FILED
                                                                  United States Court of Appeals
                                       PUBLISH                            Tenth Circuit

                    UNITED STATES COURT OF APPEALS                        July 29, 2020
                                                                     Christopher M. Wolpert
                          FOR THE TENTH CIRCUIT                          Clerk of Court
                      _________________________________

 ALEXIS STOKES, individually and as
 guardian and next friend of Baby Boy
 D.S., a minor; TAYLOR STOKES,
 individually and as guardian and next
 friend of Baby Boy D.S., a minor,

       Plaintiffs - Appellees/Cross-
       Appellants,

 v.                                                 Nos. 19-7034 & 19-7035

 UNITED STATES OF AMERICA, ex
 rel. Indian Health Service & Chickasaw
 Nation Medical Center, an agency of the
 Chickasaw Nation,

       Defendant - Appellant/Cross-
       Appellee.
                     _________________________________

                   Appeal from the United States District Court
                      for the Eastern District of Oklahoma
                          (D.C. No. 6:17-CV-00186-JH)
                     _________________________________

Casen B. Ross, Attorney, Appellate Staff, United States Department of Justice, Civil
Division, Washington, D.C. (Robert P. Charrow, General Counsel, Brian Stimson,
Principal Deputy General Counsel for Litigation, United States Department of Health
and Human Services; Joseph H. Hunt, Assistant Attorney General, Brian J. Kuester,
United States Attorney, Susan Stidham Brandon, Assistant United States Attorney,
and Abby C. Wright, Attorney, Appellate Staff, United States Department of Justice,
Civil Division, Washington, D.C., with him on the briefs), for Defendant-Appellant.

George W. Braly of Braly, Braly, Speed & Morris, PLLC, Ada, Oklahoma (William
W. Speed and Sheila Southard of Braly, Braly, Speed & Morris, PLLC, Ada,
Oklahoma, and Lawrence R. Murphy Jr. of Smolen Law, Tulsa, Oklahoma, with him
on the briefs), for Plaintiffs-Appellees.
                         _________________________________

Before BRISCOE, McHUGH, and MORITZ, Circuit Judges.
                 _________________________________

MORITZ, Circuit Judge.
                    _________________________________

      The district court awarded damages to Baby Boy D.S. (Baby Stokes) and his

parents, Alexis Stokes and Taylor Stokes, (collectively, the Stokes) in this Federal

Tort Claims Act (FTCA), 28 U.S.C. § 2674, action. The government appeals, arguing

that the district court erred in structuring damage payments. The Stokes cross appeal,

arguing that the district court erred both by miscalculating the present value of a

portion of the award and by awarding too little in noneconomic damages. For the

reasons explained below, we affirm in part, vacate in part, and remand.

                                      Background

      An employee of a federally supported health center failed to properly

administer a drug to Alexis Stokes while she gave birth to Baby Stokes. As a result,

Baby Stokes suffers from “cerebral palsy and spastic quadriplegia,” along with other

disabilities, and his life expectancy is 22 years. App. vol. 2, 79.

      The Stokes brought this FTCA case against the government. After a bench

trial, the district court found the government liable and ordered it to pay a total of

$15.9 million in damages, including the cost of Baby Stokes’s future care and

noneconomic damages. As relevant to this appeal, the district court (1) ordered the

government to pay the cost of Baby Stokes’s future care into a trust, granting the

                                            2
government a diminishing reversionary interest in the trust and permitting the trustee

to withdraw funds as needed to provide for Baby Stokes; (2) applied a zero-percent

discount rate in calculating the present value of the award for Baby Stokes’s future

care; and (3) awarded Baby Stokes $1,000,000 in noneconomic damages, Alexis

Stokes $500,000 in noneconomic damages, and Taylor Stokes $400,000 in

noneconomic damages. 1 The government and the Stokes both appeal.

                                        Analysis

      On appeal, the government does not challenge liability or the amount of

damages. It appeals only how the trust is structured with respect to the future-care

award. In their cross-appeal, the Stokes argue that (1) the district court applied the

wrong discount rate when calculating the present value of the future-care award and

(2) their noneconomic damages are erroneously low. 2

I.    Structure of the Trust

      The government argues that the district court erroneously structured Baby

Stokes’s future-care award by not approximating Oklahoma’s periodic-payment

statute to the fullest extent possible. The FTCA generally requires courts to hold the


      1
         The district court initially awarded each of the three Stokes $350,000 in
noneconomic damages because of a state-law statutory cap. See Okla. Stat. tit. 23,
§ 61.2(B). But shortly thereafter, the Oklahoma Supreme Court found that statutory
cap unconstitutional. Beason v. I. E. Miller Servs., Inc., 

, 1109 (Okla.
2019). At the Stokes’s request, the district court then increased the noneconomic-
damages award.
       2
         Citing the limited nature of their appeals, the parties filed an agreed
stipulation for partial summary disposition, requesting that we order that the
government pay the portion of the damages awarded in the judgment that are
uncontested on appeal. We granted that request.
                                            3
government liable for tort claims “in the same manner and to the same extent as a

private individual under like circumstances,” which includes applying relevant state

law. § 2674; see Hill v. United States, 

, 120–21 (10th Cir. 1996). In

Oklahoma, private individuals ordered to pay more than $100,000 in future-care

damages can request that they pay those damages through periodic payments instead

of as a lump sum. Okla. Stat. tit. 23, § 9.3(C). 3 Those payments may not continue for

more than seven years.

dies before all periodic payments are made,

the statute explains that the “obligation of the defendant [payor] to make further

payments ends.” § 9.3(H). But courts cannot neatly apply this Oklahoma law to the


      3
          As relevant here, § 9.3 provides:
      C. Upon request of a party, the court may order that future damages be paid
      in whole or in part in periodic payments rather than by a lump-sum
      payment. Periodic payments shall not exceed seven (7) years from the date
      of entry of judgment.
      D. The court shall make a specific finding of the dollar amount of periodic
      payments that will compensate the plaintiff for the future damages. The
      court shall specify in its judgment ordering the payment of future damages
      by periodic payments the:
      1. Recipient of the payments;
      2. Dollar amount of the payments;
      3. Interval between payments; and
      4. Number of payments or the period of time over which payments must be
      made.
      ...
      H. On the death of the recipient, money damages awarded for loss of future
      earnings shall continue to be paid to the estate of the recipient of the award
      without reduction. Following the satisfaction or termination of any
      obligations specified in the judgment for periodic payments, any obligation
      of the defendant health care provider to make further payments ends and
      any security given reverts to the defendant.

                                              4
federal government because they may not order the government to make periodic

payments. Hull ex. rel. Hull v. United States, 

, 1505 (10th Cir. 1992).

When facing a situation like this, courts must “approximate the result contemplated

by the” state statute. 

. And one common way to approximate

periodic-payment statues is for the government to pay future damages into an account

as a lump sum and then model disbursements from that account around the state’s

periodic-payment statute. See, e.g., Dixon v. United States, 

, 1260–61

(11th Cir. 2018) (noting that district court “granted the government’s request to make

a single payment into a trust for periodic disbursement” to approximate Florida’s

periodic-payment statute); Lee v. United States, 

, 527 (5th Cir. 2014)

(vacating district court order because it “should have structured the [FTCA] damage

award in a manner resembling” Texas’s periodic-payment statute); Dutra v. United

States, 

, 1092 (9th Cir. 2007) (explaining that “FTCA authorizes courts

to craft remedies that approximate the results contemplated by state statutes, and

nothing in the FTCA prevents district courts from ordering the United States to

provide periodic payments in the form of a reversionary trust”; requiring that FTCA

award approximate Washington’s periodic-payment statute); 

(ordering district court to create reversionary trust for FTCA award to approximate

Colorado’s periodic-payment statute).

      Here, after trial and in anticipation of a damages award, the Stokes created a

trust that (1) permitted the trustee to withdraw funds to care for Baby Stokes as

needed and (2) granted the government a 14-year fractional reversionary interest in

                                           5
the future-care award. 4 When the district court awarded damages, it approved this

trust structure with one exception: citing § 9.3, it ordered that the trust be modified to

ensure that “at the end of seven years no amount is payable to the government.” App.

vol. 2, 91.

       Citing the requirement that it be treated the same as a private individual, the

government asked the court to amend its order to more closely model the trust on

Oklahoma law. See § 2674. Specifically, the government requested that the district

court (1) order it to pay a lump sum for the amount of Baby Stokes’s future-care

award into an account, (2) order the trustee of that account to pay one-seventh of the

lump sum to Baby Stokes each year for seven years, and (3) order that in the event

Baby Stokes were to die before all payments were made, the remaining funds from

the lump sum would revert to the government. The district court declined to modify

the trust structure, stating that it “fashioned a remedy [that] approximated the result

contemplated by state law but complied with federal law. Strict adherence to state

law is not mandated.” App. vol. 2, 153.

       On appeal, the government argues that the district court erred by not

“approximat[ing] the result contemplated by” § 9.3 as required by the FTCA. 

; see also § 2674. The parties agree that the district court failed to



       4
         Specifically, the trust provided that, in the event of Baby Stokes’s premature
death, the government would recover a portion of the future-care award. The trust
agreement reduced the amount the government would recover by one-seventh every
two years such that after fourteen years, the government would not recover any of the
future-care award if Baby Stokes were to die prematurely.
                                            6
approximate all of § 9.3’s provisions. 5 But they disagree on whether the district court

was required to do so. Because we must hold the government liable for tort claims

“in the same manner and to the same extent as a private individual under like

circumstances,” § 2674, we first consider how the statute would apply to a private

party. Next, we determine whether the district court fully approximated that result for

the government. In doing so, we interpret § 9.3 de novo. 6 See Burlington N. & Santa

Fe Ry. Co. v. Grant, 

, 1024 (10th Cir. 2007); 

(reviewing de novo whether the district court erred in applying Florida’s periodic-

payment statute); Vanhoy v. United States, 

, 451 (5th Cir. 2008) (noting

that “[t]he question whether [Louisiana’s Medical Malpractice Act] requires the

district court to provide protection to the government in the form of a reversionary



       5
         The government asserts that the district court failed to approximate
(1) § 9.3(C) and (D) when it permitted the trustee to withdraw funds from Baby
Stokes’s future-care award on an as-needed basis, as opposed to structuring
withdrawals as set periodic payments, and (2) § 9.3(H) when it granted the
government a diminishing fractional reversionary interest, as opposed to granting the
government a full reversionary interest in the amount of the future-care award that a
private party would not have yet paid if Baby Stokes were to die prematurely. The
Stokes do not argue that that district court fully approximated § 9.3(C), (D), or (H).
       6
         The Stokes argue that we should review the trust’s structure for abuse of
discretion. It is true that in Hull we reviewed the district “court’s decision as to the
structure of the [FTCA] award . . . only for an abuse of discretion.” 

. But there was no state statute at issue in Hull, so our review did not involve
statutory interpretation. Moreover, in Dixon, which the Stokes rely on for this point,
the Eleventh Circuit interpreted Florida’s periodic-payment statute de novo. 

. Only after determining—under de novo review—that Florida’s
statute “leaves the trial court free to exercise its discretion in fashioning the periodic-
payment schedule” did the Dixon court review the periodic-payment schedule for
abuse of discretion.

Thus, Dixon does not support the Stokes’s argument
and instead provides further support for reviewing this issue de novo.
                                             7
trust” is one of legal interpretation; reviewing “district court’s ruling de novo”). And

because we are interpreting an Oklahoma statute, we use Oklahoma’s “rules of

statutory construction.” Ward v. Utah, 

, 1248 (10th Cir. 2005). So we

will not look beyond the statute’s text if “the language of the statute is plain and

unambiguous.” Antini v. Antini, 

, 60 (Okla. 2019).

      A.     Application of § 9.3 to a Private Party

      Turning to the text of § 9.3, the statute provides that “[u]pon request of a

party, the court may order that future damages be paid . . . in periodic payments.”

§ 9.3(C) (emphasis added). As the Stokes argue, because § 9.3 permits a private party

to request periodic payments, it gives the district court discretion whether to grant

that request. See Okla. Pub. Emps. Ass’n v. State ex rel. Okla. Office of Pers. Mgmt.,

, 845 n.18 (Okla. 2011) (“The term ‘may’ is ordinarily construed as

permissive . . . .”). Focusing on the word “may,” the Stokes then argue that none of

§ 9.3 is mandatory because it “uses the word ‘may’ rather than ‘shall,’ thus defining

the discretionary nature of implementing (or not) the statute.” Aplee. Br. 16 (quoting

§ 9.3). But the Stokes’s argument ignores the remainder of the statute, which

repeatedly uses the term “shall” in describing the court’s responsibilities. For

example, it states that “[p]eriodic payments shall not exceed [seven] years” and that

the “court shall specify . . . [the r]ecipient of the payments,” the “[d]ollar amount of

the payments,” the “[i]nterval between payments,” and the “[n]umber of payments or

the period of time over which payments must be made.” § 9.3(C)–(D) (emphases

added). The statute’s use of “shall” in these provisions indicates that once a court

                                            8
grants a private party’s request to apply the statute, it must apply these provisions

when structuring payments. See Keating v. Edmondson, 

, 888 (Okla.

2001) (explaining that “shall” “signifies a mandatory directive or command”). As the

government notes, under § 9.3, “a private party would never be required to make

periodic payments for future life-care expenses on anything other than a fixed

schedule . . . . Rather, a private party would either make a lump-sum payment or be

ordered to make periodic payments on a specific schedule set by the court.” Aplt. Br.

25.

      We agree and conclude that, as applied to a private party, § 9.3 permits the

district court discretion to grant or deny a party’s request to apply the statute. But,

once a court grants that request, the statute’s provisions become mandatory.

      B. Approximation of the Result of § 9.3

      Having considered and determined how § 9.3(C) would apply to a private

party, we must now decide whether the district court fully “approximate[d] the result

contemplated by” § 9.3 as required by the FTCA. 

; see also

§ 2674. As such, we must first determine whether the district court granted the

government’s request to apply the statute. If it did, we must then determine whether

the district court fully approximated § 9.3.

       The Stokes argue that the district court declined to apply § 9.3 because it

“explicitly declined to order periodic payments.” Aplee. Br. 23. Specifically, the

Stokes point to the district court’s order denying the “establishment of a specific

payment schedule.” App. vol. 2, 154. But, in its initial order awarding damages—and

                                            9
in response to the government’s request to structure the trust according to § 9.3—the

district court noted that it structured the trust “based on Oklahoma’s statute” and, in

doing so, cited to § 9.3. App. vol. 2, 90. Moreover, a later order modifying the

damages award stated that “the court fashioned a remedy [that] approximated the

result contemplated by state law but complied with federal law. Strict adherence to

state law is not mandated.” App. vol. 2, 153. Thus, the district court clearly did not

reject application of § 9.3 in the first instance. Rather, the district court’s deviation

from § 9.3 appears to stem from a failure to understand its duty to fully approximate

the result contemplated by the statute, as required by the FTCA. Accordingly, we

conclude that the district court granted the government’s request to apply § 9.3.

Because, as explained above, this decision renders the remainder of § 9.3 mandatory

as applied to a private party, we now must determine whether the district court

approximated that result for the government. 

.

       Here, the district court approved a trust that (1) permitted the trustee to

withdraw funds to care for Baby Stokes as needed and (2) granted the government a

seven-year fractional reversionary interest in the future-care award. The government

argues that this structure did not adequately approximate § 9.3. As described above,

courts often approximate periodic-payment statues by ordering the government to pay

future damages into an account as a lump sum and then modeling disbursements from

that account around the state’s periodic-payment statute. See, e.g., 

(ordering district court to create reversionary trust for FTCA award to

approximate Colorado’s periodic-payment statute). And, citing cases where other

                                            10
courts approximated periodic-payment statutes, the government asserts that the

district court should have approximated § 9.3 by doing just that. See, e.g., Askew v.

United States, 

, 1093 (8th Cir. 2015) (“[T]o best approximate the

results contemplated by the Missouri statutes, the district court should have specified

[the tort victim’s] future medical damages, created a reversionary trust to hold those

funds, and ordered periodic payments of future medical damages from the trust, with

the corpus of the trust to revert to the United States upon [victim’s] death.”); 

(“Although the district court could not impose a continuing

obligation on the government, it should have structured the damage award in a

manner resembling the periodic[-]payment scheme.”). Thus, the government

concludes, the district court did not “approximate the result contemplated by” § 9.3

as required by the FTCA. 

; see also § 2674.

       The Stokes argue that the district court was not required to approximate § 9.3

by creating a reversionary trust. In doing so, they first assert that the discretionary

nature of § 9.3 makes it “fundamentally different” from the statutes at issue in the

FTCA cases the government cites. Aplee. Br. 18. In those cases, courts generally

either approved of or ordered the district court to create trust structures similar to

what the government requests here. See, e.g., 

; 

. But, as the Stokes point out, those cases involved periodic-payment statutes

that require a court to order periodic payments once a party has requested them,

whereas § 9.3 allows a court to deny a party’s request. Compare, e.g., 

–21 (applying Colorado’s periodic-payment statute, Colo. Rev. Stat. § 13-64-203,

                                            11
which mandates periodic payments in certain circumstances), with § 9.3(C)

(explaining that “the court may order that future damages be paid . . . in periodic

payments”). Contrary to the Stokes’s argument, however, this distinction does not

make § 9.3 “fundamentally different” from the statutes at issue in the government’s

cases. Aplee. Br. 18. As explained above, § 9.3’s provisions became mandatory once

the district court granted the government’s request to apply the statute. And because

§ 9.3’s provisions are mandatory here, the government’s cases are both relevant and

support its contention that the district court should have more fully approximated

§ 9.3.

         The Stokes next explain that the government is not actually requesting

application of all of § 9.3 because the government does not argue that the district

court should have better approximated § 9.3(J), a provision which requires periodic

payments to include postjudgment interest. And therefore, they maintain, the

government’s argument that the district court failed to fully approximate § 9.3 is

internally inconsistent. But the Stokes do not explain how the district court could

have better approximated § 9.3(J). 7 And even assuming that the government’s

argument is internally inconsistent, the Stokes do not demonstrate how such

inconsistency impacts how a court should approximate § 9.3. Thus, the Stokes’s

§ 9.3(J) argument does not alter our analysis.


         7
         In this section of their brief, the Stokes assert that the government’s proposed
trust structure would deprive them of earning postjudgment interest, but nothing in
§ 9.3 limits the Stokes’s ability to invest the lump-sum payment, and the government
never suggests otherwise.
                                           12
      Finally, the Stokes argue that Hull and Hill require courts to find that any

reversionary interest be in the best interests of the tort victim and that granting the

government a full reversionary interest here is not in the best interest of Baby Stokes.

Therefore, the Stokes conclude, the district court did not err by failing to fully

approximate § 9.3. But the FTCA says nothing about the best interests of the victim,

and Hull and Hill do not suggest otherwise. True, Hull acknowledged that district

courts may structure FTCA awards according to the best interests of the tort victim.

. But, as discussed above, see supra note 6, Hull did not

involve a state statute. Thus, in Hull, the fact that future-care awards may be

structured in the victim’s best interest was not displaced by state law, like it is here.

       Hill similarly does not indicate that the victim’s best interests supersedes the

FTCA’s requirement to approximate state statutes. True, in Hill, we considered the

victim’s best interests in declining to exercise our inherent authority to order the

district court to grant the government a reversionary interest in the victim’s future-

earnings award. 

. But the state periodic-payment statute at issue

in Hill did not permit reversion of future-earnings awards, so there was no conflict

between our decision and the FTCA’s requirement to approximate state statutes.

Hill does not give courts permission to deviate from the FTCA’s

approximation requirement, even if doing so is in the best interests of the victim.

Accordingly, we decline to deviate from this requirement here.

       Based on § 9.3’s plain language, we conclude that if a court applies § 9.3 to a

private party then it must apply all of § 9.3’s requirements. See 

60. And that conclusion, in this case, means the court must approximate that result

for the government. 

. But, as noted above, the district court did

not fully approximate § 9.3. Instead, although it granted the government’s request to

apply the statute, it did not establish a specific payment schedule, as contemplated by

§ 9.3(C) and (D). Nor did it provide the government, in the event of Baby Stokes’

premature death, a full reversionary interest in any portion of the future-care award

that a private party making periodic payments would not yet have made, as

contemplated by § 9.3(H). Thus, the district court erred by not approximating the

result contemplated by the statute. See 

.

                                         ***

      We therefore vacate the portion of the district court’s order approving the

trust’s structure with respect to the future-care award and remand with instructions to

approximate § 9.3 to the fullest extent possible. Such approximation should specify

the “[r]ecipient of the payments,” the “[d]ollar amount of the payments,” the

“[i]nterval between payments,” and the “[n]umber of payments or the period of time

over which payments must be made.” § 9.3(D). In structuring these payments, the

district court should make “a specific finding of the dollar amount of periodic

payments that will compensate” Baby Stokes for his future care.

structure should provide that in the event of Baby Stokes’s premature death, the

government will receive a full reversionary interest in any portion of the future-care

award that a private party making periodic payments would not yet have made, as

contemplated by § 9.3(H).

                                          14
II.    Discount Rate

       In their cross appeal, the Stokes argue that the district court erred when it

calculated the present value of Baby Stokes’s future-care award. When a court

awards damages to compensate for losses that will be suffered in the future, the court

must reduce the award to present value “to account for the effects of investment” and

“the effects of inflation.” 

. The parties agree that the proper

way to calculate present value is to apply a discount rate, which is the interest rate

minus the inflation rate. But the Stokes argue that the district court (1) committed

legal error by determining that the discount-rate rule from Jones & Laughlin Steel

Corp. v. Pfeifer, 

(1983), did not apply to FTCA cases and

(2) committed factual error by relying on an interest rate that is not currently

available.

       Here, the district court applied a zero-percent discount rate. In doing so, it did

not state which interest rate or inflation rate it used, nor did it clearly state the rule it

used for determining a discount rate. Instead, the district court stated that it “t[ook]

into account the current inflation rate of medical costs, current interest rates, and the

disparity between the experts’ opinions,” and then it concluded “that a net discount

rate of [zero] percent is appropriate.” App. vol. 2, 88. But the district court did

explicitly state that Pfeifer does not apply because that case “was not an action

brought under the FTCA.”

n.38.

       On appeal, the Stokes first argue that the district court erred by determining

that Pfeifer does not apply to FTCA cases. Because this is a legal issue, we review it

                                             15
de novo. 8 See El Encanto, 

(noting that courts “decide the

presence or absence of legal error de novo”).

      In Pfeifer, the Supreme Court stated that, when calculating present value of

future awards, the discount rate should reflect the rate of interest that would be

“earned on ‘the best and safest investments’” that are “available” and should also

“represent the after-tax rate of return.” 

–38 (quoting

Chesapeake & Ohio Ry. Co. v. Kelly, 

, 491 (1916)). As noted above, the

district court concluded that Pfeifer did not apply here because Pfeifer was not an

FTCA case. But, as the Stokes explain, we have previously applied Pfeifer in an

FTCA case. 9 See 

(citing Pfeifer when determining discount

rate in FTCA case); cf. Trevino v. United States, 

, 1519 (9th Cir. 1986)

(noting that Pfeifer’s “discussion of discount rates technically is only an

interpretation of . . . the Longshoremen’s and Harbor Workers’ Compensation Act”

and thus “is not binding in an FTCA case” but that “the Court’s guidance on the issue

of economic predictions and discount rates [in Pfeifer] cannot be disregarded”). And


      8
          The government argues for a clearly erroneous standard, asserting that the
appropriate interest rate is a question of fact. See Hull, 

(“We review the
district court’s choice of a discount rate under a clearly erroneous standard.”). But
this argument assumes we are reviewing the district court’s choice of a zero-percent
discount rate for factual error. Instead, we are reviewing the district court’s method of
selecting a discount rate for “the presence or absence of legal error” and thus apply de
novo review. El Encanto, Inc, v. Hatch Chile Co., Inc., 

, 1162 (10th
Cir. 2016).
        9
          The district court also implied that Pfeifer requires courts to use the
“Treasury Bill rate” when calculating interest rates. App. vol. 2, 88 n.38. But Pfeifer
makes no mention of which investment vehicle is appropriate, as long as that vehicle
is a safe investment. See 

–38.
                                           16
the government does not argue that Pfeifer is inapplicable here—indeed, it cites

favorably to Pfeifer in its briefing on appeal. Given our reliance on Pfeifer in Hull,

we conclude that Pfeifer applies to FTCA cases and that the district court committed

legal error in determining otherwise.

       We thus vacate the portion of the district court’s order calculating the present

value of Baby Stokes’s future-care award 10 and remand with instructions to apply

Pfeifer in determining the discount rate. But we express no opinion on whether a

zero-percent discount rate itself is incorrect. Although the Stokes argue that, contrary

to Pfeifer, this zero-percent discount rate is based on an interest rate that is not

currently available, the district court did not state which interest or inflation rates it

used. Because the district court did not state the basis for its chosen discount rate, we

cannot determine whether this rate is an error. See 

(“[W]e

remand for the district court to make appropriate findings of fact explaining its

method of calculating present value for all awards for future damages . . . .”).

III.   Noneconomic Damages

       The Stokes argue that the district court’s noneconomic-damages award was too

low. As noted above, the district court ultimately awarded Baby Stokes $1,000,000 in

noneconomic damages, Alexis Stokes $500,000 in noneconomic damages, and Taylor

Stokes $400,000 in noneconomic damages. We “review the award of noneconomic


       10
         In addition to the future-care award, the district court reduced both Baby
Stokes’s and Alexis Stokes’s lost-future-income awards to present value. But neither
party contests these other calculations on appeal, and thus we do not vacate the
portion of the order reducing these awards to present value.
                                             17
damages for clear error, to determine whether ‘the award shocks the judicial

conscience.’” Deasy v. United States, 

, 360 (10th Cir. 1996) (quoting

Miller v. United States ex rel. Dep’t of the Army, 

, 897 (10th Cir.1990)).

This standard is difficult to meet as “[t]rial courts are vested with broad discretion in

awarding damages, and [we] do not lightly engage in a review of a trial court’s

actions.” Dolenz v. United States, 

, 1321 (10th Cir. 2006) (quoting

Hoskie v. United States, 

, 1354 (10th Cir. 1981)).

       The Stokes argue that the amounts of their noneconomic damages are so low

that they shock the judicial conscience because (1) they are substantially lower than

awards in factually similar cases and (2) Baby Stokes’s injuries are so severe. The

Stokes first identify other cases that they assert are factually similar, noting that the

noneconomic damages awarded to children in those similar cases average about $9.3

million and range from about $7.6 million to $11 million. In response, the

government asserts that it is inappropriate under both Tenth Circuit and Oklahoma

law to compare awards in factually similar cases when determining whether the

award shocks the conscience. But caselaw from both our circuit and the Oklahoma

Supreme Court suggest otherwise. See Hill v. J.B. Hunt Transp., Inc., 

,

670–71 (10th Cir. 2016) (noting that although “[b]oth this court and Oklahoma courts

discourage comparisons to awards from other cases, . . . [w]e have made exceptions

where a previous case is similar enough to serve as a meaningful benchmark”); Okla.

Ry. Co. v. Strong, 

, 952 (Okla. 1951) (noting that when reviewing



                                            18
damages, “[appellate] courts may . . . consider the amount of verdicts rendered in

similar cases”).

      However, we need not resolve this dispute here because even considering the

difference between the amounts awarded in the cases the Stokes urge us to consider

and the amounts awarded here, the Stokes’s noneconomic damage awards do not

shock the judicial conscience. This is true even taking into consideration the extent of

Baby Stokes’s injuries, as the Stokes also urge us to do. Although not necessarily an

award we would endorse on de novo review, the district court has wide discretion in

determining noneconomic damages. See 

. And the Stokes do

not meet their high burden of demonstrating that the district court erred in exercising

that discretion here. We therefore affirm the portion of the district court’s order

regarding noneconomic damages.

                                      Conclusion

      For the reasons stated above, we (1) vacate and remand the portion of the

district court’s order structuring the trust with respect to Baby Stokes’s future-care

award, with instructions to fully approximate § 9.3; (2) vacate and remand the

portion of the district court’s order calculating the present value of Baby Stokes’s

future-care award, with instructions to apply Pfeifer; and (3) affirm the portion of the

district court’s order regarding noneconomic damages.




                                           19
19-7034, 19-7035, Stokes, et al. v. United States of America, ex rel. Indian Health Service
                  & Chickasaw Nation Medical Center

McHUGH, Circuit Judge, concurring:

       I join in much of the majority’s thoughtful analysis and in its conclusion that the

district court’s judgment should be vacated and the case remanded. I write separately,

however, to explain where my analysis differs.

       First, with respect to the structure of the Stokes’s life care trust: Rather than order

the district court to modify the trust so that it complies to the extent possible with the

requirements of Oklahoma’s periodic payment statute, I would leave the district court

free to clarify on remand whether it intended to apply that statute. In my view, the record

is not clear in that regard.

       Second, with respect to the discount rate applicable to the Stokes’s damage award:

Rather than leave the district court free to again apply a zero percent discount rate, I

would find that its use of a zero percent discount rate for Baby Stokes’s life care costs

was clear error.

       I agree with the panel majority’s discussion of noneconomic damages and join that

portion of the opinion in full.

                          I.      STRUCTURE OF THE TRUST

       I agree with the panel majority that if the district court had ordered periodic

distributions from the life care trust pursuant to Okla. Stat. Ann. tit. 23, § 9.3, the district

court would then be required to approximate the result that would obtain against a private

party under all that statute’s provisions. But I cannot agree that the district court “granted
the government’s request to apply § 9.3.” Maj. Op. at 11. The district court did not order

periodic trust distributions. And the district court’s treatment of Oklahoma law in its

various orders is, at best, ambiguous. Accordingly, I would remand so that the district

court has an opportunity to clarify its intentions.

       Consider the structure of Oklahoma’s periodic payments statute. Okla. Stat. Ann.

tit. 23, § 9.3 provides that, “[u]pon request of a party, the court may order that future

damages be paid in whole or in part in periodic payments rather than by a lump-sum

payment.”

added). If a court does order periodic payments, it must

“specify in its judgment . . . the: 1. Recipient of the payments; 2. Dollar amount of the

payments; 3. Interval between payments; and 4. Number of payments or the period of

time over which payments must be made.”
 Id. § 9.3(D). But, 
in any event, “[p]eriodic

payments shall not exceed seven (7) years from the date of entry of judgment.”
 Id.

§ 9.3(C).

       Section 
9.3 also imposes several conditions to guarantee that a defendant makes

court-ordered periodic payments:

       As a condition to authorizing periodic payments of future damages, the
       court shall require a defendant who is not adequately insured to provide
       evidence of financial responsibility in an amount adequate to assure full
       payment of damages awarded by the judgment. The judgment shall
       provide for payments to be funded by:

              1. An annuity contract issued by a company licensed to do
              business as an insurance company, including an assignment
              within the meaning of Section 130, Internal Revenue Code of
              1986, as amended;

              2. An obligation of the United States;


                                              2
              3. Applicable and collectible liability insurance from one or more
              qualified insurers; or

              4. Any other satisfactory form of funding approved by the court.

       On termination of periodic payments of future damages, the court shall
       order the return of the security, or as much as remains, to the defendant.


Id. § 9.3(F)-(G).

       When 
the recipient of periodic payments dies, “damages awarded for loss of future

earnings shall continue to be paid to the estate of the recipient of the award without

reduction.”
 Id. § 9.3(H). By 
contrast, “[f]ollowing the satisfaction or termination of any

obligations specified in the judgment for periodic payments, any obligation of the

defendant health care provider to make further payments ends and any security given

reverts to the defendant.”
 Id.

       There are three 
reasons we should be skeptical the district court intended to adopt

all, or even most, of § 9.3. First, under our decision in Hull v. United States, 
971 F.2d

1499 
(10th Cir. 1992), a court may not order the United States to make periodic

payments. See
 id. at 1505. 
Strict compliance with § 9.3 is therefore impossible in FTCA

actions. The United States concedes as much. But that concession also defeats the United

States’ argument that the district court “opted-in” to § 9.3 by limiting the partial

reversionary interest to seven years. The district court did not “opt-in” to § 9.3; it could

not do so because it had no power to order periodic payments.

       Second, § 9.3—by its plain terms—never requires periodic payments. It states that

a court “may order that future damages be paid in whole or in part in periodic payments.”

added). Because a similarly situated Oklahoma court would not be

                                              3
required to order periodic payments in a non-FTCA case analogous to this one, the

district court was not required to do so here.

       Third, it would be anomalous for the district court to impose all the various

requirements of § 9.3 on this life care trust. For example, § 9.3(F) requires that periodic

payments be funded by one of four means, including “[a]n obligation of the United

States.”
 Id. § 9.3(F)(2). The 
United States does not argue that it should be required to

make a lump sum payment to the life care trust using Treasury bills or some other

government-backed instrument. But that is the logical endpoint of its all-encompassing

position that the FTCA requires the district court’s judgment to approximate Oklahoma

law in every possible respect.

       To be sure, the district court determined that it would order the life care trust to

make payments within seven years “based on Oklahoma’s statute.” App., Vol. II at 90. In

addition, the district court stated that it was doing so “on the basis of the statute rather

than plaintiffs’ consent.” App., Vol. II at 91 n.41. And, the district court later described

its prior order as “fashion[ing] a remedy which approximated the result contemplated by

state law.” App., Vol. II at 153. But the district court was permitted to draw inspiration

from Oklahoma’s periodic payment statute without adopting it wholesale. Indeed, how

could it be otherwise when § 9.3(C) states that periodic payments “shall not exceed seven

(7) years” but says nothing about the duration of reversionary interests.

       The panel majority sees clarity in the district court’s confusing and scattered

references to § 9.3 because it proceeds as if the district court were faced with a binary

choice between applying some of § 9.3 or none of it. See Maj. Op. at 11 (referring to the

                                                 4
district court’s “duty to fully approximate the result contemplated by the statute”). But if

§ 9.3 does not apply to this life care trust, then the district court possessed “inherent

authority” to structure the trust in an alternative way that furthered Baby Stokes’s

interests. 1 
Hull, 971 F.2d at 1505
.

       Because the district court did not order periodic trust distributions, we should not

assume that it intended to apply § 9.3. Instead, we should vacate and remand so the

district court can clarify its intentions. On remand, the district court should be free (1) to

order periodic trust distributions based on § 9.3 or (2) to explain that it elects not to

approximate the statutory scheme and to instead show how its alternative arrangement

advances Baby Stokes’s interests.

                              II.        THE DISCOUNT RATE

       I agree with the panel majority that the district court misstated the law when it

calculated the applicable discount rate, but I would go further and also hold that—on this

record—the district court committed clear error in assessing a zero percent discount rate

applicable to the damage award for Baby Stokes’s future care.

                                    A.     Legal Framework

       I agree with the panel majority that the district court was wrong to disregard Jones

& Laughlin Steel Corp. v. Pfeifer, 
462 U.S. 523 
(1983), when calculating the applicable



       1
          Unlike the majority, I do not view Oklahoma’s statute as a rejection of the “best
interests” test. Instead, I see it as the Oklahoma Legislature’s attempt to codify a one-
size-fits-all procedure to protect the best interests of judgment beneficiaries, while
allowing the court to opt out of that procedure if the best interests of the particular
beneficiary would be better served outside that framework.
                                               5
discount rate. In Jones & Laughlin, the Supreme Court discounted a lump sum damages

award “based on the rate of interest that would be earned on ‘the best and safest

investments’” because an injured plaintiff “is entitled to a risk-free stream of future

income to replace his lost wages.”
 Id. at 537 
(quoting Chesapeake & Ohio Ry. Co. v.

Kelly, 
241 U.S. 485
, 491 (1916)). The rate should therefore “not reflect the market’s

premium for investors who are willing to accept some risk of default.”
 Id.

       In Hull, we 
applied Jones & Laughlin to an FTCA damages award for injuries to a

child. “In calculating damages,” we explained,

       the trier of fact should discount awards for future damages to present
       value to account for the effects of investment, but the effects of inflation
       should also be considered, and, with regard to wages, other factors may
       need to be considered that may reasonably lead to a projection of
       increased wages over 
time.

971 F.2d at 1510
–11.

       Hull cited favorably to our pre-Jones & Laughlin decision in Hoskie v. United

States, 
666 F.2d 1353 
(10th Cir. 1981). 
See 971 F.2d at 1511
. There, we explained that

the object of discounting a damages award is to identify “an amount of money that can be

invested in a reasonably safe long-term investment available to the average person.”

Hoskie, 666 F.2d at 1355
. Applying that rule, we held it was not clear error for a district

court to adopt a “9.5 percent discount rate . . . based on the current yield of triple A-rated

corporate bonds” when calculating the net present value of an FTCA damage award for

lost future earnings.
 Id. at 1355 
(emphasis added).

       Together, these cases teach that the first step in calculating an appropriate discount

rate is to identify the best and safest investment available to the plaintiff. After the district

                                               6
court identifies such an investment, the second step is to identify the rate of return that

can be earned on that investment. The third step is to subtract the resulting rate of return

from the projected rate of inflation (whether for wages, healthcare, general expenses, or

any other relevant bundle of goods and services).

       I offer two additional observations to assist the district court on remand. First, I

disagree with the United States’ contention that municipal bonds are categorically a safe

and available investment. Appellant Resp. at 26 (citing Chesapeake & Ohio Ry. 
Co., 241

U.S. at 490
). Courts must make specific findings about the current availability of a safe

investment or safe set of investments. Cf. 
Hoskie, 666 F.2d at 1355 
(referring to “the

current yield of triple A-rated corporate bonds” (emphasis added)).

       I therefore also disagree with the plaintiffs’ contention that “the only risk free and

available investment vehicle” is a portfolio of Treasury bills. Appellee Resp. at 34. In

fact, the testimony at trial indicated otherwise. For example, on cross-examination, the

defense expert agreed that certain high-grade municipal bonds are “representative of a

highly secure investment.” App., Vol. V at 181.

       Second, I note that historical rates of return are one tool that can help a district

court estimate future returns but are not an end unto themselves. “We cannot deny

history, nor can history provide an always reliable basis for predicting the future.”

Trevino v. United States, 
804 F.2d 1512
, 1518 (9th Cir. 1986). In this field of “rough

approximation,” Jones & 
Laughlin, 462 U.S. at 546
, expert testimony on the usefulness

of historical data deserves careful consideration. E.g., App., Vol. IV at 121 (expert

testimony that it would not be appropriate to calculate a discount rate using historic

                                               7
municipal bond yields because “we don’t know if [those rates] will ever return”); see also

Trevino, 804 F.2d at 1518 
(criticizing a district court’s interest-rate calculation because

“it relied on an unrepresentative timespan”).

                                      B.     Clear Error

       “We review the district court’s choice of a discount rate under a clearly erroneous

standard.” 
Hull, 971 F.2d at 1512
. The district court committed clear error by adopting a

rate of return that the parties’ experts agreed is not currently available to the plaintiffs.

       First, some housekeeping. The panel majority states: “Because the district court

did not state the basis for its chosen discount rate, we cannot determine whether this rate

is an error.” Maj. Op. at 18. But in Hull, we stated: “Because the district court adopted

the figure proposed by [an] expert, we may conclude that the district court also adopted

the expert’s methodology.” 
Hull, 971 F.2d at 1512
.

       The district court adopted a figure within the defense expert’s range of proposed

discount rates (zero to 1.82 percent) and above the plaintiffs’ expert’s proposed discount

rate (below zero). Consequently, we may review the correctness of the defense expert’s

methodology, which employed historical average rates of return on high-grade municipal

bonds. 2




       2
        The panel majority observes, accurately, that the defense expert calculated “three
separate discount rates each applying to a different component of the future-care award.”
Maj. Op. at 19 n.13. But the defense expert calculated all three of those discount rates
using the historical rate of return on high-grade municipal bonds. That is where the error
lies.
                                                8
       On this record, the district court committed clear error by relying on the defense

expert’s methodology. The defense expert calculated a historical rate of return on high-

grade municipal bonds of 5.92 percent. At trial, plaintiffs’ counsel asked the plaintiffs’

expert whether there are “any Triple A rated municipal bonds that are presently available

with an interest rate anywhere close to 5.92 percent,” and he answered, “No. None that

have a decent rating anyway. Maybe you could buy some, what they call garbage bonds.

But that wouldn’t be satisfactory either because they’re not risk free.” App., Vol. IV at

145. Likewise, the defense expert acknowledged on cross examination that he would “be

surprised if by the end of the year [the Stokes could] find a high-grade municipal bond

yielding 5.92 percent.” App., Vol. V at 173.

       In sum, the district court committed clear error by adopting a discount rate that

relied on a rate of return not available to the plaintiffs were they to promptly invest the

life care portion of their damage award in the best and safest investment. I therefore

concur in the panel’s decision to vacate and remand so the district court may calculate a

new discount rate.




                                               9