United States v. James D. PaulsonUnited States v. James D. Paulson
FOR PUBLICATION
OPINION
UNITED STATES OF AMERICA, Plaintiff-Appellee, v. JOHN MICHAEL PAULSON, individually; and as Executor of the Estate of Allen E. Paulson; JAMES D. PAULSON, individually; and as statutory executor of the Estate of Allen E. Paulson, MADELEINE PICKENS, individually; and as statutory executor of the Estate of Allen E. Paulson; and as Trustee of the Marital Trust created under the Allen E. Paulson Living Trust; and as Trustee of the Madeleine Anne Paulson Separate Property Trust, Defendants, and VIKKI E. PAULSON, individually; and as statutory executor of the Estate of Allen E. Paulson; and as Co-Trustee of the Allen E. Paulson Living Trust; CRYSTAL CHRISTENSEN, individually; and as statutory executor of the Estate of Allen E. Paulson; and as Co-Trustee of the Allen E. Paulson Living Trust, Defendants-Appellants.
No. 21-55230
D.C. No. 3:15-cv-02057-AJB-NLS
Appeal from the United States District Court for the Southern District of California
Anthony J. Battaglia, District Judge, Presiding
Argued and Submitted February 11, 2022
San Francisco, California
Filed May 17, 2023
Before: Kim
SUMMARY*
Tax
The panel reversed the district court‘s judgment in favor of defendants, and remanded with instructions to enter judgment in favor of the government on its claims for estate taxes, and to conduct any further proceedings necessary to determine the amount of each defendant‘s liability for unpaid taxes.
The United States sued several heirs of Allen Paulson, alleging that they were trustees of Paulson‘s trust or received estate property as transferees or beneficiaries, and were thus personally liable for estate taxes under
Allen Paulson died with an estate valued at nearly $200 million, most of which was placed in a living trust. The estate was distributed among Paulson‘s heirs over the years. When the estate filed its tax return, it also paid a portion of its tax liability, and elected to pay the remaining balance in installmеnts with a fifteen-year plan under
The United States filed an action against the beneficiaries, seeking a judgment against the estate and living trust for the outstanding balance of the estate‘s tax liability. The United States also sought judgment against the individual defendants under
The panel held that
Judge Ikuta dissented. Disagreeing with the majority‘s statutory interpretation, she explained that the taxpayers’ reading of the statute is more plausible, avoids an illogical result (namely, that a person who receives estate property years after the estate is settled could be held personally liable for estate taxes that potentially exceed the current value of the property received), and is is a better indication of Congress‘s intent to impose such personal liability only on the date of the decedent‘s death.
COUNSEL
Lauren E. Hume (argued), Joan I. Oppenheimer, and Ivan C. Dale, Attorneys; David A. Hubbert, Acting Assistant Attorney General; Tax Division, United States Department of Justice; Washington, D.C.; Randy S. Grossman, Acting United States Attorney; Office of the United States Attorney; Washington, D.C.; for Plaintiff-Appellant/Cross-Appellee.
Glen A. Stankee (argued), Akerman LLP, Fort Lauderdale, Florida; Katherine E. Giddings, Akerman LLP, Tallahassee, Florida; Donald N. David, Akerman LLP, New York, New York; Joshua R. Mandell, Akerman LLP, Los Angeles, California; Lisa M. Coyle, Blank Rome LLP, New York, New York; for Defendant-Appellee Madeleine Pickens.
John C. Maloney Jr. (argued), Zuber Lawler LLP, New York, New York, for Defendants-Appellees/Cross-Appellants Vikki E. Paulson and Crystal Christensen.
James D. Paulson, Woodland Hills, California, pro se Defendant-Appellant.
OPINION
BADE, Circuit Judge:
Allen Paulson died with an estate valued at nearly $200 million, with most of his assets placed in a living trust. But years later more than $10 million in estate taxes, interest, and penalties remained unpaid. The United States of America (the United States or the government) sued several of Paulson‘s heirs—John Michael Paulson, James D. Paulson, Vikki E. Paulson, Crystal Christensen, and Madeleine Pickens—alleging that they controlled the trust, as trustees, or received estate property, as transferees or beneficiaries, and thus are personally liable for the estate taxes under
As relevant to this appeal, the district court granted in part Vikki Paulson‘s Crystal Christensen‘s, and Madeleine Pickens‘s motions to dismiss, concluding that they were not liable for the estate taxes under
We hold that
I
A
Allen Paulson died on July 19, 2000. He was survived by his third wife Madeleine
marriage—Richard Paulson, James Paulson, and John Michael Paulson—and several grandchildren, including Crystal Christensen. Richard Paulson died after his father, and Vikki Paulson is Richard Paulson‘s widow. At the time of Allen Paulson‘s death, his gross estate was valued at $193,434,344 for federal estate tax purposes. Nearly all his assets, which included real estate, stocks, bonds, cash, and receivables, were held in a living trust.2 The living trust was revocable during Allen Paulson‘s lifetime and, according to its terms, the trust was to pay any estate taxes.
When Allen Paulson died, his son John Michael Paulson became a co-trustee of the living trust and was appointed co-executor by the probate court. In October 2001, John Michael Paulson became the sole executor of the estate, with a different co-trustee. That same month, he filed an estate tax return, or Form 706, with the Internal Revenue Service (IRS). On October 23, 2001, the IRS received the estate‘s Form 706 estate tax return, which reported a total gross estate of $187,729,626, а net taxable estate of $9,234,172, and an estate tax liability of $4,459,051. The estate paid $706,296 with the return and elected to defer the remaining balance of $3,752,755 to be paid in installments with a fifteen-year plan under
2001, the IRS assessed the reported estate tax liability of $4,459,051.
The IRS audited the estate tax return and asserted a deficiency in the estate tax reported on the return, which the estate challenged in Tax Court. In December 2005, the Tax Court entered a stipulated decision and determined that the estate owed an additional $6,669,477 in estate taxes. The IRS assessed the additional liability in January 2006, and the estate elected to pay this amount through the remaining
Meanwhile, various disputes arose between Madeleine Pickens and Allen Paulson‘s other heirs. In settlement of those disputes, Madeleine Pickens received assets that the government asserts were worth approximately $19 million, including $750,000 in cash, two residences and the personal property located at those residences, and an ownership interest in the Del Mar Country Club.5 Vikki Paulson and Crystal Christensen assert that the assets Madeleine Pickens received were worth
2003, John Michael Paulson and the co-trustee transferred these assets from the living trust to Madeleine Pickens, as trustee of her personal living trust. Between 2003 and 2006, John Michael Paulson distributed at least $7,261,887 in cash from the living trust to other trust beneficiaries, including $990,125 to Crystal Christensen.6
In March 2009, the probate court removed John Michael Paulson as trustee of the living trust for misconduct and appointed Vikki Paulson and James Paulson as co-trustees. The government asserts that, at that time, the trust contained assets worth more than $13.7 million, which exceeded the estate tax liability. Vikki Paulson and Crystal Christensen claim that by this time the living trust was insolvent, with $10.8 million in assets, but $28.3 million in liabilities, including $9.6 million in federal tax liability.
In May 2010, because of the missed installment payments, the IRS terminаted the
In February 2011, the probate court appointed Crystal Christensen co-trustee of the living trust with Vikki Paulson. At that time, according to the government, the living trust
held assets worth at least $8.8 million. In June and July 2011, the IRS recorded notices of federal tax liens against the estate under
B
In September 2015, the United States filed this action against John Michael Paulson, Madeleine Pickens, James Paulson, Vikki Paulson, and Crystal Christensen in their individual and representative capacities. The complaint sought a judgment against the estate and the living trust for the outstanding balance of the 2006 estate tax liability, which then exceeded $10 million, as well as judgments against the individual defendants under
James Paulson, Vikki Paulson, Crystal Christensen, and Madeleine Pickens filed motions to dismiss and argued that they were not personally liable for the estate taxes under
II
These appeals raise questions of statutory interpretation, which we review de novo. Mada-Luna v. Fitzpatrick, 813 F.2d 1006, 1011 (9th Cir. 1987).
III
Section 2001 of the Internal Revenue Code imposes a tax on a decedent‘s taxable estate, which the executor is required to pay.
operates to protect the government‘s ability to collect estate and gift taxes. See
The statutory provision at issue here,
If the estate tax imposed by chapter 11 is not paid when due, then the spouse, transferee, trustee (except the trustee of an employees’
trust which meets the requirements of section 401(a)), surviving tenant, person in possession of the property by reason of the exercise, nonexercise, or release of a power of appointment, or beneficiary, who receives, or has on the date of the decedent‘s death, property included in the gross estate under sections 2034 to 2042, inclusive, to the extent of the value, at the time of decedent‘s death, of such property, shall be personally liable for such tax.
The United States argues the limiting phrase “on the date of decedent‘s death” modifies only the immediately preceding verb “has,” and not the more remote verb “receives.” Therefore, in its view, the statute imposes personal liability on those listed in the statute who (1) receive estate property at any time on or after the date of the decedent‘s death, or (2) have estate property on the date of the decedent‘s death. Thus, it contends,
The defendants, in contrast, argue that the limiting phrase “on the date of the decedent‘s death” modifies both the immediately preceding verb “has,” and the more remote verb “receives.” Thus, under their interpretation, the statute imposes personal liability for the unpaid estate taxes only on those who receive or have property included in the gross estate on the date of the decedent‘s death. But those who receive property from the estate at any point after the date of the decedent‘s death have no personal liability for the unpaid estate taxes.
We conclude that the most natural reading of the statutory text, and other indicia of its meaning, supports the United States’ interpretation. Therefore, we hold that
A
“Statutory construction must begin with the language employed by Congress and the assumption that the ordinary meaning of that language accurately expresses the legislative purpose.” Engine Mfrs. Ass‘n v. S. Coast Air Quality Mgmt. Dist., 541 U.S. 246, 252 (2004) (internal quotation marks omitted) (quoting Park ‘N Fly, Inc., v. Dollar Park & Fly, Inc., 469 U.S. 189, 194 (1985)); see also, e.g., Facebook, Inc. v. Duguid, 141 S. Ct. 1163, 1169 (2021) (explaining that when interpreting a statute, “[w]e begin with the text.“); United States v. Ron Pair Enters., Inc., 489 U.S. 235, 241 (1989) (“The task of resolving the dispute over the meaning of [a statute] begins where all such inquiries must begin: with the language of the statute itself.“).
Here, the statutory text at issue states that a person (who fits within a category listed in the statute) “who receives, or has on the date of the decedent‘s
This reading of the statute is supported by the canon of statutory construction known as “the rule of the last antecedent.” The Supreme Court has long applied this “timeworn textual canon” to interpret “statutes that include a list of terms or phrases followed by a limiting clause,”
Lockhart v. United States, 577 U.S. 347, 351 (2016). The “rule of the last antecedent” provides that “a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows.”12 Id. (alteration in original) (quoting Barnhart v. Thomas, 540 U.S. 20, 26 (2003)); see also id. (“[Q]ualifying words or phrases modify the words or phrases immediately preceding them and not words or phrases more remote, unless the extension is necessary from the context or the spirit of the entire writing.” (alteration in original) (quoting BLACK‘S LAW DICTIONARY 1532-33 (10th ed. 2014))). The rule of the last antecedent supports the conclusion that the limiting phrase “on the date of the decedent‘s death” modifies only the immediately preceding antecedent “has,” and not the more remote antecedent “receives.”
Vikki Paulson and Crystal Christensen, however, argue that we should apply the series-qualifier canon and conclude that the limiting phrase “on the date of the decedent‘s death” modifies both the immediately
series.‘” Facebook, 141 S. Ct. at 1169 (alteration in original) (quoting ANTONIN SCALIA & BRYAN A. GARNER, READING LAW: THE INTERPRETATION OF LEGAL TEXTS 147 (2012)).
In Facebook, the Court interpreted the Telephone Consumer Protection Act of 1991,
Here, however, the limiting phrase in
clause that contains both antecedents. Instead, the limiting phrase is set off by commas with the immediate antecedent, “has,” from the rest of the sentence (“who receives, or has on the date of the decedent‘s death, property included in the gross estate“).
Moreover, accepting the defendants’ interpretation would require us to read the statute as if it were punctuated differently—to essentially rewrite the statute. Specifically, we would either need to read the statute as if the two verbs “receives” and “has” appeared together in an integrated clause and were separated from the limiting phrase by a comma (i.e., a person who receives or has, on the date of the decedent‘s death, property included in the gross estate is liable for the unpaid estate taxes) or as if the statute included an additional comma that separated the limiting phrase from the antecedents (i.e., a person, who receives, or has, on the date of the decedent‘s death, property included in the gross estate is liable for the unpaid estate taxes). Cf. In re Bateman, 515 F.3d 272, 277 (4th Cir. 2008) (reading a provision in the bankruptcy code so that “[n]o punctuation needs to be added or deleted” (internal quotation marks and citation omitted)). But Congress did not structure the statute this way. See Int‘l Primate Prot. League v. Adm‘rs of Tulane Educ. Fund, 500 U.S. 72, 79–80 (1991) (explaining that Congress would have added a comma if it had intended a meaning other than the natural reading);14 see also In re
Sanders, 551 F.3d 397, 400 (6th Cir. 2008) (“Congress no doubt could have worked around [the rule of the last antecedent] had it wished ....“).
We therefore conclude that the rule of the last antecedent is the canon of interpretation that is most consistent with the text, structure, and punctuation of
B
This conclusion, however, does not end our inquiry. As the Court has explained, canons of statutory interpretation are not absolute and can be “overcome by other indicia of meaning.” Lockhart, 577 U.S. at 352 (citations omitted); see also Facebook, 141 S. Ct. at 1170 n.5 (“Linguistic canons are tools of statutory interpretation whose usefulness depends on the particular statutory text and context at issue.“). Here, however, applying the rule of the last antecedent results in an interpretation of
This is so because we are also bound by the canon that requires us to “strive to ‘giv[e] effect to each word and mak[e] every effort not to interpret a provision in a manner supported the conclusion that the phrase ‘Any officer of the United States or any agency thereof, or person acting under him,’ did not permit agencies to remove civil suits from state to federal court. 500 U.S. at 79-80. As the Court explained, ‘[i]f the drafters of
that renders other provisions of the same statute inconsistent, meaningless or superfluous.‘” R.J. Reynolds Tobacco Co. v. County of Los Angeles, 29 F.4th 542, 553 (9th Cir. 2022) (alterations in original) (quoting Shelby v. Bartlett, 391 F.3d 1061, 1064 (9th Cir. 2004)). The defendants’ narrow interpretation of
That clause applies
The statute also explicitly applies to those who already have or possess estate property on the date of the decedent‘s death, such as a “surviving tenant” or a “person in possession of the property.”
Thus, the context and structure of the statute provide additional indicia of its meaning and further clarify that personal liability for the estate tax applies to those who receive estate property, on or after the date of the decedent‘s death (i.e., through annuities, other receivable payments, powers of appointment, or insurance policies), and to those who have estate property on the date of the decedent‘s death (e.g., through a survivorship tenancy).
Vikki Paulson and Crystal Christensen acknowledge that
But the statute does not state that liability for unpaid estate taxes attaches only to those who can pay the taxes on the date of the decedent‘s death. Instead, the statute imposes personal liability for the unpaid estate taxes based on the receipt or possession of property from the gross estate. See
Madeleine Pickens, on the other hand, argues that “[§§] 2039 and 2042 do not bring within the gross estate insurance proceeds and annuity payments received on the date of death, but rather insurance payments and annuity payments receivable on the date of the decedent‘s death.” Although she acknowledges that these payments are receivable at the decedent‘s death and “may not actually be paid until some later point,” she maintains “[i]t is that receivable“—the receivable available at the decedent‘s death “that is brought within the grоss estate by [§§] 2039 and 2042.” But the statute does not impose personal liability on those who “receive a receivable” on the date of the decedent‘s death. See
Madeleine Pickens also argues that the statute‘s incorporation of
includes this distinction. Instead, the statute explicitly applies to property that trustees, transferees, beneficiaries, and others listed in the statute have or receive. Property that exists on the date of the decedent‘s death, including property within the scope of
Therefore, we conclude that the context and structure of
C
Vikki Paulson and Crystal Christensen also argue that applying the rule of the last antecedent to interpret the statute, as in the government‘s proposed “overly broad interpretation,” would result in “two absurd situations.” First, they argue that if
person to whom a property interest is conveyed, which, in their view, includes “purchasers.” Second, they argue that because the estate property is valued “at the time of the decedent‘s death,” if the property later depreciates, those who receive estate property after the date of the decedent‘s death could be personally liable for estate taxes that exceed the value of the property they received.
Although not expressly stated in their briefing, it appears these defendants are impliedly invoking the canon against absurdity. See United States v. Middleton, 231 F.3d 1207, 1210 (9th Cir. 2000) (explaining that a court should avoid an interpretation of a statute that would produce “an absurd and unjust result which Congress could not have intended“) (quoting Clinton v. City of New York, 524 U.S. 417, 429 (1998)). The defendants, however, fail to address long-standing Supreme Court and Ninth Circuit case law that strictly limits the circumstances in which the absurdity canon may apply. See, e.g., Crooks v. Harrelson, 282 U.S. 55, 60 (1930) (explaining that the absurdity doctrine is applied “only under rare and exceptional circumstances,” and that “the absurdity must be so gross as to shock the general moral or common sense“); see also id. (explaining that the application of the absurdity doctrine “so nearly approaches the boundary between the exercise of the judicial power and that of the legislative power as to call rather for great caution and circumspection in order to avoid usurpation of the latter.).16
As the Court explained in Crooks, Congress may enact legislation that “turn[s] out to be mischievous, absurd, or otherwise objectionable. But in such case the remedy lies with the lawmaking authority, and not with the courts.” Id. (citations omitted); see also Griffin v. Oceanic Contractors, Inc., 458 U.S. 564, 571, 574-75 (1982) (concluding that an interpretation of federal maritime statute that resulted in $300,000 award to seaman for back wages penalty, when he had incurred only $412 in unpaid wages, did not present an “exceptional case” that allowed court to apply the absurdity doctrine); see also id. at 576 (“The remedy for any dissatisfaction with the results in particular cases lies with Congress and not with this Court. Congress may amend the statute; we may not.“).
As we explain next, without even reaching the absurdity canon, the defendants’ first argument—suggesting tax liability could be applied to bona fide purchasers of estate assets—fails based on the plain language of
“patently absurd” results, such as shown by the “few examples of true absurdity...given in the Holy Trinity decision,” of prosecuting a sheriff for obstruction of the mail when he was executing a warrant to arrest a mail carrier for murder, or applying “a medieval law against drawing blood in the streets” to a physician treating “a man who had fallen down in a fit“).
have intended the result‘“) (quoting In re Hokulani Square, Inc., 776 F.3d 1083, 1088 (9th Cir. 2015)).
1
The defendants’ first argument fails because
Moreover, the tax code provides different definitions for “transferees” and “purchasers.” In
In
only if the estate‘s executor has been discharged from personal liability pursuant to
2
a
The defendants’ second argument also fails. The defendants correctly state that the statutory language imposes estate tax liability “to the extent of the value, at the time of the decedent‘s death, of such property.”
The defendants, however, dispute that Congress could have also anticipated that estate property could depreciate after the date of the decedent‘s death and thus potentially result in tax liability for the recipient that exceeds the property‘s value.20 The defendants argue that an
interpretation of
This is not a situation where it is “quite impossible” that Congress could have intended the result. See Lopez, 998 F.3d at 438 (citation omitted). Here, Congress clearly could have anticipated that the value of estate property could change after the date of the decedent‘s death—either by increasing or decreasing in value—and thus could have
of loss would apply equally to those who receive estate property on the date of the decedent‘s death and to those who receive estate property after the date of the decedent‘s death. There is nothing about the risk of accepting property that may decline in value that would apply unfairly to those who receive such property after the date of the decedent‘s death.
anticipated that the value
And while it is “not our job to find reasons for what Congress has plainly done,” Lopez, 998 F.3d at 447 (M. Smith, J., concurring) (internal quotation marks and citation omitted), Congress rationally could have concluded that such risk is acceptable or is effectively mitigated by other provisions of the tax code, and thus is outweighed by the benefit of ensuring the collection of estate taxes. This is not an irrational tax policy. Indeed, we have previously recognized that “[§] 6324 is structured to assure collection of the estate tax.” Vohland, 675 F.2d at 1076. Moreover, even if it were to conclude that such a policy is “odd,” or “not wise,” Lopez, 998 F.3d at 447 (M. Smith, J., concurring) (citation omitted), or simply unfair, we cannot rewrite the statute to advance a different policy, id. at 440 (majority opinion). See also Hokulani Square, 776 F.3d at 1088 (“The absurdity canon isn‘t a license for us to disregard statutory text where it conflicts with our policy preferences . . . .“). And if Congress determines that its tax policy leads to
unintended or unfair results, it is for Congress, not the courts, to rewrite the tax code. See Crooks, 282 U.S. at 60; Griffin, 458 U.S. at 576. Therefore, we conclude that applying the rule of the last antecedent to
b
But our conclusion—that this is not the “exceptional” case where we can invoke the absurdity canon to reject the interpretation of a statute that is most consistent with its text, structure, punctuation, and other indicia of meaning—does not mean that the defendants’ “the sky is falling”23 arguments are based on anything other than remote hypotheticals. And even if the defendants could demonstrate that applying
As an initial matter, before those who receive estate property could be subjected
First, the property must have depreciated after the date of the decedent‘s death to the point that it is worth less than the tax liability, which is calculated as a percentage of the amount of the taxable estate.25 See
Second, the executor must have failed to pay the estate tax before distributing estate property. See
Third, the estate must have “divest[ed] itself of the assets necessary to satisfy its tax obligations,” Geniviva, 16 F.3d at 524, thus defeating the lien for estate taxes under that would apply under
Fourth, the statute of limitations must not have expired by the time the property is distributed or the government attempts collection. See
Fifth, a transferee, beneficiary, or other recipient of the estate property must not have disclaimed or refused the property. See
Sixth, the government must successfully seek to impose tax liability on a transferee, beneficiary, or other recipient of estate property in an amount that exceeds the value of the property they received.
Focusing on the final factor—whether the government would later seek to impose tax liability that exceeds the value of the property received and would be successful in advancing that argument—we rely on the government‘s avowals in its briefing and at oral argument that estate tax liability cannot exceed the value of the property received. Specifically, the government asserted in its briefing that the language in
the person‘s liability is capped at the value of the property had or received.”27
moment.”29 Id. at 749-50 (internal quotation marks and citations omitted).
The Court has identified three non-exclusive factors that should “inform” a court‘s decision whether to apply judicial estoppel: (1) “a party‘s later position must be ‘clearly inconsistent’ with its earlier position“; (2) “the party has succeeded in persuading a court to accept that party‘s earlier position, so that judicial acceptance of an inconsistent position in a later proceeding would create ‘the perception that either the first or the second court was misled‘“; and (3) “the party seeking to assert an inconsistent position would derive an unfair advantage or impose an unfair detriment on
the opposing party if not estopped.” Id. at 750-51 (internal quotation marks and citations omitted).
If these considerations were applied to the government‘s representations here—that
Moreover, there are cases that, while not directly addressing the issue before us now, include statements that lend support to the government‘s argument that it does not seek to impose liability for estate taxes that exceed the value of the property received. See Geniviva, 16 F.3d at 523 (construing
Finally, defendants have not identified, and our research has not uncovered, any case in which the government has attempted to impose personal liability for estate taxes that exceeded the value of the property received. The absence of any case law on this point supports the conclusion that this situation has never been litigated because the government has never taken this position, which in turn, supports the conclusion that it is unlikely that the government will attempt to assert this argument in future litigation.
Thus, we conclude that applying the rule of the last antecedent does not lead to absurd results, but instead results in the most natural reading of the statute, consistent with its structure and context.
D
The defendants also argue that to interpret the statute we must consider its purpose and intent. Madeleine Pickens argues that “the purpose of
But nothing in the statutory text supports her argument that Congress‘s purpose in enacting
Instead,
Vikki Paulson and Crystal Christensen also argue that we should interpret the statute based on Congress‘s intent. They baldly assert that “Congress did not intend that individuals who had no control over estate property at the date of the decedent‘s death be held liable for unpaid estate taxes.” This argument, like Madeleine Pickens’ “purpose of the statute” argument, fails because it has no support in the statutory text. There is nothing in the statute that suggests that liability for unpaid estate taxes is based on the opportunity to ensure that taxes are paid at a particular time; instead, the statute imposes personal liability on those who receive or have estate property.
E
The defendants also argue that ambiguities in tax statutes must be resolved in favor of the taxpayer and against the government. However, as the United States argues, the “modern validity” of the “taxpayer rule of lenity” is “questionable.” See Colgate-Palmolive-Peet Co. v. United States, 320 U.S. 422, 429–30 (1943) (resolving ambiguity in taxing statute in favor of the government); Maloney v. Portland Assocs., 109 F.2d 124, 126 (9th Cir. 1940) (“[T]here is considerable doubt as to the present existence of the old rule to the effect that ambiguities in a taxing act are to be resolved in favor of the taxpayer.“); SCALIA & GARNER, supra, at 299–300, & nn.17–19 (explaining that the Court previously construed tax laws “strict[ly]” and in “case[s] of doubt . . . against the government,” but the rule “can no longer be said to enjoy universal approval.” (footnotes omitted)); see also Fang Lin Ai v. United States, 809 F.3d 503, 507 (9th Cir. 2015) (“[W]e do not mechanically resolve doubts in favor of the taxpayer but instead resort to the ordinary tools of statutory interpretation.“).
Vikki Paulson and Crystal Christensen acknowledge that “the rule of lenity is sometimes called into question,” but they argue that the Ninth Circuit “still strictly construes tax provisions to resolve ambiguity in the taxpayer‘s favor.” To support this broad assertion they cite our decision in United States v. Boyd, 991 F.3d 1077, 1085 (9th Cir. 2021). But defendants’ arguments, if accepted, would require us to stretch Boyd beyond its language and reasoning—in Boyd, we did not state that the rule of lenity applies to all ambiguous “tax provisions” or that all such provisions must be strictly construed. See id. at 1085–86. Instead, our discussion was limited to “tax provision[s] which impose[] a penalty.” Id. at 1085 (emphasis added).
To be sure, we explained that “our circuit strictly construes tax penalty provisions
But we need not decide the modern validity of the rule of lenity as applied to all tax provisions because that rule does not apply to the statute at issue here. That is because “[t]he rule ‘applies only when, after consulting traditional canons of statutory construction, we are left with an ambiguous statute.‘” Shular v. United States, 140 S. Ct. 779, 787 (2020) (quoting United States v. Shabani, 513 U.S. 10, 17 (1994)); see id. at 788 (Kavanaugh, J., concurring) (“Of course, when a reviewing court employs all of the traditional tools of construction, the court will almost always reach a conclusion about the best interpretation, thereby resolving any perceived ambiguity. That explains why the rule of lenity rarely comes into play.” (internal quotation marks and citation omitted)). As previously explained, after reviewing the text of
F
Finally, the defendants argue that we must accept their interpretation of
In both cases, without any attempt to construe the statutes by applying the traditional tools—namely the canons of statutory interpretation—the courts concluded that because the statutory language could support different interpretations, the statutes must be deemed ambiguous, and thus “any doubt as to the meaning of the statutes” must be resolved in the taxpayer‘s favor.32 Englert, 32 T.C. at 1016; see also Johnson, 2013 WL 3924087, at *5 (“Where there is ambiguity as to the meaning of a tax statute, the court must resolve the issue in favor of the taxpayer.“). But, as discussed above, even if the rule of lenity validly applies to taxing statutes, it does so “only when, after consulting traditional canons of statutory construction, we are left with an ambiguous statute.” Shular, 140 S. Ct. at 787 (internal quotation marks and citation omitted). Because the courts in Englert and Johnson made no attempt to “resolv[e] any perceived ambiguity,” see id. at 788 (Kavanaugh, J., concurring), they erroneously concluded that they were required to construe the statutes at issue in the taxpayer‘s favor. Therefore, we decline the defendants’ suggestion that we adopt the reasoning of these cases.
* * * *
After starting our analysis with the text of
IV
Our holding that
A
The government argues that the defendants are liable under the statute as trustees, transferees, and beneficiaries. Vikki Paulson and Crystal Christensen acknowledge that they are successor trustees, and James Paulson has not submitted a brief contesting the district court‘s finding that he was a successor trustee. Thus, these defendants do not dispute that, if
We therefore conclude that James Paulson, Vikki Paulson, and Crystal Christensen are liable, as trustees, for the unpaid estate taxes on property from the gross estate, held in the living trust, “to the extent of the value, at the time of the decedent‘s death, of such property.”
B
The government also argues that the ordinary meaning of “beneficiary” includes “trust beneficiaries” and therefore
Because the statute does not define “beneficiary,” “we look first to the word‘s ordinary meaning.” See Schindler Elevator Corp. v. United States, 563 U.S. 401, 407 (2011) (citing Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 175 (2009) (“Statutory construction must begin with the language employed by Congress and the assumption that the ordinary meaning of that language accurately expresses the legislative purpose” (internal quotation marks omitted))); Asgrow Seed Co. v. Winterboer, 513 U.S. 179, 187 (1995) (“When terms used in a statute are undefined, we give them their ordinary meaning“). At this first step, we conclude that dictionary definitions support the government‘s broad interpretation, rather than the defendants’ narrow interpretation limiting liability to insurance beneficiaries. See Beneficiary, BLACK‘S LAW DICTIONARY (11th ed. 2019) (defining “beneficiary” as “[s]omeone who is designated to receive the advantages from an action or change; esp., one designated to benefit from an appointment, disposition, or assignment (as in a will, insurance policy, etc.), or to receive something as a result of a legal arrangement or instrument,” and “[s]omeone designated to receive money or property from a person who has died“); see also Beneficiary, AMERICAN HERITAGE DICTIONARY (5th ed. 2018) (“One that receives a benefit” or “the recipient of funds, property, or other benefits, as from an insurance policy or trust“); Beneficiary, WEBSTER‘S NEW WORLD COLLEGE DICTIONARY (5th ed 2014) (“[A]nyone receiving benefit” or “a person named to receive the income or inheritance from a will, insurance policy, trust, etc. . . .“); Beneficiary, WEBSTER‘S NEW WORLD DICTIONARY (4th ed. 2003) (“[A]nyone receiving or to receive benefits, as funds from a will or insurance policy . . . .“); Beneficiary, 2 OXFORD ENGLISH DICTIONARY (2d еd. 1989) (“[O]ne who receives benefits or favours; a debtor to another‘s bounty . . . .“). Therefore, we conclude that the ordinary meaning of “beneficiary” includes a “trust beneficiary.”
C
But we must also consider whether “there is any textual basis for adopting a narrower definition” of “beneficiary.” See Schindler, 563 U.S. at 409; see also SCALIA & GARNER, supra, at 70 (“One should assume the contextually appropriate ordinary meaning unless there is reason to think otherwise. Sometimes there is reason to think otherwise, which ordinarily comes from context.” (emphasis in original)). The government argues that the text of
We start with Higley v. Commissioner, in which the Eighth Circuit interpreted the word “beneficiary” in § 315(b) of the Revenue Act of 1926. 69 F.2d at 162. The text of this predecessor statute, however, differs significantly from the text of
If (1) the decedent makes a transfer, by trust or otherwise, of any property in contemplation of or intended to take effect in possession or enjoyment at or after his death . . . or (2) if insurance passes under a contract executed by the decedent in favor of a specific beneficiary, and if in either case the tax in respect thereto is not paid when due, then the transferee, trustee, or beneficiary shall be personally liable for such tax[.]
Id. (quoting 26 U.S.C. § 1115(b) (emphasis added)). As the court recognized in its analysis of the statute, § 315(b) expressly addressed two types of property dispositions: (1) “transfers,” including “trusts,” and (2) “insurance,” and imposed liability on the “transferee, trustee, or beneficiary.” Id. Indeed, the statute specifically referred to “insurance . . . in favor of a specific beneficiary.” Id. The court concluded that this structure meant that the word “trustee” was “employed in connеction with trust only,” and the word “beneficiary” “applies only to insurance policy beneficiaries.” Id.
But this direct textual and structural correlation between (1) dispositions by “transfers” and “trusts” to the liability of a “transferee” or “trustee,” and (2) dispositions of “insurance in favor of a specific beneficiary” to the liability of a “beneficiary,” is not present in
We next consider Englert v. Commissioner, in which the Tax Court interpreted another predecessor statute,
If the tax herein imposed is not paid when due, then the spouse, transferee, trustee, surviving tenant, person in possession of the property by reason of the exercise, nonexercise, or release of a power of appointment, or beneficiary, who receives, or has on the date of the decedent‘s death, property included in the gross estate under section 811(b), (c), (d), (e), (f), or (g), to the extent of the value, at the time of the decedent‘s death, of such property, shall be personally liable for such tax.
Id. at 1017 n.4 (quoting 26 U.S.C. § 827(b)).
As the Tax Court noted,
The court stated its belief that Congress “studiously chose a classification applicable to each of such subsections and included them in section 827(b) in the same order as the related property interests appeal in subsections (b) through (g), inclusive, of section 811.” Id. at 1016. Applying this reasoning, and as petitioner argued, the court concluded that a person liable under the statute as a beneficiary would be limited to the beneficiary of a life insurance policy under § 811(g). See id. at 1013, 1016.
But
As an initial matter, in Englert, the tax court found compelling the direct correlation of the six categories of persons liable to the six categories of property included in the gross estate, and concluded it was the result of Congress‘s “studious[] cho[ice.]” Id. at 1016. That direct correlation is not present in
does not provide a textual or structural basis for us to conclude that the word “beneficiary” in
Despite the textual and structural differences between
D
We must also apply the presumption of consistent usage that “a word or phrase is presumed to bear the same meaning throughout a text.” SCALIA & GARNER, supra, at 170; see also id. at 172 (“The presumption of consistent usage applies also when different sections of an act or code are at issue.“). In this case, we note that the use of the term “beneficiary,” in different sections of the tax code and in the regulations, supports the broader, ordinary meaning of the word.
First, the defendants argue that
Second, the same is true for
Finally, the regulations addressing liability for estate taxes use the term “beneficiary” broadly to indicate those who receive distributions from the estate, or in other words, trust beneficiaries. See
E
Finally, the defendants offer policy arguments to support their interpretation of the statute. Crystal Christensen argues that because trust beneficiaries have “no power to take estate property,” or “to distribute it,” they should not be liable fоr the estate taxes if a trustee mismanages the estate and distributes property before “ensuring the estate‘s taxes [are] paid in full.” But the statute does not condition personal liability for the unpaid estate taxes on the power to take or distribute estate property. Instead, it imposes personal liability on categories of persons who receive or have estate property, and those categories include persons who do not have the power to take or distribute estate property.
Indeed, the defendants recognize that life insurance beneficiaries are “beneficiaries” under
The defendants also argue that we should reject the government‘s argument that
* * * *
We conclude that the ordinary meaning of beneficiary, which includes trust beneficiaries, applies to
V
Because
REVERSED and REMANDED.
IKUTA, Circuit Judge, dissenting:
Our only task in interpreting
Rather than adopt a reasonable interpretation of the statute that is more likely to reflect congressional intent, the majority adopts a “hypertechnical reading” of statutory lаnguage that loses sight of the “fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” Davis v. Mich. Dep‘t of Treasury, 489 U.S. 803, 809 (1989) (citation omitted). In order to justify this approach, the majority and the government proffer a number of unpersuasive rationales. First, the government provides a non-responsive description of its litigating position: it states it “has consistently argued” that it would not impose liability greater than the value of the property received. The majority, in turn, suggests that the result of its interpretation is not likely to occur. But neither the government‘s nor the majority‘s assurances about the future (that individuals are unlikely to be held personally liable for estate taxes that potentially exceed the current value of the property received from a decedent‘s estate) impacts the interpretation of the statute.
Because the taxpayers’ reading is more plausible and avoids the majority‘s illogical result, it is a better indication of Congress‘s intent. The inquiry should end there. Therefore, I respectfully dissent.
I
A
When an individual dies, an estate tax lien automatically arises and attaches to the decedent‘s gross estate.
In addition to a lien,
tax purposes. See
B
In this case, the estate elected to defer payments over fourteen years. But the government failed to use the options available to protect its unsecured interests in deferred payments. See supra, at 59. It also failed to hold Michael Paulson, the trustee of the decedent‘s trust on the date of the decedent‘s death, personally liable for the estate taxes due, United States v. Paulson, 445 F. Supp. 3d 824, 831 (S.D. Cal. 2020), even though such liability may extend after the expiration of the ten-year estate tax lien provided for in
To compensate for its failures to use the available statutory options to collect estate taxes, the government here adopted a novel reading of
The majority justifies its adoption of the government‘s novel reading based on the lack of a comma after the word “has.” The majority views the absence of a comma as triggering the doctrine of the last antecedent, a rule of statutory construction which states that “a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows.” Lockhart v. United States, 577 U.S. 347, 351 (2016) (citation omitted). But while “[p]unctuation is a permissible indicator of meaning,” Navajo Nation v. U.S. Dep‘t of Interior, 819 F.3d 1084, 1093 (9th Cir. 2016) (citing Antonin Scalia & Bryan A. Garner, READING LAW: THE INTERPRETATION OF LEGAL TEXTS 161-65 (2012)), it “can assuredly be overcome by other indicia of meaning,” Barnhart v. Thomas, 540 U.S. 20, 26 (2003) (citation omitted). The “last antecedent principle is merely an interpretive presumption based on the grammatical rule against misplaced modifiers.” Payless Shoesource, Inc. v. Travelers Cos., Inc., 585 F.3d 1366, 1371-72 (10th Cir. 2009). “At the same time, though, we know that grammatical rules are bent and broken all the time,” and we should not rely solely on grammar in interpreting a text “when evident sense and meaning require a different construction.” Id. (citation and internal quotation marks omitted).
Like other circuits, we have acknowledged that the last antecedent canon is inapplicable when it creates illogical results and the statute‘s plain language gives rise to a more logical reading. See One Sentinel, 416 F.3d at 979. In One Sentinel, the government brought a civil forfeiture action against a Sentinel Arms Striker-12 shotgun on the ground that it was “a ‘destructive device’ possessed in violation of the National Firearms Act.” Id. at 978. The Act defined a destructive device as
any type of weapon by whatever name known which will, or which may be readily converted to, expel a projectile by the action of an explosive or other propellant, the barrel or barrels of which have a bore of more than one-half inch in diameter, except a shotgun or shotgun shell which the Secretary finds is generally recognized as particularly suitable for sporting purposes[.]
Id. at 979 (citing
We rejected that argument because following the last antecedent doctrine would have created the illogical result that no shotgun could be a “destructive device.” Id. We explained that “the doctrine of the last antecedent must yield to the most logical meaning of a statute that emerges from its plain language and legislative history.” Id. at 979 (citation and quotation marks omitted). Therefore, we declined to apply the last antecedent canon and interpreted the relevant clause as if an omitted comma after “shell” were included. Id.
The same principle applies here. The government and majority implicitly concede that the government‘s reading of the statute potentially results in allowing the government to impose personal liability for unpaid estate taxes on trust asset recipients in excess of the value of the assets received. This could occur under the government‘s interpretation, for instance, if property of the estate had a high value at the time of the decedent‘s death but decreased precipitously by the time it was received by a beneficiary. In such a case, the beneficiary would nevertheless be personally liable for the unpaid estate taxes based on the value of the property on the date of death, even if the property were worth mere cents on the dollar when received by the beneficiary. Congress could not have intended to make a person who receives property many years after a settlor‘s death personally liable for estate taxes that exceed the value of the property received.
The majority claims the taxpayers “are impliedly invoking the canon against absurdity,” and then refutes this strawman argument by pointing to the “high bar” for invoking this canon. But because the canon against absurdity applies only when a court departs from the plain meaning of a statute, see, e.g., Lamie v. U.S. Tr., 540 U.S. 526, 534 (2004); Taylor v. Dir., Off. of Workers’ Comp. Programs, 201 F.3d 1234, 1241 (9th Cir. 2000), it is not implicated here. The taxpayers do not ask the court to disregard the text of
C
While the majority primarily focuses on the doctrine of the last antecedent to support its interpretation of
D
As an alternative to its textual arguments, the majority attempts to defend its interpretation by predicting that its illogical results are unlikely to occur.2 But the majority cites no support for its approach of interpreting statutes based on predictions regarding future events. Nor can it, because our job is merely to discern the most reasonable interpretation of the statute, which requires us to take into account its “most logical meaning.” One Sentinel, 416 F.3d at 979 (citation and quotation marks omitted).
In any event, the majority‘s assurances are unpersuasive, even on their own terms. First, the majority claims that the illogical result caused by the government‘s interpretation can be avoided because an individual poised to receive trust assets “must not have disclaimed or refused [trust] property.” In other words, according to the majority, prospective recipients of trust assets are amply protected becausе they can simply refuse assets that will suffer too great a decrease in value.
The majority‘s argument does not survive scrutiny. Federal disclaimer law applies in this context. See
whether their interests are vested or contingent, must disclaim no later than 9 months after the original transfer creating an interest.“); see also Breakiron v. Gudonis, No. 10-cv-10427, 2010 WL 3191794, at *1 (D. Mass. Aug. 10, 2010) (“Under Treasury Regulation
The majority fails to explain how a person would have the prescience to know within nine months from the date of decedent‘s death that the value of the interest in property to be transferred to that person at some point in the future will dramatically decline many years later (assuming that person even knows оf the existence of such an interest). Without this prescience, the person would not be able to disclaim such an asset within the required time frame. At bottom, a person‘s right to disclaim an asset within nine months of decedent‘s death does not avoid the result caused by the government‘s and majority‘s interpretation of the statute.
The majority also contends that it “rel[ies] on the government‘s avowals in its briefing and at oral argument that estate tax liability cannot exceed the value of the property received.” According to the majority, this promise, coupled with “judicial estoppel, provides additional safeguards” against the unfair application of personal liability under
But even if the government had offered an authoritative interpretation, the majority misunderstands how the doctrine of judicial estoppel (which the government does not raise) would apply in this case. Judicial estoppel is an equitable doctrine that generally “prevents a party from prevailing in one phase of a case on an argument and then relying on a contradictory argument to prevail in another phase.” New Hampshire v. Maine, 532 U.S. 742, 749 (2001) (quoting Pegram v. Herdrich, 530 U.S. 211, 227 n.8 (2000)). “Courts apply the doctrine where a party‘s ‘later inconsistent position’ presents a ‘risk of inconsistent court determinations.‘” New Edge Network, Inc. v. FCC, 461 F.3d 1105, 1114 (9th Cir. 2006). The doctrine is “invoked by a court at its discretion” to “protect the integrity of the judicial process.” Russell v. Rolfs, 893 F.2d 1033, 1037 (9th Cir. 1990).
Judicial estoppel is not applicable here. In future cases, a court would be bound only by the majority‘s interpretation of
But even if the government had provided (and the majority had adopted) an interpretation of
Finally, instead of explaining why its statutory interpretation does not lead to a nonsensical result, the majority also argues
II
The majority has overemphasized a single canon of statutory construction—the rule of the last antecedent—to ignore that “fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 133 (2000) (citing Davis, 489 U.S. at 809). Although the punctuation chosen by Congress is important, we must also give due regard to sense and meaning. As our sister circuit has explained, “while the rules of English grammar often afford a valuable starting point to understanding a speaker‘s meaning, thеy are violated so often by so many of us that they can hardly be safely relied upon as the end point of any analysis of the parties’ plain meaning.” Payless Shoesource, Inc., 585 F.3d at 1372. Our binding precedent requires this approach; we may not read a statute as defining a “destructive device” to include shotgun shells but not shotguns merely because of a misplaced comma. One Sentinel, 416 F.3d at 979. And the Tenth Circuit offers an example that speaks volumes: “Groucho Marx could joke in Animal Crackers, ‘One morning I shot an elephant in my pajamas. How he got into my pajamas I‘ll never know,’ leaving his audience at once amused by the image of a pachyderm stealing into his night clothes and yet certain that Marx meant something very different.” Payless Shoesource, Inc., 585 F.3d at 1372. Because I would interpret the statute according to the most likely intent of Congress, rather than adopt the majority‘s mechanical adherence to the rule of the last antecedent, I respectfully dissent.
Notes
If the tax herein imposed is not paid when due, then the spouse, transferee, trustee, surviving tenant, person in possession of the property by reason of the exercise, nonexercise, or release of a power of appointment, or beneficiary, who receives, or has on the date of the decedent‘s death, property included in the gross estate under section 811(b), (c), (d), (e), (f), or (g), to the extent of the value, at the time of the decedent‘s death, of such property, shall be personally liable for such tax.
the power of appointment shall be considered to exist on the date of the decedent‘s death even though the exercise of the power is subject to a precedent giving of notice or even though the exercise of the power takes effect only on the expiration of a stated period after its exercise, whether or not on or before the date of the decedent‘s death notice has been given or the power has been exercised.
Upon transfer of non-probate property to a purchaser, the property is divested of the lien, so that a purchaser of such property is fully protected.
[26 U.S.C.] § 6324(a)(2) . Property that was part of the ‘probate’ estate, i.e.,[§] 2033 property, is divested of the lien when it is transferred to a subsequent purchaser, but