Summary
affirming district court's admission of expert opinion testimony because the expert "did more than just repeat information gleaned from external sources"
Summary of this case from Camowraps, LLC v. Quantum Digital Ventures LLCOpinion
No. 11–11080.
2013-01-23
FACTORY MUTUAL INSURANCE COMPANY, Plaintiff–Appellee v. ALON USA L.P.; Alon USA GP L.L.C.; Alon USA Refining Incorporated, Defendants–Appellants.
Bruce H. Rogers, John Clifton Hart, Brown, Dean, Wiseman, Proctor, Hart & Howell, L.L.P., Fort Worth, TX, for Plaintiff–Appellee. Jessica Ann Zavadil, Richard Russell Hollenbeck, Thomas Clark Wright, Wright & Close, L.L.P., Houston, TX, for Defendants–Appellants.
Bruce H. Rogers, John Clifton Hart, Brown, Dean, Wiseman, Proctor, Hart & Howell, L.L.P., Fort Worth, TX, for Plaintiff–Appellee. Jessica Ann Zavadil, Richard Russell Hollenbeck, Thomas Clark Wright, Wright & Close, L.L.P., Houston, TX, for Defendants–Appellants.
Appeal from the United States District Court for the Northern District of Texas.
Before WIENER, CLEMENT, and PRADO, Circuit Judges.
PRADO, Circuit Judge:
Plaintiff–Appellee, Factory Mutual Insurance Company (“FM”), was awarded damages stemming from an industrial accident that destroyed a waste treatment plant at an oil refinery plant owned by Defendants–Appellants, Alon USA LP, Alon USA GP LLC, and Alon USA Refining Inc. (collectively, “Alon”). Alon appealed the court's damages determination. We AFFIRM.
I
Alon owns and operates an oil refinery in Big Spring, Texas. It relied on the equipment and services of a third party, Veolia North America–West (“Veolia”), for on-site water treatment and waste management. The equipment located in the waste treatment facility (“the Scalfuel plant” or “the Scalfuel facility”) was owned and operated by Veolia and insured by FM. On February 18, 2008, a cloud of vapor exploded at the Scalfuel facility, destroying it. Veolia filed a claim with FM in the amount of $6,106,880, which FM paid in accordance with the insurance policy. Thereafter, on February 17, 2010, FM filed a subrogation claim against Alon to recover damages stemming from the explosion, alleging that Alon's negligence both directly and proximately caused the damages at issue.
Before the bench trial began, Alon stipulated to liability, leaving only the issue of damages to be determined. At trial, the parties agreed that damages would be determined by the fair market value of the Scalfuel plant before the explosion, but they fundamentally disagreed as to how fair market value should be calculated in this context. FM contended that it was entitled to the Scalfuel plant's replacement cost, i.e., the cost of new parts and labor adjusted downward to account for the original plant's depreciation at the time of the explosion, since there is no market for Scalfuel plants that can be used as a measure of value. On the other hand, Alon argued that FM was only entitled to the cost of the Scalfuel plant's component parts. FM sought $6,106,880, whereas Alon claimed FM could only recover $877,882.
The district court found that, even though there is a market for specific used components, there is no market for used Scalfuel systems. Since the sum price of a Scalfuel system's components does not reflect the full value of an operational Scalfuel plant, the district court found that the fair market value is determined by the replacement cost adjusted for improvements in value beyond the destroyed plant and depreciation reflecting the remaining useful life of the plant before its destruction. Accordingly, the district court found Alon liable for $3,790,391.96, plus interest. To reach this figure, the district court started with an estimate for new equipment, including taxes and shipping, of $2,356,110. Ten percent was added to this amount as a contingency. The combined sum was then multiplied by 2.25 to account for the costs of installation, testing, and startup and the result was then multiplied by 0.65 to account for the original Scalfuel plant's 35% depreciation. Alon timely appealed the district court's judgment, challenging the measure of damages and calculation of fair market value.
Contingency costs generally refer to unforeseeable capital costs that arise during the course of construction.
II
The district court had jurisdiction pursuant to 28 U.S.C. § 1332, as the parties are diverse and the amount in controversy exceeds $75,000. This Court has jurisdiction pursuant to 28 U.S.C. § 1291, as the district court entered final judgment on October 12, 2011 and Alon timely filed a Notice of Appeal on November 10, 2011.
III
On appeal, Alon challenges both the district court's use of replacement cost to determine the market value of the Scalfuel plant, as well as two figures that went into calculating the replacement cost. Specifically, Alon claims that expert testimony concerning the 35% depreciation figure should have been excluded and that the 2.25 multiplier lacked an underlying factual basis.
A
While the parties agree that fair market value is the measure of damages here, they disagree as to how fair market value should be calculated. If a market exists for the property destroyed, i.e., if willing buyers and willing sellers engage in the sale of the property at issue, then comparable sales are the usual measure of value. If a market does not exist, however, then replacement cost is the appropriate measure of value. Here, the district court found that no market for Scalfuel plants exists based on the evidence presented at trial. Accordingly, the district court utilized replacement cost to calculate FM's damages.
1
This Court reviews the district court's findings of fact for clear error. Mid–Continent Cas. Co. v. Davis, 683 F.3d 651, 654 (5th Cir.2012). “A finding is clearly erroneous if it is without substantial evidence to support it, the court misinterpreted the effect of the evidence, or this court is convinced that the findings are against the preponderance of credible testimony.” Becker v. Tidewater, Inc., 586 F.3d 358, 365 (5th Cir.2009) (quoting Bd. of Trs. New Orleans Employers Int'l Longshoremen's Ass'n v. Gabriel, Roeder, Smith & Co., 529 F.3d 506, 509 (5th Cir.2008)). Thus, when the fact finder is faced with two permissible views of the evidence, the choice between them cannot be clearly erroneous. Davis, 683 F.3d at 654.
2
“A plaintiff whose property has been destroyed by the tort [ious] acts of another is generally entitled to recover the market value of the property at the time of its loss.” Seminole Pipeline Co. v. Broad Leaf Partners, Inc., 979 S.W.2d 730, 754 (Tex.App.–Houston [14th Dist.] 1998, no pet.); see also Waples–Platter Co. v. Commercial Standard Ins. Co., 156 Tex. 234, 294 S.W.2d 375, 376 (1956). The measure of damages is “the difference in its market value immediately before and immediately after the injury, at the place where the damage occurred.” Thomas v. Oldham, 895 S.W.2d 352, 359 (Tex.1995). Market value is the amount a willing buyer, who is under no obligation to buy, would pay to a willing seller, who is under no obligation to sell. Id. Importantly, however, not all property has a market value. Id. “[I]n situations where a market value does not exist, ... replacement value is the means of assessing damages.” Id. (quotation omitted); see also Gulf States Utils. Co. v. Low, 79 S.W.3d 561, 569 (Tex.2002). Such a situation arises when, for example, “comparable sales figures are lacking....” City of Harlingen v. Estate of Sharboneau, 48 S.W.3d 177, 183 (Tex.2001).
Here, the district court determined—and the parties do not dispute—that market value is the correct measure of damages. That is, FM is entitled to recover the value of the Scalfuel facility immediately before the explosion, since the facility was worth nothing after the explosion occurred. Whether replacement cost or the estimated price of used component parts constitutes the appropriate measure of value prior to the explosion forms the foundation of the parties' dispute.
This appeal centers on a determination of fact made by the district court in calculating market value, namely that “[t]here is no market for used Scalfuel systems.” The court so held because each Scalfuel system is unique and
includes some standard constituent subsystems and some subsystems that are specifically constructed for the Scalfuel system. While there is a market for some of the used subsystems, the market price of such used subsystems does not reflect the market value of a running Scalfuel system. The price of the used subsystems does not reflect the expense of integrating the parts into a working Scalfuel system. No reasonable person would construct a working Scalfuel system out of used components.
If, as the district court found, “there is no market price for a generic Scalfuel system,” then replacement cost—adjusted for betterment and depreciation—is the correct measure of value. On the other hand, if the value of a Scalfuel plant can be measured by pricing the plant's component parts, as Alon argues, then that approach is the appropriate method for measuring damages. As explained below, the district court did not commit clear error in finding that no market for Scalfuel facilities exists. Therefore, replacement cost was an appropriate measure of damages, and we affirm.
Alon argues that the Scalfuel system at issue here was comprised of a number of component parts, many of which were readily procurable on the open market. Because used versions of most Scalfuel system parts could be procured from vendors, Alon asserts that market value is more appropriately measured by pricing the individual components likely present at the Scalfuel plant before the explosion.
The measure of value is based on components “likely” present at the time of the explosion because there do not appear to be any documents in existence that comprehensively detail the component parts of the Scalfuel plant prior to the explosion. As a result, the parties' experts were forced to estimate what equipment was present.
In support of this claim, Alon relies heavily on Hartford Insurance Co. v. Jiminez, 814 S.W.2d 551 (Tex.App.–Houston [1st Dist.] 1991, no pet.), a case in which the appellate court affirmed a trial court's finding that there was no evidence as to the amount of damage the plaintiffs sustained, despite their demonstrating that the automobile had been totaled and that the insurance company paid approximately $7,000 as a result. 814 S.W.2d at 552. The court so held because “[w]hat an insurance company paid is not evidence of reasonable market value” and the plaintiffs' evidence of damages consisted solely of proof regarding payment from his insurance company. Id. Thus, the court determined that “no evidence whatever was admitted showing [the car's] reasonable market value at the time of the collision.” Id. Alon, however, relies on Jiminez for a much broader proposition than the case supports. Whereas Jiminez can at best be read to support the claim that insurance payments may not be used to prove damages, Alon goes so far as to claim that Jiminez means “evidence of replacement cost is no evidence of market value damages.” To the contrary, while it is a disfavored method, replacement cost can indeed provide a competent measure of damages. See Gulf States, 79 S.W.3d at 569.
Furthermore, Jiminez is readily distinguishable from this case in two ways. First, unlike in Jiminez, FM did not rely on evidence of its payment to Veolia to prove the amount of damages. Instead, FM relied on expert opinions regarding valuation. Indeed, one witness identified some 74% of the equipment initially present at the Scalfuel plant and sought quotations from vendors regarding price. The plaintiff in Jiminez merely presented evidence regarding the amount his insurance company paid for the car, 814 S.W.2d at 552, whereas FM utilized the services of an appraiser to determine the value of the Scalfuel facility prior to the explosion. FM did not rely on proof of its payment to Veolia to prove market value.
Alon's appeal includes challenges to some of those experts' opinions. Those specific contentions are discussed in Parts III.B and III.C, infra.
Second, the automobile in Jiminez unequivocally had an ascertainable market value that can be determined using comparable sales in a specific geographic area. See City of Harlingen, 48 S.W.3d at 183 (holding that market value using comparable sales is preferable, but that replacement cost may be used when comparable sales are lacking). But unlike an automobile, the Scalfuel facility here is a specialized, integrated set of systems for which no market exists. Scalfuel plants are not regularly bought or sold, each Scalfuel facility has proprietary component parts, and the plant's underlying process is itself patented. Moreover, Alon's estimate was limited to the cost of procuring component parts; the figure given by Alon did not account for installation, on-site engineering, or startup, all of which are critical to the value of a Scalfuel facility. Alon's expert conceded that his $877,882 figure amounted to equipment that was “sitting on the ground, not assembled.” The price for an automobile in Jiminez would account for a complete, working vehicle rather than a pile of used parts. Anthony Foster, President of ChemTech Consultants, further stated that it is “unwise” and highly uncommon to build a Scalfuel plant “completely out of used equipment.” These characteristics readily distinguish FM's claim from Jiminez.
In their brief, appellants claims that their expert “estimated the number of man-hours needed to construct the unit.” This is an untrue statement. The testimony Alon cites merely points to Alon's expert estimating the number of man-hours it would take to estimate the cost of completely rebuilding the Scalfuel plant. That is, Alon's expert did not estimate the number of man-hours it would actually take to build a new Scalfuel plant. This distinction is reiterated later in his testimony: “If you're asking me if I have included the cost of installing the plant and getting the plant running in [the damages figure], the answer is, no, I have not.”
Based on the evidence presented, the district court did not err when it found that no market for Scalfuel systems exist. Ample evidence was presented to support such a finding. Furthermore, Alon never actually addresses the finding that no market exists. Rather, they reiterate their ability to price out component parts, assuming that this fact alone precludes the use of replacement cost as a measure of damages. This position effectively glosses over the substance underlying the district court's conclusion that a market for complete, operational Scalfuel plants does not exist.
While Alon presented evidence that the component parts of a Scalfuel plant could be priced individually, weighed against FM's proffer, the district court was presented with two permissible views of the evidence. In comparing the parties' arguments, the district court gave weight to factors such as labor, layout, and installation because the market value of a fully operational Scalfuel plant is greater than the sum of its component parts. Considering these factors alongside the unique layout, structure, and design of each Scalfuel plant, the district court determined that a market for Scalfuel plants does not exist. In light of the deferential standard of review applicable, no manifest error was committed. It necessarily follows that replacement cost was the appropriate measure of damages. Therefore, we affirm.
B
1
“This Court reviews the admissibility of expert testimony for abuse of discretion.” Primrose Operating Co. v. Nat'l Am. Ins. Co., 382 F.3d 546, 561 (5th Cir.2004) (citing Vogler v. Blackmore, 352 F.3d 150, 153 (5th Cir.2003)). The district court's discretion will not be disturbed on appeal unless the its decision was manifestly erroneous. Id. (citing United States v. Tucker, 345 F.3d 320, 326 (5th Cir.2003)).
2
Expert witnesses may base opinions on facts or data that the expert “has been made aware of or personally observed.” Fed.R.Evid. 703. If the facts and data relied upon are the sort that experts in that field would reasonably rely on, then those facts “need not be admissible for the opinion to be admitted.” Id. Accordingly, experts may base their opinions on otherwise-inadmissible information, such as hearsay, so long as the information is the sort reasonably relied upon in the experts' field.
The purpose of this rule is largely practical: experts generally base their opinions on information which, to be admissible in court, would entail “the expenditure of substantial time in producing and examining various authenticating witnesses.” Fed.R.Evid. 703 advisory committee's note. Because experts may use their past experience and professional judgment to make critical decisions on the basis of such information outside of court, Rule 703 was intended “to bring the judicial practice into line with the practice of the experts themselves when not in court.” Id. Courts nevertheless must serve a gate-keeping function with respect to Rule 703 opinions to ensure “the expert isn't being used as a vehicle for circumventing the rules of evidence.” In re James Wilson Assocs., 965 F.2d 160, 173 (7th Cir.1992). Rule 703 “was not intended to abolish the hearsay rule and to allow a witness, under the guise of giving expert testimony, to in effect become the mouthpiece of the witnesses on whose statements or opinions the expert purports to base his opinion.” Loeffel Steel Prods., Inc. v. Delta Brands, Inc., 387 F.Supp.2d 794, 808 (N.D.Ill.2005). The rule “was never intended to allow oblique evasions of the hearsay rule.” Id.
Here, FM presented an expert appraiser, Leslie H. Miles, Jr. (“Miles”), to testify regarding the value of the Scalfuel plant. Part of the testimony offered by Miles dealt with the remaining life of the original Scalfuel plant at the time of the explosion. This testimony was a necessary part of calculating damages because, without adjusting the replacement cost of the new Scalfuel plant downward to account for the original plant's depreciation, FM would reap a windfall in its overall recovery. Since the original equipment was no longer available, Miles met with individuals who were purportedly familiar with the original Scalfuel plant and attempted to educate them regarding “what depreciation is made up of and how you calculate it.” This is a method Miles has apparently used in the past.
The Veolia employees estimated that, prior to the explosion, the Scalfuel plant still had 65% of its remaining life; in other words, it was 35% depreciated at the time of the explosion. Based on the information he provided the employees and the discussion he witnessed, Miles deemed the employees' estimate reliable, and even expressed surprise at the “aggressiveness of it....” Miles did not expect the employees to estimate such a low remaining life given his experience with various chemical processing plants. That said, Miles appears merely to have adopted the depreciation number provided by Veolia's employees; it is not clear how he used past experience as an appraiser to deem the employees' estimate reliable. It is on this ground that Alon challenges the district court's admission of Miles's testimony as violative of Rule 703.
Alon argues that the district court abused its discretion by allowing Miles to testify regarding the Scalfuel plant's estimated depreciation because “Miles knew that physical depreciation had occurred ... but he did nothing to calculate that depreciation” beyond relying on a figure estimated by Veolia employees. In support of this claim, Alon relies on United States v. Mejia, 545 F.3d 179 (2d Cir.2008), for the proposition that experts may not simply transmit hearsay to the jury. 545 F.3d at 197. Specifically, “the expert must form his own opinions by applying his extensive experience and a reliable methodology to the inadmissible materials.” Id. (internal quotation marks omitted). Otherwise the expert may simply parrot impermissible hearsay evidence, thereby allowing a party to circumvent the rules against hearsay. Id. In Mejia, the Second Circuit ruled that the district court had abused its discretion in allowing testimony from one of the government's expert witnesses regarding gang structure and operations. Id. at 197–98. The court held that some of the expert's testimony ran afoul of Rule 703 because he had merely repeated information gathered from other sources without articulating how he applied his expertise to the underlying information being relayed. Id.
Alon also relies on Loeffel Steel Products v. Delta Brands, Inc., for similar reasons.In Loeffel, which involved an expert's testimony regarding economic loss in a steel factory, a magistrate judge determined that the expert's testimony violated Rule 703 because the numbers he relied upon to determine economic loss “came from the defendants' employees, on whom [the expert] uncritically relied.” 387 F.Supp.2d at 807. Furthermore, the expert had little understanding of the underlying industrial processes he was evaluating and broadly “brought no expertise to bear on the underlying assumptions on which his economic loss theory was based....” Id. The expert had very little to offer beyond uncritical reliance on figures provided by others.
In response, FM argues that Miles “clearly explained how he arrived at the factors that he used to determine the depreciated value of the Scalfuel plant.” This statement is somewhat misleading, though. On the one hand, Miles did clearly articulate what depreciation means and how it is usually calculated. On the other, however, Miles's testimony does not reveal any particular expertise brought to the process of evaluating the number provided by Veolia's employees. Miles did provide guidance to the employees regarding depreciation theory, but he relied solely on the employees' estimation, a figure he deemed reliable based on the discussion he observed. There was no way to verify that the Veolia employees competently applied the considerations on which Miles had instructed them, though Miles did state that he was surprised by the “aggressiveness” of the employees' estimate.
That said, Miles also testified that the estimates of others constitute the sort of information reasonably relied upon by appraisers approaching valuation questions. Neither Mejia nor Loeffel is entirely apposite, and neither is binding upon this Court. Insofar as he educated and interviewed Veolia employees, Miles did more than just repeat information gleaned from external sources. Cf. Mejia, 545 F.3d at 198 (The witness “did not analyze his source materials so much as repeat their contents. [He] thus provided evidence to the jury without also giving the jury the information it needed” to consider the reliability of the underlying sources.). Furthermore, Miles demonstrated his familiarity with the appraisal of heavy industrial plants broadly, even if he had little experience with Scalfuel plants in particular. Cf. Loeffel, 387 F.Supp.2d at 807 (stating as “undisputed” the fact that the expert witness had no experience with the relevant machinery and was “incapable of assessing the validity of the information provided....”).
In light of the deferential standard on appeal, we affirm because the district court did not abuse its discretion. Miles did clearly state that the sort of information relied upon here—the opinions of others—is the sort of information reasonably relied upon by appraisers. Moreover, Miles's investigation must be viewed in light of what was feasible. Since the original Scalfuel plant was destroyed and scrapped after the explosion, there was very little room for investigation of any sort. While Miles could have doubtless come to a more accurate estimate by inspecting records or the equipment itself, neither was available. Miles thus consulted one of the few sources of information available: employees who had worked at or near the Scalfuel facility. In light of these considerations, the district court was best placed to evaluate whether Miles uncritically relied upon the depreciation figures given to him by Veolia's employees through his testimony.
The parties do not address why it was not possible to directly depose the employees on whom Miles relied for his depreciation figure, though Alon does briefly raise the issue. Indeed, the third case relied upon by Alon, In re James Wilson Associates, 965 F.2d 160 (7th Cir.1992), involved a similar issue. In that case, which centered on estimating the physical deterioration of a building for valuation purposes, the testimony of an architect was deemed violative of Rule 703 because the architect was merely parroting information given to him by the engineer who had actually inspected the building at issue. 965 F.2d at 173. As the court stated, “The issue was the state of the building, and the expert who had evaluated that state—the consulting engineer—was the one who should have testified.” Id. The problem was rooted in the fact that the architect was seeking to repeat information on a topic outside of his expertise instead of using “what the engineer told him to offer an opinion within the architect's domain of expertise....” Id. This case is distinguishable, however, since Miles sought to use information outside of the employees' expertise to offer an opinion within his own expertise.
C
1
This Court reviews the district court's findings of fact for clear error. Mid–Continent Cas. Co., 683 F.3d at 654;see also Part III.B.1, supra.
2
At trial, the district court heard from multiple experts concerning the propriety of using a multiplier when calculating replacement costs. In this context, a multiplier refers to a number that, when applied to the underlying cost figure, is intended to account for anticipated costs associated with construction, including installation, startup, overhead, and testing. Alon's expert, Dean Harris (“Harris”), would have applied a multiplier of 1.25 or 1.5 at most. FM's witness, Tony Foster (“Foster”), initially recommended a 2.0 multiplier, but later changed his mind and advised that a 2.5 multiplier was more appropriate. Veolia, which has substantial experience in the construction of Scalfuel plants, has apparently encountered multipliers ranging from 1.8 to 3.2. The district court ultimately used a multiplier of 2.25, which is the compromise figure that FM used when evaluating Veolia's claim. On appeal, Alon contends that FM's proffered 2.5 figure and the district court's adopted 2.25 figure lack factual bases. As explained below, clear error was not committed.
First, Alon's attempt at undermining FM's witness testimony is unavailing because it misconstrues the evidence at issue. During trial, Alon pressed Foster regarding a pair of letters that ostensibly recommended a 1.0 multiplier for calculating the cost of a new Scalfuel facility. Specifically, one document stated: “With this design concept in mind, ChemTech believes that a fair Total Installed Cost is 1 x the estimated equipment cost, or $2.6 million equipment plus $2.6 installation.” The second letter said: “ChemTech ... recommends a multiplier of 1.0 times the skidded equipment cost shown on the summary sheet at $2.6 million.” Alon implied during this line of questioning that FM's own witnesses did not believe a multiplier between 2.0 and 2.5 was warranted. However, Foster claimed in his response that the 1.0 text was a “typo.” Both documents in fact recommend doubling the estimated equipment cost to account for concomitant costs associated with installing the equipment. This means applying a 2.0 multiplier, even if FM misused the term in its correspondence. Indeed, a 1.0 multiplier would do no multiplying at all.
The first letter stated that “Total Installed Cost is 1 x the estimated equipment cost, or $2.6 million equipment plus $2.6 million installation.” While not a model of clarity, it is difficult to dispute that the final cost figure should be double the estimated cost of equipment. That is, the estimated equipment cost is $2.6 millionand the estimated installation cost should be the same amount, i.e., “1 x the estimated equipment cost”, effectively doubling the total installed cost.
The same is true of the second letter, which made the same mistake. It states that “ChemTech ... recommends a multiplier of 1.0 times the skidded equipment cost shown on the summary sheet at $2.6 million.” As Foster explained, multipliers can be conceptually understood in numerous ways: “The common thing is you can take a percentage of value, say, a hundred percent of the cost increase. If you bought a wallet for $10, a hundred percent of it would be another $10, you know? ... The other simpler method of doing it is two times” the original amount. The letters clearly convey the opinion that the estimated equipment cost should be doubled to account for installation costs, regardless of how they used the term multiplier. Alon's reliance on these documents is thus misplaced and does nothing to undermine FM's evidence or the district court's conclusion concerning an appropriate multiplier.
Finally, Alon relies on the testimony of Harris, its own expert, who testified that multipliers are generally disfavored for their inaccuracy and should only be used for broad cost estimates. While Harris clearly believed that use of a multiplier was not appropriate in these circumstances, his testimony does not definitively settle the issue. FM's witnesses agreed that multipliers are not ideal, but suggested that they are nevertheless appropriate in cases such as this in which no better information is available. Given the lack of useful records and resources pertaining to this particular Scalfuel plant, it was entirely reasonable to conclude that use of a multiplier was appropriate. As experts for both parties testified, multipliers are best suited for just such a situation.
The district court was clearly presented with two permissible views of the evidence. Few records were available to estimate the cost of rebuilding the Scalfuel plant, which counsels in favor of using a multiplier. Furthermore, the 2.25 multiplier used by the district court is well within the range recommended by the witnesses and is consistent with Veolia's past experiences. Accordingly, the district court did not clearly err, and we affirm.
IV
For the foregoing reasons, the judgment of the district court is AFFIRMED.