ORDER AND JUDGMENT
TERRENCE L. O’BRIEN, United States Circuit Judge.
INTRODUCTION
This appeal concerns the tax treatment of a conservation easement. The taxpayer, Trout Ranch, LLC (Trout Ranch), purchased some 450 acres of land in Gunnison County, Colorado, with the aim of developing home sites on a small segment while preserving the rest, approximately 85 percent of the property, by means of a conservation easement. Trout Ranch claimed a charitable deduction for the easement, which it valued at approximately $2.2 million. The IRS determined the easement did not reduce the value of the taxpayer’s property, appraised the charitable contribution at zero, and disallowed the deduction. The tax court concluded the proper value of the easement was $560,000. Trout Ranch challenges that figure on appeal and insists the correct valuation is higher, closer to the $2.2 million estimate it originally offered. Finding no errors in the tax court’s decision, we affirm.
BACKGROUND
A.
Gunnison Riverbanks Ranch
Gunnison Riverbanks Ranch sits on 453 acres in Gunnison County, Colorado. The property is situated between the towns of Gunnison, six miles to the south, and Crested Butte, 20 miles to the north. The crown jewel of the property is two miles of frontage on the Gunnison River, a stream beloved by fisherman for its world-class Rainbow and German Brown Trout.
Since Gunnison County has no zoning laws, land use is governed by Colorado law, under which developers may subdivide land into 35-acre parcels as a matter of right, with additional subdivision subject to the approval of county planning commissions. Colo. Rev. Stat. Ann. § 30-28-101. Gunnison County uses a system of transferable development rights called the Large Parcel Initiative Process (LPIP). The system promotes conservation by encouraging developers to permanently restrict development of large parcels of land (preserved tracts) in exchange for the right to develop unpreserved tracts more densely. Under the LPIP, a developer who preserves 75 percent of his property can subdivide the remainder into three lots for every 75 acres he owns (rounded down to the nearest multiple of 35). If a developer preserves 85 percent, he gets an additional lot for every 140 acres. So while Colorado law allows a property with Gun-nison Riverbanks Ranch’s dimensions to be divided into twelve 35-acre lots, the LPIP would permit as many as twenty-two 3-acre lots.
Trout Ranch purchased Gunnison Riverbanks Ranch in 2003 with plans to develop a residential subdivision under the LPIP and preserve the remainder with a conservation easement. The price was $3,953,268. According to the development plan, the land not encumbered by the easement would be subdivided into 21 residential lots, with an additional lot for a club
house. Lots would average approximately three acres and owners would have access to a host of shared amenities, including a clubhouse, a boat house, riding stables, duck blinds, an archery range, and three ponds. In addition, the lot owners would enjoy preferred, if not exclusive, right to use the “conserved lands,” in particular the right of access to nearly two miles of frontage on the Gunnison River. The deal was consummated in December 2003, when Trout Ranch conveyed an easement encumbering approximately 85 percent of the property — 384 of 453 acres — to the Crested Butte Land Trust. The plan was approved by Gunnison County in April 2004.
B.
The Tax Return
Trout Ranch elected to be taxed as a partnership for the 2003 tax year. On its 2003 return, the partnership claimed a charitable deduction of $2,179,849 for the conservation easement. Years later, in March 2008, the Commissioner issued a notice of adjustment reducing the 2003 deduction to $485,000. The Commissioner would later receive permission from the tax court to reduce the adjusted value to zero and disallow the charitable deduction entirely. Trout Ranch filed a petition in the tax court objecting to the adjustment and asserting the Commissioner had substantially undervalued the conservation easement. The value of the easement was the only issue raised in the petition; both parties agree the easement is a “qualified conservation contribution” and therefore a proper basis for a deduction under I.R.C. § 170(f)(3), which creates an exception to the general rule that a taxpayer will not be granted a deduction for a charitable contribution of a partial interest in property.
C.
Battle of the Appraisers
Both sides introduced the testimony of expert appraisers in support of their asserted valuations. The experts worked under the framework for valuing conservation easements set forth in the Treasury regulations. Treas. Reg. § 1.170A-14(h)(3)(i). Under this framework, the value of a conservation easement is “the fair market value of the [easement] at the time of the contribution.”
Id.
The regulations prescribe two methods for ascertaining fair market value: “If there is a substantial record of sales of easements comparable to the donated easement ... the fair market value of the donated easement is based on the sales prices of such comparable easements.”
Id.
But if no substantial record of sales is available, “as a general rule (but not necessarily in all cases) the fair market value ... is equal to the difference between the fair market value of the property it encumbers before the granting of the restriction and the fair market value of the encumbered property after the granting of the restriction.”
Id.
Finding the before and after values requires a determination of the property’s income-producing potential. To this end, appraisers will generally construct discounted cash-flow models, which estimate the present value of a property by estimating future revenue (in this case, revenue from lot sales) and discounting it based on the cost of capital. Naturally, the before-and-after approach is more common in remote areas like Gunni-son County where records of comparable sales are scarce.
Trout Ranch appraiser Jonathan Lengel was the only expert to perform a valuation based on the comparable-sales method. (Lengel also provided a valuation using the before-and-after method, but the comparable sales figure is what ultimately formed the basis for his opinion). This was Len-gel’s first time using the comparable-sales method, and he admitted at trial that he would probably have stuck to the before-
and-after approach had his client not urged him to consider an alternative method.
Lengel based his comparable-sales valuation on four sales of conservation easements in Gunnison County, but, on cross-examination, it was apparent that he had not carefully vetted the comparisons. The sales had been recorded by a local conservation group and listed on a matrix that omitted most of the information necessary for a meaningful comparison, including the consideration exchanged, the relevant restrictions, and the profitability of the underlying land. Worse yet, Lengel admitted he had not examined the deeds or appraisals for any of the easements and all of the comparable transactions involved bargain sales to charities in which the purchaser paid less than the appraised price, making it difficult to determine the actual value of the easement.
In spite of these flaws, Lengel’s comparable-sales analysis estimated the value of the easement between $1.59 million and $2.3 million, a figure in line with the before-and-after appraisal he submitted with Trout Ranch’s tax return ($2.2 million), and not far from the before-and-after appraisal he had submitted in a supplemental report in anticipation of trial ($3 million). In the end, he placed the value of the easement at $2.2 million, an estimate based primarily on comparable sales.
In the Commissioner’s corner were expert appraisers Lou Garone and Michael Nash. Both Garone and Nash performed before-and-after appraisals using discounted cash-flow models to determine the most profitable use of the property before and after the easement. In evaluating the most profitable use before the easement, Garone considered configurations as sparse as 12 lots and as dense as 60, but concluded a 22-lot model, much like the one prepared by Trout Ranch, would be the most lucrative use of the land, with a projected value around $5 million. Since everyone agreed the same 22-lot configuration represented the best use of the property after the easement, Garone concluded the value of the easement was zero.
Because Nash agreed with Garone — a 22-lot configuration was the best use of the land both before and after contributing the easement — he concluded the value of the easement was zero. In his view, Trout Ranch wisely availed itself of the LPIP process, which allowed it to avoid the costs and procedural risk of a prolonged approval process. Nash also constructed a discounted cash-flow model which yielded a $5,650,000 value for the property, to which he added $725,000 for improvements made by Trout Ranch, for a total value of $6,300,000.
D.
Tax Court Opinion
The tax court refused to accept any of the expert appraisals in their entirety. Rather, it achieved a different result by borrowing bits of data from each report to construct its own discounted cash-flow model.
As a preliminary matter, the tax court concluded the before-and-after approach was the appropriate valuation method because there was “no substantial record of sales of easements comparable to the donated easement.” Doc. 42 at 12. The comparable sales identified by Lengel were off the mark, the court explained, because most of the underlying easements were more restrictive and, in turn, more detrimental to the profitability of the encumbered land. In three of the examples, the easement reduced developable land by
at least 89 percent, compared to the Trout Ranch easement which reduced potential development by only 45 percent. In a fourth example, the only sale involving an easement that did not all but foreclose the possibility of development, the easement still restricted development of the most valuable areas of the property, including the creek frontage. This was the exact opposite of what happened in this case, where Trout Ranch
concentrated
development in the most valuable areas of the property. Moreover, each of the comparable sales identified by Lengel were bargain sales in which the purchaser paid less for the land than the appraised value, making it difficult to ascertain the true value of the easement.
Turning to the before-and-after method, the court agreed with the experts’ consensus that the most profitable use of the land post contribution would be a 22-lot residential subdivision. Recognizing disagreement among the experts over the potential value of such a development, the court conducted its own analysis using sales data from the nearby Hidden Valley Ranch, where similarly sized lots had been selling in a range between $320,000 and $430,000. Since the court viewed the Hidden Valley property as comparable but less desirable than Gunnison Riverbanks Ranch, the court added a $60,000 premium to the top of the sales range in order to arrive at an estimated lot price of $490,000. That figure was then considered in conjunction with factors such as the number of lots, absorption rate, overhead, appreciation, and discount rate to reach a final post-contribution value of $3.89 million. With respect to the pre-contribution value, the court used a hypothetical 40-lot residential subdivision with an estimated value of $4.45 million. The court then subtracted from that figure the $3.89 million post-contribution value to arrive at a total before-and-after value of $560,000.
DISCUSSION
Trout Ranch challenges the tax court’s valuation on several bases. First, it contends the tax court should have excluded the testimony of Nash and Garone on the ground it was unreliable and contrary to Treasury regulations. Second, it disagrees with the tax court’s valuation and contends the court erred by failing to consider comparable easement sales and by relying on data from after the date Trout Ranch donated the easement.
A.
Whether the tax court erred by admitting the testimony of Commissioner experts Garone and Nash
According to Trout Ranch, the testimony of Nash and Garone was inadmissible because the experts failed to follow the method prescribed in the Treasury regulations in conducting their appraisals. Specifically, it contends the two experts used the before-and-after analysis without giving due consideration to sales of comparable easements, the first appraisal method set forth in the regulations.
As previously explained, the value of a conservation easement is the “fair market value of the perpetual conservation restriction at the time of the contribution.” Treas. Reg. § 1.170A-14(h)(3)(i). That value is often ascertained by finding the “difference between the fair market value of the property [the easement] encumbers before the granting of the restriction and the fair market value of the encumbered property after the granting of the restriction.”
Id.
But where a record of sales of comparable easements is available for meaningful comparison, the regulations re
quire the appraisal be based on the comparable sales rather than the before-and-after method.
Id.
Determining fair market value is a factual inquiry governed by the Federal Rules of Evidence, Tax Court Rule 143(a). The tax court serves as a gatekeeper for admitting expert evidence, much as the trial judge does in civil and criminal trials.
See Daubert v. Meirell Dow Pharm., Inc., 509
U.S.
579, 597,
113 S.Ct. 2786, 125 L.Ed.2d 469 (1993);
Kumho Tire Co. v. Carmichael,
526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). We review admissibility determinations for abuse of discretion and afford wide latitude to its decisions on the admissibility of expert testimony.
See Bitter v. A.O. Smith Corp.,
400 F.3d 1227, 1232 (10th Cir.2004).
The tax court properly admitted the expert testimony of Nash and Garone. As professional appraisers, they determined the value of the easement by calculating and comparing the best use of the property before and after Trout Ranch donated the easement. To this end, they constructed discounted cash-flow models to determine the present value of the ranch’s future revenue. This remains an accepted method for determining the value of a real estate interest where no comparable sales are available,
see
Treas. Reg. § 1.170A-14(h)(3)(i);
Marine v. Comm’r,
92 T.C. 958, 983 (1989), and there is nothing in the record suggesting Nash and Garone were wrong to use it in this case. Having settled on a valid method, Nash and Garone reliably applied it, pooled data from sales at Gunnison Riverbanks Ranch and nearby properties, separated relevant figures from irrelevant ones, and then analyzed the remaining data in light of a host of factors to determine the most profitable use of the land before and after the easement.
Trout Ranch’s main objection is that Nash and Garone applied the before-and-after method without giving proper consideration to comparable sales. In this regard, Trout Ranch is not merely challenging the decision to use the before-and-after method, though it does contend the comparable sales method was more appropriate here; rather, it asserts Nash and Garone were not reliable experts because they failed even to
consider
the comparable sales method.
This assertion is false. Both experts acknowledged the comparable-sales method in explaining their appraisal; they simply chose not to follow it. In any event, we cannot see what this argument has to do with the tax court’s reliability determination under Rule 702. Trout Ranch seems to be saying the analysis submitted by Nash and Garone was unreliable because neither expert discussed their reasons for not using the comparable-sales method. But under the Treasury regulations, they had no duty to explain why they chose one method over the other; they had only to choose the appropriate method and apply it correctly. Treas. Reg. § 1.170A-14(h)(3)(i). The notion that some form of preliminary analysis was required and that an expert’s failure to provide it could undermine the reliability of his analysis, seems to have been pulled from thin air.
In any event, Nash and Garone were correct to overlook comparable sales and focus their attention on the most profitable value of the property before and after the easement was granted. Trout Ranch’s argument to the contrary is based on a misplaced reliance on the comparable sales Lengel identified in his report. As the tax court observed, the easements involved in those sales were substantially more re
strictive than the one at issue here, both in terms of the quantity of land covered and the quality of the land restricted. In all but one case the “comparable” easement effectively eliminated the possibility of commercial development, and in the one case where there was still room to develop, the developable land comprised the least desirable areas of the property. In other words, the easements on record were considerably more detrimental to the value of their underlying properties than the easement encumbering Gunnison Riverbanks Ranch.
B.
Whether the tax court’s opinion was clearly erroneous.
Recall that the tax court set out to determine the value of the easement by finding the difference in the fair market value of the property before and after the easement was donated. All agreed a residential subdivision represented the best use of the land, and the three experts used potential-income models to calculate the before-and-after values. Unconvinced by any of the expert reports in their entirety, the tax court took data and analysis from each and constructed its own discounted cash-flow model to project the revenue from a hypothetical residential subdivision. Although lot price dominates the analysis, it is but one of several factors assessed in the court’s model, which also includes lot density, the absorption rate, appreciation, capital expenses, overhead, developer’s profit, and the discount rate. The court’s before-donation appraisal, $4.45 million, was based on a 40-lot configuration; the after-donation appraisal, $8.89 million, on a 22-lot configuration. The court appraised the conservation easement at $560,000, the difference between the before-donation value and the after-donation value.
We review the tax court’s valuation of property as we would any question of fact — for clear error.
See Estate of Holl v.
Comm’r, 54 F.3d 648, 650 (10th Cir.1995). The court has broad discretion in forming its own conclusions about the record. It can find facts and accept or reject expert testimony as it sees fit.
Helvering v. Nat’l Grocrery Co.,
804 U.S. 282, 294-95, 58 S.Ct. 932, 82 L.Ed. 1346 (1938);
Kiva Dunes Conservation, LLC v. Comm’r,
97 T.C.M. (CCH) 1818, 1820 (2009).
Trout Ranch contends the tax court erred by relying on data involving sales that took place after the easement was donated. This assertion appears to embrace two related arguments: first, a legal argument that data from after the date of valuation is categorically inadmissible in determining fair market value; and second, a factual argument that the post-valuation-date data used by the court in this case distorted its method.
Neither argument has merit. No authority supports the contention that land appraisals cannot take account of data released after the date of valuation. Trout Ranch relies on Treas. Reg. § 1.170A-14(h)(3)(i), which provides, “[t]he value of the contribution under section 170 in the case of a charitable contribution of a perpetual conservation restriction is the fair market value of the perpetual conservation restriction at the time of the contribution.” But a regulation fixing the fair market value at the time of contribution does not necessarily restrict the evidence to be considered in determining fair market value. Section 1.170A-14(h)(3)(i) is neutral on the question of relevant evidence; it sets no limitations on the information informing fair market value.
Treasury Regulation § 1.170A-l(c)(2) is likewise unhelpful: “The fair market value is the price at which the property would change hands between a willing buyer and
a willing seller, neither being under any compulsion to buy or to sell and both having reasonable knowledge of relevant facts.” Stressing the last clause about “reasonable knowledge of relevant facts,” Trout Ranch questions how a reasonable buyer at the time of contribution could be expected to account for land sales that had yet to occur. But Trout Ranch reads the “reasonable buyer” language too literally. The reasonable buyer is a conceptual device meant to illustrate the objective nature of the inquiry, not a limitation on the evidence that can inform it. The question for the real estate appraiser and ultimately the tax court is not what a reasonable buyer could say about a particular sale but what the particular sale could say about the reasonable buyer. Data unavailable to a reasonable buyer may still tell us something about the price such a hypothetical buyer would have been willing to pay at the time of the contribution.
Fair market value is at bottom a question of fact, and questions of fact are governed by the rule of relevance.
Polack v. Comm’r of Internal Revenue,
366 F.3d 608, 612 (8th Cir.2004);
Gilford v. Comm’r of Internal Revenue,
88 T.C. 38, 50 (1987). While evidence of subsequent sales may not always be probative of a prior fair market value, whether such evidence should factor into an appraisal is not a categorical question of law but a simple question of relevance: Does unfair prejudice substantially outweigh the probative value of the evidence? As the Seventh Circuit has explained, the question to be asked in a valuation case “is whether the admission of the evidence would make more or less probable the proposition that the property had a certain fair market value on a given date?”
First Nat’l Bank of Kenosha v. United States,
763 F.2d 891, 894 (7th Cir.1985). To this end, data from events after the donation can be just as informative (and just as distorting) as data from events before the donation. The challenge for the trier of fact is determining whether the data makes the asserted fair market value more or less likely. If there is a hard-and-fast rule governing the admission of subsequent events, it is that such events will be considered to the extent they were reasonably foreseeable on the date of donation valuation.
See Saltzman v. Comm’r,
131 F.3d 87, 93 (2d Cir.1997).
The Supreme Court’s decision in
Ithaca Trust Co. v. United States
is not to the contrary. 279 U.S. 151, 49 S.Ct. 291, 73 L.Ed. 647 (1929). That case addresses whether post-valuation data can be considered in determining the taxable value of an estate. The testator had left his estate to his wife for life with authority to draw from the principal enough to keep her comfortable, the remainder to transfer in trust to a group of charities upon the wife’s death.
Id.
at 154, 49 S.Ct. 291. The wife died shortly thereafter, within the one-year window for filing a return reflecting the charitable deduction, the value of which had originally been diminished by the amount of money the wife, based on actuarial data, was expected to draw from the principal. This raised a question: Should the value of the charitable deduction be determined based on what the charities were expecting to receive given the wife’s life expectancy at the time of the testator’s death (a smaller amount), or based on what the charities
actually
received in the wake of the wife’s untimely death (the larger amount)? The Supreme Court decided that the charitable contribution should be valued at the time of the testator’s death, when the estate is settled. “The tax is on the act of the testator,” the Court explained, “not on the receipt of
property by the legatees.”
Id.
at 155, 49 S.Ct. 291. The Court continued: “Tempting as it is to correct uncertain probabilities by the now certain fact, we are of opinion that it cannot be done, but that the value of the wife’s life interest must be estimated by the mortality tables.”
Id.
Citing
Ithaca Trust
for the proposition that subsequent events are
never
relevant in valuation cases stretches the decision beyond elastic limits. The dilemma in
Ithaca Trust
was whether to base the valuation on an estimate made at the testator’s death or on the actual value discovered later on. In other words, the court decided whether fair market value was the appropriate metric even where actual valuable was knowable. This was a policy question whose answer hinged on the purpose of the underlying tax. The same question does not arise in real estate appraisal, because everyone agrees the relevant figure is the estimated value of the land at the time of the taxable event. When we speak of subsequent events in this context, we speak of data that could
inform,
fair market value,
not data that could replace it.
This court’s decision in
Estate of McMorris v. Comm’r
is distinguishable for the same reason. 243 F.3d 1254 (10th Cir.2001). Directly applying
Ithaca Trust, McMorris
holds that events occurring after the decedent’s death may not be considered in determining the value of a claim against the decedent’s estate.
Id.
at 1261. For the same reasons we have addressed, the holding should not be understood to extend beyond the estate-tax setting.
Trout Ranch argues that, even if post-valuation evidence is not categorically inadmissible, such evidence should have been excluded in this case because the post-valuation data cited by the tax court inaccurately inflated the appraisal. Trout Ranch contends the purchase of a nearby ski resort in Crested Butte increased property values in Gunnison County at a rate that could not have been foreseen when the easement was donated, with real property in the county appreciating by more than 50 percent between 2004 and 2006.
Before turning to the data, it is important to note that the tax court was mindful of the risks associated with post-valuation data. It devoted an entire section of its opinion to addressing Trout Ranch’s concerns about the effect post-contribution lot sales might have had on the Commissioner’s valuation. Wary of placing too much weight on such sales, the court resolved to “give the most weight to lot sales within a year of the date of valuation (i.e., sales in 2003 and 2004) and less weight to sales outside that range.” (Doc. 42 at 30.) The court explained that whether evidence relating to subsequent lot sales is admissible in determining fair market value is a question of relevance. “We find that the evidence of lot sales within a reasonable period after the date of valuation (especially those at Gunnison Riverbanks Ranch itself) tends to make a given estimate of the lot prices more or less likely; that is, such evidence is relevant.” Doc. 42 at 29 (footnote omitted).
With respect to Trout Ranch’s contention about real property prices increasing at an unforeseeable clip following the donation of the easement, the tax court accepted the Commissioner’s argument that most of the appreciation was driven by development in and around Crested Butte (a premier ski resort), which the court treated as a separate economic zone within Gunnison County. Relying on a report from an appraiser at the Gunnison County Assessor’s Office, it found the economic
area encompassing Gunnison Riverbanks Ranch experienced only modest upward pressure. “We find no evidence,” the court concluded, “that the lots at Gunnison Riverbanks Ranch appreciated at more than a reasonable rate after the date of valuation.”
Id.
at 30. Moreover, although the sale of the ski resort at Crested Butte closed after Trout Ranch donated the easement, the court observed the sale was
announced
prior to the donation, making any related appreciation in land value foreseeable.
In summary, the tax court properly exercised its discretion by incorporating post-valuation data into the income analysis. Having addressed Trout Ranch’s concerns about unforeseen appreciation, the court could focus on the lot sales from Hidden River Ranch and Gunnison Riverbanks Ranch it deemed most indicative of the price of a lot at Gunnison Riverbanks Ranch on the date of valuation. Excluding post-valuation data, it seems, would have left the court unequipped to make an informed finding as to value. While reasonable minds could differ about the weight to be assigned to such data, it would be a stretch to argue the use of post-valuation data in this case amounted to an abuse of discretion.
C.
Whether the charitable deduction should be subject to the limitations in I.R.C. § 170(b)(1)(B)
When the Commissioner sent Trout Ranch the notice disallowing a charitable deduction beyond $485,000, he also asserted the deduction of the charitable contribution would be subject to the 30 percent limitation in § 170(b)(l)(B)(i), rather than the 50 percent limitation in 170(b)(1)(A).
Upon receiving notice from the Commissioner, Trout Ranch had 90 days to file a petition in the appropriate court contesting the adjustment and identifying any errors it thought the Commissioner committed.
See
Tax Court Rule 241(d)(1)(C) (“Any issues not raised in the assignments of error ... shall be deemed to be conceded.”).
Id.
Trout Ranch’s petition for readjustment did not include an objection to the Commissioner’s application of § 170(b)(1)(B), and the tax court properly concluded the matter had been conceded.
Trout Ranch argues the issue should never have been addressed by the tax court because it is a “non-partnership” item more appropriately resolved on the partners’ individual tax returns. We disagree. The Code considers partnership items to be those best determined at the partnership level, and the pertinent regulations provide that partnership items include “the legal and factual determinations that underlie the determination of the amount, timing, and characterization of items of income, credit, gain, loss, deduction, etc.” Treas. Reg. § 301.6231(a)(3)-1(b). I.R.C. § 6231(a)(3). As the tax court explained, the question of which limitation covers the charitable contribution depends simply on the type of organization to which the easement was donated, which is best determined at the partnership level.
D.
Whether the tax court erred in failing to shift the burden of proof to the Commissioner
Generally speaking, the Commissioner’s determinations of deficiency are presumed correct and the taxpayer bears the burden of proof. Tax Court Rule 142(a). However, in certain situations where the taxpayer introduces credible evidence relevant to the proper tax liability, the law shifts the burden to the Commissioner.
See
I.R.C. § 7491(a). Trout Ranch contends this is such a situation and the tax court erred in failing to shift the burden to the Commissioner. Not so. The tax court’s opinion was based on its take on the preponderance of the evidence and was announced as such. Had the evidence been evenly balanced, a situation not presented here, the issue would necessarily have been resolved against the party bearing the burden of persuasion. But the tax court did not consider it necessary to hold either party to its burden. Accordingly, any technical error related to the burden of proof was harmless.
See Keating v. CIR,
544 F.3d 900, 906 (8th Cir. 2008).
AFFIRMED.