In 1986, the Puerto Rico Department of Consumer Affairs (DACO) took a small, tentative step toward regulating the profit margins of gasoline wholesalers. The wholesalers treated this move as a declaration of war. They mounted a courtroom counteroffensive and succeeded in obtaining an injunction against the enforcement of DACO’s embryonic regulation. Following a series of pitched battles that stretched from San Juan to Boston to the banks of the Potomac and back again, DACO emerged victorious.
Long after the injunction had been vacated, DACO purposed to exact tribute from the vanquished. Specifically, it sought restitution from the wholesalers based on the “excess” profits that they allegedly earned while *872 shielded by the injunction. The district court declined to grant the envisioned spoils. We affirm.
I. BACKGROUND
This is presumably the final skirmish in a decade-long conflict. Other jousts are chronicled in a series of published opinions.
See, e.g., Puerto Rico Dep’t of Consumer Affairs v. Isla Petroleum Corp.,
From 1973 forward, the federal government imposed price controls on the sale of petroleum and petroleum products.
See
15 U.S.C. §§ 751-760h (as amended). At the time federal controls ended in early 1981, the regulatory scheme limited wholesalers’ gross profit margins (GPMs) on the sale of gasoline to 8.6 cents per gallon.
1
See Tenoco,
By 1985, the GPMs of gasoline wholesalers in Puerto Rico ranged from 6.9e to 16.76e per gallon. In early 1986, world oil prices plummeted — but the price of gasoline in Puerto Rico (both wholesale and retail) failed to follow suit. The Puerto Rico legislature, ostensibly concerned that the oil companies were taking unfair advantage, imposed an excise tax on crude oil and refined petroleum products. In connection with the new tax, DACO promulgated an administrative order under date of April 23, 1986. The order prohibited wholesalers from passing the tax through to retailers. It also froze wholesale and retail gasoline prices at their March 31, 1986 levels.
When, thereafter, world oil prices soared, the price freeze forced several wholesalers to sell gasoline at prices below their acquisition costs. Since large oil companies are not in business to lose money, a coterie of wholesalers (including the trio that appear as appel-lees here) wasted little time in asking the federal district court to enjoin enforcement of the April 23 order. Moving with equal celerity, the district court scheduled a trial on the merits for May 21, 1986. See Fed. R.Civ.P. 65(a)(2) (authorizing the district court to “order the trial on the merits to be advanced and consolidated with the hearing on the application [for preliminary injunction]”). On May 20, DACO reshuffled the cards; it rescinded the price freeze and issued what it called a “temporary” order that harked back to the former, federally inspired ceiling and established, in lieu of the thawed freeze, maximum GPMs of 8.6<t per gallon for petroleum wholesalers. The May 20 order also scheduled a public hearing for June 2 to “receive comments from all interested persons on the adequacy of this Temporary Order and on any modifications that should be made to attain a situation where primary reliance can be placed on competitive market forces to maintain fair margins at all levels of distribution and fair prices for the consumer.”
This maneuver did not derail the litigation. The district court merely switched tracks, trained its sights on the May 20 edict, and went forward with a three-day bench trial. On June 4 — roughly ten days after the trial ended — the court enjoined enforcement of the May 20 order on federal preemption and other constitutional grounds.
See Isla I,
*873
DACO appealed the preemption ruling to the Temporary Emergency Court of Appeals (TECA),
see
15 U.S.C. § 754(a)(1) (granting TECA exclusive jurisdiction over claims arising directly under the Emergency Petroleum Allocation Act of 1973), and appealed the remaining rulings
{e.g.,
the invalidation of the order on due process and takings grounds) to this court. We stayed proceedings pending consideration of the preemption ruling. TECA affirmed that ruling, see
Isla II,
On June 27, 1989 (the day after we issued our mandate incinerating the district court’s injunction), DACO promulgated an interim order establishing a maximum GPM of 11<C per gallon, effective forthwith. Its final order, issued on November 30, 1989, adopted a ceiling of 13$ per gallon. That order withstood a vigorous constitutional challenge by the wholesalers.
See TPR I,
An ensuing period of unaccustomed tranquility ended abruptly in mid-1992 when DACO again took up the cudgels. It issued a so-called remedial order in which it sought to recoup almost $250,000,000 in profits exceeding an 8.6c per gallon GPM that it estimated three wholesalers — Texaco Puerto Rico, Inc., Esso Standard Oil Co. (P.R.), and the Shell Company (Puerto Rico) Ltd. (appel-lees here) — had earned during the three-year life (June 1986 to June 1989) of the errant injunction. 2 The wholesalers quickly repaired to the district court and requested protection from the remedial order. Before the court could act, DACO issued a revised remedial order. Under its terms, a wholesaler could choose between paying a refund based on a retrospective GPM of 13$ per gallon for the injunction period or paying one based on whatever profit margin would have allowed it to achieve an annual return on assets equal to the average return on assets for the electric utility industry, plus one percent, during the same period.
The wholesalers were not mollified. They challenged the revised remedial order and, on April 1, 1993, DACO rescinded it. This hasty retreat did not restore the peace, for the agency simply attacked on a different front. It revivified the court action originally instituted by the oil companies and filed a motion for restitution seeking an award equal to the excess profits that the wholesalers would have been forced to disgorge but for the pendency of the improvidently issued injunction. Following a tumultuous period of discovery,
see, e.g., infra
Part III (discussing certain disputed discovery rulings), and a three-week bench trial, the court denied the motion for restitution on September 9, 1994.
See TPR II,
II. THE MERITS
Our analysis of the merits is partitioned into four segments. We discuss the nature of restitution, parse the decision below, limn the standard of review, and, finally, examine the record to determine whether the denial of restitution can be upheld.
A. The Nature of Restitution.
In its motion, DACO sought restitution based upon the hoary adage “that a party against whom an erroneous judgment or decree has been carried into effect is entitled, in the event of a reversal, to be restored by his adversary to that which he has lost thereby.”
Arkadelphia Milling Co. v. St Louis S.W. Ry. Co.,
This emphasis on the particulars of each individual case is consistent with the central feature of equity jurisdiction: “the ability to assess all relevant facts and circumstances and tailor appropriate relief on a case by case basis.”
Rosario-Torres v. Hernandez-Colon,
Claims for restitution arising out of the vacation or reversal of a judgment are tested by the same standards as other claims for restitution.
See Atlantic Coast Line,
B. The Decision Below.
The district court predicated its denial of DACO’s motion for restitution on alternative grounds. In the first place, the court determined that there was no benefit to be restored as the wholesalers had not profited from the injunction.
See TPR II,
This latter determination rested upon an analysis of five equitable factors. First, based on evidence regarding the competitiveness of the gasoline market and earnings in other industries, the court found that the wholesalers “did not benefit disproportionately from the lack of regulation.” Id. at 707. Next, the court found DACO guilty of “unreasonable delay” in seeking restitution. Id.- Third, the court concluded that DACO exhibited bad faith with regard to Texaco, Esso, and Shell. See id. Fourth, the court determined that DACO’s actions during the injunction period had lulled the wholesalers into believing that DACO would not demand restitution. See id. at 708. Finally, the court thought that the public interest did not favor a restitutionary order. See id.
C. Standard of Review.
Appellate review often calls into play a blend of rules. So it is here. We review the factual findings that undergird the trial court’s ultimate determination only for clear error.
See Lussier,
The main event evokes a different criterion. We review a district court’s ultimate decision to grant or withhold an equitable remedy for abuse of discretion.
See, e.g., Lussier,
We inspect the voluminous record with these precepts in mind to ascertain whether the denial of DACO’s motion for restitution is sustainable.
D. Discussion.
The court below began with the question of benefit, and treated that question as a discrete inquiry.
See TPR II,
With this preface, we turn to an examination of the judgment below. For ease in reference, we treat each group of factual findings as a separate integer in the equitable equation. The methodologic innovation that we have described — introducing the question of whether the wholesalers benefited from the injunction (and if so, to what extent) into the assay proper — does not require remand. The lower court made detailed factual findings on the question of benefit, and we can easily align those findings along the preferred legal matrix.
See Societe Des Produits Nestle v. Casa Helvetia, Inc.,
1. Benefit. Because restitution is founded on the concept of unjust enrichment, a court considering a request for restitution must investigate the extent to which the target “received a benefit.” Restatement, supra, § 1, cmt. a, at 12. In a case such as this, the problems of proof are readily evident. The regulation that the district court enjoined was clearly labelled as temporary when promulgated, and the injunction prevented further regulation (temporary or permanent). Thus, DACO found itself, at trial, in an epistemological quandary: it had to prove that, had the district court sent the wholesalers packing, it (DACO) would have put into effect a more durable regulation that would have capped GPMs at a level below what the wholesalers actually earned during the pendency of the injunction.
The district court found DACO’s strivings inadequate to this daunting task. In the court’s view, DACO’s adoption of temporary margin controls in 1986 did not “evidence[]
*876
an intent to implement a long-term regulatory plan” to curb profit margins, but, instead, constituted “a short-term erratic response” to an unprecedented situation.
TPR II,
For the most part, this conclusion is adequately anchored in the record. Later actions are often revelatory of earlier intentions,
see, e.g., United States v. Sutton,
The United States Supreme Court decided Isla III on April 19, 1988. Within days, DACO disseminated a press release in which its Secretary, Pedro Ortiz Alvarez (Ortiz), crowed that the Court had “restored to Puer-to Rico the historic power to regulate gasoline prices.” Shortly thereafter, DACO commenced an administrative proceeding to determine whether controls should be introduced. To this end, it requested (and received) financial data and other information from the wholesalers. It also sought industry input as to whether the commonwealth should set either price or margin controls on gasoline, and held public hearings beginning in the fall of 1988 to consider the desirability of controls, the-problems that might arise incident to them, and the reasonableness of existing profit margins in the industry.
In December, as the administrative proceeding wound down, Esso’s general manager, Charles Griffith, met with Secretary Ortiz. According to Griffith, Ortiz informed him that DACO had completed its study and decided against imposing controls because “the market was behaving.” Later that month, a daily newspaper, El Nuevo Dio, published an article based on an interview with Secretary Ortiz. The article reported that DACO had elected “not to regulate gasoline prices and to instead adopt ‘close supervision’ of the industry.” The newspaper quoted Secretary Ortiz as conceding that the wholesalers had not earned “excessive profits.”
In January of 1989, Ortiz resigned. The new Secretary, Jorge R. Ocasio Rodriguez (Ocasio), told Griffith that he was aware of the earlier study and of his predecessor’s conclusions, and that he “intend[ed] to follow [Secretary Ortiz’s] policies” in regard to petroleum wholesalers. In fact, DACO did not adopt controls until June 27, 1989 — the day after the district court’s injunction had been lifted — and then attributed the about-face to newly emergent “erratic and unstable” price fluctuations in the Puerto Rico market.
Noting this chronology of inaction laced with reassurances, and remarking bits of trial testimony such as Secretary Ortiz’ oft-stated preference for a free market system, the district court concluded that DACO’s failure to impose any controls for over a year after the Supreme Court’s decision in Isla III cleared the regulatory path demonstrated that it lacked long-term regulatory intent, and that in all likelihood it would not have regulated wholesalers’ profits during the June 1986 — June 1989 time frame even if the injunction had never issued.
We believe the district court’s finding that, during the injunction period, DACO would not have adopted a permanent regulation limiting profit margins to a level lower than those actually earned by the wholesalers is sustainable on this record. Still, DACO’s assault on this determination possesses con-victive force in one respect. The district *877 court focused almost exclusively on DACO’s actions from and after April of 1988 (when the Supreme Court overruled TECA and gave the green light to state regulation) in attempting to divine DACO’s regulatory intent dating back to mid-1986. This strikes us as a sufficiently accurate barometer of long-term regulatory intent, but fails to deal satisfactorily with the near term. The stopgap order that DACO promulgated on May 20, 1986 would have remained in effect for some period but for the injunction. Thus, even if the district court’s finding is accepted, some benefit — however small — still might have accrued to the wholesalers by reason of the district court’s abrupt suspension of this order.
That said, DACO’s proof does not permit us to quantify that presumed benefit. Because DACO, as the claimant for restitution, bears the burden of proving the conferral and extent of a benefit,
see Atlantic Coast Line,
2.
The Wholesalers’ GPMs.
The district court analyzed the wholesalers’ profit margins during the pendency of the injunction and concluded that they “did not benefit disproportionately from the lack of regulation.”
TPR II,
Once again, DACO’s conception of equity is too inelastic. “The mere fact that a person benefits another is not of itself sufficient to require the other to make restitution therefor.” Restatement,
supra,
§ 1, emt. e, at 13. As the district court noted, a finding that the wholesalers’ actual profit margins were unreasonably high would assist in showing that “the money was received in such circumstances that the possessor will give offense to equity and good conscience if permitted to retain it.”
TPR II,
The district court based its assessment that the wholesalers’ earnings during the relevant period were reasonable on a series of subsidiary findings. It gave weight to the fact that the wholesalers’ profits “were in line with profits earned during the unregulated period after federal controls were terminated, and before the 1986 regulation was enacted.”
TPR II,
DACO suggests that these findings have a tenuous basis in fact — but this is a fairly typical rejoinder of a party seeking to surmount the high hurdle of elear-error review. The district court relied mainly on the testimony of four economists presented as expert witnesses by the wholesalers. We have studied their testimony (including the plethoric exhibits associated therewith), and we are fully persuaded that, given this evidence, the district court had a solid basis for finding that, during the injunction period, the wholesale market in Puerto Rico was staunchly competitive, and that the profits earned by Texaco, Esso, and Shell were reasonable. Although DACO’s expert testified in a diametrically opposite vein, choosing between experts in a jury-waived trial is principally the business of the district court, not the court of appeals.
See, e.g., Keller v. United States,
3.
Delay.
In weighing the equities, the lower court found that “DACO’s actions in seeking restitution have been marked by unreasonable delay.”
TPR II,
It is true that laches ordinarily cannot be raised as a defense against the government in an action brought to enforce a public right or protect a public interest.
See Illinois v. Kentucky,
An equitable defense and an equitable factor are conceptually and practically distinct. The divagation is subtle, but significant. An equitable defense “bar[s] the cause of action entirely, or bar[s] ... the equitable remedy.” 1 Dan B. Dobbs, Law of Remedies § 2.4(1), at 91 (2d ed. 1993). Moreover, in evaluating an equitable defense, the court considers only the plaintiffs conduct and is free to “deny all remedies if the plaintiff does not meet equity’s standards.” Id. § 2.4(5), at 108-09. In contrast, an equitable factor must always be weighed in concert with other relevant factors. See id. at 109. Moreover, as part of balancing the equities, the court “looks at the conduct of both parties and the potential hardships that might result from a judicial decision either way.” Id. From a practical standpoint, then, “[e]ven when an equitable defense does not bar the claim, the total balance of equities and hardships might do so.” Id., § 2.4(1), at 91.
Here, the district court explicitly disclaimed any intent to apply the equitable doctrine of laches as a bar to DACO’s mo
*879
tion.
See TPR II,
DACO also contends that the district court clearly erred in finding prejudicial delay. This contention is unpersuasive. The evidence shows that DACO first raised the refund issue in its June 1989 interim order. DACO did nothing further on this score until ten months later, when it sent letters to the wholesalers conveying its “preliminary views” on the suitability of refunds. DACO then dropped the refund issue like a hot potato and did not resurrect it until August 20, 1992, when the then-Secretary, Guillermo Mojica Maldonado (Mojica), announced at a press conference that he planned to seek refunds from the wholesalers. All told, DACO waited three years after this court vacated the injunction to commit itself to the pursuit of restitution.
DACO does not dispute the accuracy of this chronology, but takes vigorous exception to the court’s conclusion that “[tjhis type of stopping and starting, delaying and then pro-eeeding[,] must be considered prejudicial to the wholesalers, who had to run their business with the threat of multimillion dollar refunds occasionally flaring up and then disappearing.”
TPR II,
Government agencies, like private corporations, have an obligation to conduct their affairs in a reasonably efficient manner.
See Potomac Elec. Power Co. v. ICC,
We will not wax longiloquent. It is trite, but true, that equity ministers to the vigilant, not to those who sleep upon their rights.
See, e.g., Sandstrom v. Chemlawn Corp.,
4.
Bad Faith.
It is old hat that a court called upon to do equity should always consider whether the petitioning party has acted in bad faith or with unclean hands.
See Precision Instrument Mfg. Co. v. Automotive Maintenance Mach. Co.,
In the case at bar, the test was met. The district court found pervasive evidence of bad faith on DACO’s part, directly related to the core elements of the dispute
sub judice. See TPR II,
In making its finding of bad faith, the lower court relied heavily on two occurrences. The court found that, in the spring of 1986, while the government of Puerto Rico was pondering the advisability of an excise tax,
see supra
p. 872, high-level officials, including the President of the Senate and the Secretary of State, summoned executives of the three appellees to a series of private audiences. The court further found that “the wholesalers were warned that they should cooperate with the government in the implementation of the new tax by refraining from further lowering gas prices, so that the government could achieve revenue from the tax....”
TPR II,
DACO claims that these thinly veiled minations had no bearing on margin regulations imposed well after the excise tax was enacted. This claim is disingenuous. Past is prologue, and the district court plausibly could find — as it did — that the 1986 meetings were part of the same overall course of conduct that led to the push for restitution six years later. After all, the meetings involved the same principals and the same subject matter, and, with the benefit of hindsight, can be viewed as a harbinger of things to come. On this basis, the district court did not err in concluding that the 1986 meetings were relevant to DACO’s good faith (or lack thereof) in seeking restitution some years thereafter. This is particularly true in that, shortly after the government “suggested” that the wholesalers refrain from lowering gasoline prices, DACO attempted to justify its regulation of GPMs on the ground that the wholesalers’ prices were too high. Thus, in effect, DACO bore a degree of responsibility for creating the “excess profits” that it later attempted to recapture, first via the excise tax, and then by dint of the motion for restitution.
The second pillar of the court’s conclusion lacks the dramatic impact of these strong-arm tactics, but affords a closer temporal link. The court thought that the actions of Secretary Mojica in and around 1992 betokened bad faith.
See TPR II,
We think that the record as a whole corroborates the district court’s determination that the 8.6$ figure was chosen as a crude club to bludgeon the wholesalers into a settlement, without regard for the economic realities of the petroleum industry. Indeed, the nisi prius roll is replete with evidence suggesting this unhappy conclusion. For one thing, DACO’s chief economist, Carlos La-santa, testified that he had advised his superiors that the 8.6$ margin was economically inadequate, yet DACO persisted in its plan. For another thing, Secretary Mojica testified that he issued the remedial order and set the ceiling without even pausing to review the administrative record. 6
In sum, the grounds relied upon by the district court pass muster. Because the remedy of restitution is premised on the concept of unjust enrichment, DACO’s actions both in 1986 and in 1992 sabotage its present attempt to seize the high ground by asserting that the wholesalers took unfair advantage of the erroneous injunction. Hence, we are unwilling to disturb the court’s determination that DACO’s actions were tinged with bad faith.
5.
Reliance.
A court considering a restitutionary remedy may properly weigh the factor of reliance in its equitable balancing.
See Moss v. Civil Aeronautics Bd.,
521
F.2d 298 (D.C.Cir.1975), cert.denied,
The record is consistent with these findings. It is not farfetched to think that Secretary Ortiz’s statements,
see, e.g., supra
p. 876, could have lulled the wholesalers into a false sense of security.
See, e.g., Insurance Co. v. Mowry,
The court’s finding of detrimental rebanee is bolstered by another circumstance. When Judge Fuste issued the injunction, DACO could have — but did not— ask him to require a bond or an escrow account.
See Inland Steel Co. v. United
*882
States,
6.
Public Interest.
It cannot be gainsaid that a court asked to dispense equitable remediation should give serious consideration to the public interest.
See Morgan,
At trial, DACO made no effort to contradict the wholesalers’ testimony on this point. In this venue, it likewise abjures any challenge to the testimony’s relevance. Instead, DACO complains about the district court’s related statement that DACO had “failed to propose a cogent plan to restore losses” to the Puerto Rico motorists who bore the brunt of the alleged overcharges. Id. In DACO’s eyes, depositing a restitutionary award into the commonwealth’s general fund comprises an entirely satisfactory trickle-down substitute for the court’s envisioned plan of direct payments to motorists.
Once again, DACO’s fascination with a single tree obscures its view of the forest. The district court’s rescript, properly read, does not hold that depositing refunds into the commonwealth’s coffers is repugnant to the public interest in an absolute sense. The court’s point is quite different. Judge Fuste expressed the belief that the clear harm to the Puerto Rico economy that would result from levying a huge restitution award outweighed the benefit accruing from refunds that would not directly compensate the injured victims. Though such a judgment call may be arguable, we are unprepared to say that it represents a clearly erroneous assessment of the evidence.
Cf, e.g., Moss,
7. Recapitulation. We have fashioned a tried-and-true framework for gauging claimed abuses of discretion:
In making discretionary judgments, a district court abuses its discretion when a relevant factor deserving of significant weight is overlooked, or when an improper factor is accorded significant weight, or when the court considers the appropriate mix of factors, but commits a palpable error of judgment in calibrating the deci-sional scales.
United States v. Roberts,
III. OTHER ISSUES
In addition to its assault upon the district court’s equitable determination, DACO mounts a more narrowly targeted offensive on a second front. In this regard, DACO assigns error to a series of discovery rulings that together forced the disclosure of eighteen agency documents, mostly in the nature of correspondence between DACO (or other government representatives) and DACO’s outside counsel. This attempt to open a second front is little more than a diversionary sortie, poorly outfitted and easily repulsed.
We set the stage. In ordering disclosure as a subset of a broader order that DACO turn over the “complete administrative file” in the case to the wholesalers, the court determined that these writings were not entitled to protection under either the attorney-client privilege or the deliberative process privilege. We consider the district court’s privilege rulings cognizant that, “[b]e-cause we regard the existence of a privilege as a factual determination for the trial court ... the district court’s finding of no privilege can be overturned only if clearly erroneous.”
United States v. Wilson,
A. Attorney-Client Privilege.
The Supreme Court has described the attorney-client privilege as “the oldest of the privileges for confidential communications known to the common law.”
Upjohn Co. v. United States,
In its unpublished order requiring revelation of the eighteen documents, the district court rejected DACO’s claim of attorney-client privilege on two grounds. First, the court found that DACO waived any such privilege because four of the documents “were inadvertently shown to Texaco’s legal representatives” during their initial review of the administrative file. 7 We examine the underpinnings of this ruling.
It is apodictic that inadvertent disclosures may work a waiver of the attorney-client privilege.
See, e.g., In re Sealed Case,
In general, a waiver premised on inadvertent disclosure will be deemed to en
*884
compass “all other such communications on the same subject.”
Weil v. Investment/Indicators, Research & Mgmt., Inc.,
The district court’s alternative ground for ordering disclosure is equally solid. The court found as a fact, after in camera inspection of the disputed documents, that outside counsel had become an integral part of the adjudicative decisionmaking process. Based on this factual finding, the court ruled that the attorney-client privilege did not apply because, when an administrative agency engaged in an adjudicative function delegates its responsibilities to outside counsel, then the work product generated by the firm is part of the adjudicative process itself and, hence, beyond the reach of the attorney-client privilege.
DACO resists this analysis, pontificating that such a doctrine “would render the attorney-client privilege meaningless where state or local governments employ counsel and rely on their advice.” Appellants’ Brief at 47. But this trumpeting misapprehends the tenor of the district court’s ruling. The attorney-client privilege attaches only when the attorney acts in that capacity.
See Bay State Ambulance,
We cannot term this finding clearly erroneous. The record shows that DACO’s counsel had, in fact, drafted remedial orders that DACO adopted verbatim; that Dr. Logan, an employee of DACO’s counsel, was the “putative author of the [1989] 13-eent regulation,”
TPRII,
B. Deliberative Process Privilege.
DACO also takes exception to the district court’s ruling that the deliberative process privilege did not exempt the same cache of documents from production. The deliberative process privilege “shields from public disclosure confidential inter-agency memoranda on matters of law or policy.”
National Wildlife Fed’n v. United States Forest Serv.,
The Supreme Court has restricted the deliberative process privilege to materials that are both predecisional and deliberative.
See EPA v. Mink,
Even if a document satisfies the criteria for protection under the deliberative process privilege, nondisclosure is not automatic. The privilege “is a qualified one,”
FTC v. Warner Communications Inc.,
At bottom, then, the deliberative process privilege is “a discretionary one.”
In re Franklin Nat’l Bank Sec. Litig.,
Assuming, arguendo, that the documents at issue are both predecisional and deliberative — a matter on which we need not opine— the district court’s rejection of the deliberative process privilege is nevertheless impervious to DACO’s attack. The court support-ably found that the wholesalers had made a “strong showing” of arbitrariness and discriminatory motives on DACO’s part. Given the discretionary nature of the deliberative process privilege, and the district court’s warranted conclusion that DACO acted in bad faith over a lengthy period of time, see supra Part 11(D)(4), we resist the urge to tinker with the court’s determination that the wholesalers’ interest in due process and fairness outweighed DACO’s interest in shielding its deliberations from public view. 8
C. Harmless Error.
We add a postscript to our discussion of the district court’s discovery rulings. In all events, we do not believe that the district court’s rejection of DACO’s privilege claims affected DACO’s substantial rights. Any error was, therefore, harmless. See Fed.R.Civ.P. 61 (explaining that a court “must disregard any error or defect in the proceeding which does not affect the substantial rights of the parties”).
In denying DACO’s claim for restitution, the district court mentioned only one of the eighteen challenged documents (a June 1989 memorandum from DACO’s outside counsel to Governor Hernandez Colon).
See TPR II,
IV. CONCLUSION
We need go no further. There are neither precise answers nor perfect solutions when a court is forced to deal with the shadowy world of what might have been. Where, as here, the customary deference accorded to the trial court as factfinder is augmented by due respect for that court’s equitable discretion, appellate courts should hesitate to meddle. In this instance, the judge, who had handled the ease from its inception, weighed and balanced the equities, and juxtaposed the parties’ rights with painstaking care. Thus, whether or not we, if writing on a pristine page, might have concluded otherwise, we are unable to tease an abuse of discretion out of what is quintessentially a judgment call.
Affirmed.
Notes
. A GPM represents the difference between the sales price and the seller's acquisition cost. The latter cost includes the price of the gasoline plus excise taxes, but excludes operating costs.
See Tenoco,
. We refer to the three oil companies collectively as "the wholesalers," and individually as "Texaco,” "Esso," and "Shell.”
. DACO's estimate of the benefit received — it says that the wholesalers charged their customers anywhere from $64,500,000 to $250,000,000 more during the injunction period than DACO would have permitted — is not only unproven but also deserves to be taken with a good deal of salt. DACO whips up the lower of these frothy figures by suggesting that, absent the injunction, it would have limited the wholesalers' GPMs to a level no higher than 13$ per gallon throughout the relevant period, and, therefore, that any earnings above that plateau are the fruits of the errant injunction. The higher figure is presumably derived in the same way, but using a projected regulatory ceiling of 8.6$ per gallon (the ceiling imposed in the May 20, 1986 temporary order) for the entire three-year span. These gaudy claims enjoy little or no record support.
. The court used as congeners such benchmarks as returns on assets in the electric utility industry, returns on government bonds, and returns on investments in the industrial distribution services, and fuel industries.
See TPR II,
. The court below offered sound reasons for siding with the wholesalers’ experts. Equally as important, it viewed the testimony of appellant's expert, Dr. Logan, "with some skepticism in light of his intimate involvement with DACO," his former employment by DACO's counsel, and his status as "the putative author of the 13-cent regulation."
TPR II,
. DACO asserts that, because the remedial order "was only the starting point for [its] consideration of the appropriate level of refunds,” the terms of the order “cannot rationally be considered evidence of bad faith.” Appellants’ Brief at 38-39. This ipse dixit does not withstand scrutiny. When Secretary Mojica announced the promulgation of the remedial order, he presented the 8.6$ figure not as a guidepost to a determination of the eventual measure, but as a fait accom-pli. Moreover, the order itself described "a maximum profit margin of 8.6 cents per gallon” as "conclusive and undebatable." DACO retreated from this figure only after the wholesalers sought judicial protection.
. At trial, the district court described how this bevue occurred:
You people [DACO] told them [Texaco’s representatives], here is a room full of papers, you can take a look at them. They looked at them, they found them and then when you discovered that they had seen them and that they wanted copies of those, then you came running here seeking an order.
. We note in passing that the district court's waiver analysis, made in connection with DACO’s claim of attorney-client privilege,
see supra
Part 111(A), arguably applies to the deliberative process privilege as well. Because the privilege lacks vitality here, we will not pursue the question of waiver beyond noting that it is apparently unsettled.
Compare, e.g., Clark v. Township of Falls,
