (footnote omitted).

This case is unusual because Dean Dairy's delay in constructing a pretreatment plant was not beneficial to its "bottom line"; in effect, Dean Dairy was actually penalizing itself in failing to promptly build the pretreatment plant. Our general assumption of the reasonable capitalist went awry with this company.

There are methods other than the delayed or avoided capital expenditure for ascertaining economic benefit, a fact the appellant and the amici decline to acknowledge. It is significant that neither the statute nor the case law supports the contention that the cost-avoidance method is the only permissible method of determining the amount a polluter has gained from violating the law. In Smithfield Foods, though the court applied the delayed or avoided costs method, it "acknowledge[d] there are various methods for calculating defendants' economic benefit gained from noncompliance." 972 F. Supp. at 349.

In contrast to the situation in Smithfield Foods, the "cost-avoided" method of determining economic benefit is not a method that fits the facts that were presented to the district court because Dean Dairy did not profit by delaying its construction of the pretreatment plant. But it clearly gained other economic benefits by failing to adopt the method that was readily available. The wastewater from the Fairmont plant is created by the required daily cleaning of its vats and other processing equipment. A reduction of production would reduce the wastewater. Thus, if Dean Dairy wanted to avoid the cost entailed by the purchase of new equipment, it had the option of reducing volume.

The approach adopted by the district court is not in conflict with the CWA or basic economic principles. A violator who chooses to continue to violate its permit while experimenting with less costly remedies necessarily subjects itself to the surrender via penalty of any economic benefit it acquired. The fact that the violator has also penalized itself by failing to implement cost-effective methods that would have put it into compliance with its permit and thereby save it money is certainly no basis to mitigate its penalty.

Requiring a company to reduce the amount of pollution it creates to comply with its permit is not unreasonable. As the court in Tyson Foods stated: "There was one simple and straightforward way for Tyson to avoid paying civil penalties for violations of the Clean Water Act: After purchasing the plant, Tyson could have ceased operations until it was able to discharge pollutants without violating the requirements of its . . . permit. Tyson chose not to do this and it must now bear the consequences of that decision." 897 F.2d at 1141-42. Similarly, Dean Dairy chose neither what proved to be the economically sensible option (building the pretreatment facility) nor the alternative option of reducing the amount of wastewater produced. Accordingly, it must bear the consequences.

Appellant and the amici argue that the use of the "wrongful profits" method to calculate economic benefit contravenes EPA policy. They base this argument on the EPA's interim Clean Water Act Settlement Penalty Policy Guide published in 1995, which states that the "objective of the economic benefit calculation is to place violators in the same financial position as they would have if they complied on time." App. at 244. The appellant focuses on language in the Guide that explains that "[p]ersons that violate the Clean Water Act are likely to have obtained an economic benefit as a result of delayed or completely avoided pollution control expenditures during the period of noncompliance." Id. Thus, this reference to delay and avoidance and the EPA's creation of computer software ("the BEN model") to calculate a violator's economic benefit from delaying or avoiding compliance has led appellant to conclude that this is the only approach to determining economic benefit.

However, the EPA guideline itself notes that this approach may not always be applicable. The policy states:

[I]f the violator is a privately-owned regulated utility, the standard BEN model may not be appropriate. In this situation, the Agency should consider a wrongful profits analysis and seek to recover the profits and other competitive market benefits the violator obtained as a result of operating during the period of operation. . . . In a few unusual cases, economic benefit may be negative: this means, e.g., operating the old efficient system was more expensive than purchasing and operating a new, more efficient treatment system. When economic benefit is negative, the settlement calculation enters zero as the economic benefit.

App. at 245-46 (emphasis added).

Even the EPA guideline, on which Dean Dairy relies, is receptive to a wrongful profits analysis where appropriate. But the EPA policy is not applicable here because, by its own express terms, it is not "intended for use by EPA, violators, courts or administrative judges in determining penalties at a hearing or trial." App. at 243. See also Laidlaw I, 956 F. Supp. at 601 ("the policy expressly states that it is not to be used by a court in determining a penalty at a trial . . . . [therefore] it will not be considered in arriving at an appropriate penalty amount"). Finally, the conclusive evidence that the "wrongful profits" approach to economic benefit does not conflict with EPA's policy is the fact that its representative is on the brief for the government arguing in support of that approach here.

We conclude that the district court's method of calculation of the penalty was within its discretion. We do not suggest that we have any dissatisfaction with the cost-avoided method of determining a violator's economic benefit in the usual case. However, under these unusual circumstances, we see no legally significant difference in measuring the economic benefit achieved by avoiding the costs of anti-pollution equipment, and the economic benefits achieved by failing to reduce the volume of pollution created. Both methods aim to recoup any benefits a violator gained by breaking the law and which gave the violator an advantage vis-a-vis its competitors. The penalty thus achieves the leveling of the playing field intended by Congress.

Finally, we reject Dean Dairy's claim that it was "ambushed" and unfairly surprised by the government's reliance on the "wrongful profits" theory of economic benefit. Although the government first identified Joint Exhibit 18 as the document on which it based the $417,000 figure in its penalty calculation in its closing argument, the document had already been admitted into evidence and was the subject of the government's trial examination of Koontz, Dean Dairy's plant manager, quoted earlier in this opinion. Dean Dairy did not question Koontz on this document nor did it call Ron Crock, its own controller, who prepared the figures in the document, for an explanation. If Dean Dairy was content to allow those figures to go to the factfinder without cross-examining Koontz and examining Crock, then it cannot reasonably argue only after the district court's opinion, forthcoming six months later, that the figures don't mean what they say.

Dean Dairy vehemently contends that it was given inadequate notice of the government's theory. Our examination of the record establishes that the government gave Dean Dairy ample notice that it was pursuing this theory of economic benefit. The government's pretrial memorandum states that "Dean Dairy was operating its Fairmont plant at a percentage over capacity, thus increasing its profits at the expense of violating" the CWA, and that this "economic benefit, separate from capital costs, was substantial for each year of production." Supp. App. at 748. The government's trial brief also emphasized that Dean Dairy "enjoyed financial benefits during the period of noncompliance" because its "production operations were neither halted nor adversely impacted by its noncompliance." App. at 768. The government's opening statement promised to "present evidence showing what Dean itself estimated as the amount of revenue it would lose if it decreased production to levels that would comply with its permit." Supp. App. at 734. Dean Dairy never requested a continuance of the bench trial to answer those contentions.

Moreover, although Dean Dairy repeatedly refers to the government's stipulation that Dean Dairy did not realize any economic benefit from delaying the capital expenditures needed to achieve compliance with the IU permit, the government carefully reserved the argument that Dean Dairy could have received economic benefit from other actions. Indeed, at trial government counsel specifically stated:

[T]he United States is preserving the argument with respect to economic benefit to other actions that the dairy could have taken. In other words, we are talking about production or other matters aside from the capital expenditures, as well as the maintenance costs associated with the actual hardware that was brought on-line at the dairy facility.

App. at 288.

Finally, although six months elapsed after the close of the trial during which the parties prepared and submitted post trial briefs, which we have examined, Dean Dairy never argued that it had been unprepared by the government's theory or method of proof. Only after the district court's opinion fixing the penalty did it raise these claims. By then, it was too late. Under these circumstances, we cannot accept Dean Dairy's claim that it was unfairly surprised by either its use of the "wrongful profits" approach of economic benefit or the basis of the calculation of the penalty.

B. Consideration of the Finances of Dean Dairy's Parent Company

In analyzing the "economic impact on the violator," one of the six factors the statute lists as relevant to the CWA penalty, the district court stated it would look to the finances of Dean Dairy's parent, Dean Foods. Among the reasons it gave were that "Dean Foods was closely involved with Fairmont [Dean Dairy's plant] in evaluating the options for resolving Fairmont's wastewater problems, and it had complete control over whether and when Fairmont would achieve compliance with its IU permit." 929 F. Supp. at 808. The district court also stated that "[a]t the same time, Dean Foods was siphoning off profits from Fairmont." Id.

Dean Dairy contends that it was legal error for the district court to consider the financial condition of Dean Foods because the parent corporation was not a party to the action, it did not violate the Clean Water Act, and there was insufficient evidence to pierce the corporate veil. We reject its contention. Notwithstanding the arguments made in the brief and before us orally, it is important to remember that it was not Dean Foods, but only Dean Dairy, the violator, who was penalized. The reference to Dean Foods' financial statement merely assured that the penalty would not be set at a level above Dean Dairy's ability to pay. If the subsidiary does not retain its revenues, as the evidence showed in this case, then its parent's financial resources are highly relevant. Other courts in CWA cases have looked to the assets and finances of the violator's parent in evaluating the economic impact of the penalty on a violator, see Universal Tool, 786 F. Supp. at 753; PIRG v. Powell Duffryn, 720 F. Supp. 1158, 1166 (D.N.J. 1989), reversed in part on other grounds, 913 F.2d 64 (3d Cir. 1990), and we see nothing improper in the same action here.

The consideration of a parent's financial condition in assessing a penalty on a subsidiary is a far cry from piercing the corporate veil and holding the parent liable for the actions of its subsidiary; here the penalty was assessed against the subsidiary alone. Furthermore, the district court only considered Dean Foods' assets as one factor among others; they were not dispositive. The court simply undertook a fact-specific assessment that examined the role of the parent in the operations of the subsidiary, particularly with regard to the issue of pollution of the nearby waters and actions that could have resolved it.

Although Dean Dairy contends that any evaluation of its financial condition would show that it lost over one million dollars between 1992 and 1995, there was also evidence that it had over twenty million dollars in assets infiscal year 1995. App. at 586-88. In light of the discretion afforded a district court in determining a penalty, we conclude that the court's examination of the parent's assets, the basis for which it was done, and the manner in which the information was used was neither clear error nor an abuse of discretion.

III.

For the foregoing reasons, the order of the district court will be affirmed.

A True Copy:

Teste:

Clerk of the United States Court of Appeals for the Third Circuit