holding that “federal authorities are not bound” by state-court determination of facts affecting federal tax liabilities if the United States was not a party to the state proceedingsSummary of this case from Marren v. Stout
Together with No. 240, Second National Bank of New Haven, Executor v. United States, also on certiorari to the same court.
Where federal estate tax liability turns upon the character of a property interest held and transferred by the decedent under state law, held, federal authorities are not bound by the determination made of such property interest by a state trial court; if there is no decision by the State's highest court federal authorities must apply what they find to be the state law after giving "proper regard" to relevant rulings of other courts of the State. Pp. 457, 462-466.
No. 673, 363 F.2d 1009, reversed and remanded; No. 240, 351 F.2d 489, affirmed.
Jack S. Levin argued the cause for petitioner in No. 673 and for the United States in No. 240. With him on the briefs were Acting Solicitor General Spritzer, Assistant Attorney General Rogovin, Richard C. Pugh, Meyer Rothwacks, Robert N. Anderson and Thomas Silk, Jr.
Curtiss K. Thompson argued the cause for petitioner in No. 240. With him on the briefs was John H. Weir.
John W. Burke, Jr., argued the cause and filed a brief for respondent in No. 673.
These two federal estate tax cases present a common issue for our determination: Whether a federal court or agency in a federal estate tax controversy is conclusively bound by a state trial court adjudication of property rights or characterization of property interests when the United States is not made a party to such proceeding.
In No. 673, Commissioner of Internal Revenue v. Estate of Bosch, 363 F.2d 1009, the Court of Appeals for the Second Circuit held that since the state trial court had "authoritatively determined" the rights of the parties, it was not required to delve into the correctness of that state court decree. In No. 240, Second National Bank of New Haven, Executor v. United States, 351 F.2d 489, another panel of the same Circuit held that the "decrees of the Connecticut Probate Court . . . under no circumstances can be construed as binding" on a federal court in subsequent litigation involving federal revenue laws. Whether these cases conflict in principle or not, which is disputed here, there does exist a widespread conflict among the circuits over the question and we granted certiorari to resolve it. 385 U.S. 966, 968. We hold that where the federal estate tax liability turns upon the character of a property interest held and transferred by the decedent under state law, federal authorities are not bound by the determination made of such property interest by a state trial court.
Illustrative of the conflict among the circuits are: Gallagher v. Smith, 223 F.2d 218 (C.A. 3d Cir., 1955); Faulkerson's Estate v. United States, 301 F.2d 231 (C.A. 7th Cir.), cert. denied, 371 U.S. 887 (1962); Pierpont v. C. I. R., 336 F.2d 277 (C.A. 4th Cir., 1964), cert. denied, 380 U.S. 908 (1965).
(a) No. 673, Commissioner v. Estate of Bosch.
In 1930, decedent, a resident of New York, created a revocable trust which, as amended in 1931, provided that the income from the corpus was to be paid to his wife during her lifetime. The instrument also gave her a general power of appointment, in default of which it provided that half of the corpus was to go to his heirs and the remaining half was to go to those of his wife. In 1951 the wife executed an instrument purporting to release the general power of appointment and convert it into a special power. Upon decedent's death in 1957, respondent, in paying federal estate taxes, claimed a marital deduction for the value of the widow's trust. The Commissioner determined, however, that the trust corpus did not qualify for the deduction under § 2056(b)(5) of the 1954 Internal Revenue Code and levied a deficiency. Respondent then filed a petition for redetermination in the Tax Court. The ultimate outcome of the controversy hinged on whether the release executed by Mrs. Bosch in 1951 was invalid — as she claimed it to be — in which case she would have enjoyed a general power of appointment at her husband's death and the trust would therefore qualify for the marital deduction. While the Tax Court proceeding was pending, the respondent filed a petition in the Supreme Court of New York for settlement of the trustee's account; it also sought a determination as to the validity of the release under state law. The Tax Court, with the Commissioner's consent, abstained from making its decision pending the outcome of the state court action. The state court found the release to be a nullity; the Tax Court then accepted the state court judgment as being an "authoritative exposition of New York law and adjudication of the property rights involved," 43 T.C. 120, 124, and permitted the deduction. On appeal, a divided Court of Appeals affirmed. It held that "[t]he issue is . . . not whether the federal court is `bound by' the decision of the state tribunal, but whether or not a state tribunal has authoritatively determined the rights under state law of a party to the federal action." 363 F.2d, at 1013. The court concluded that the "New York judgment, rendered by a court which had jurisdiction over parties and subject matter, authoritatively settled the rights of the parties, not only for New York, but also for purposes of the application to those rights of the relevant provisions of federal tax law." Id., at 1014. It declared that since the state court had held the wife to have a general power of appointment under its law, the corpus of the trust qualified for the marital deduction. We do not agree and reverse.
Section 2056(b)(5) of the Internal Revenue Code of 1954, 26 U.S.C. § 2056 (b)(5), provides: "(5) Life estate with power of appointment in surviving spouse. — In the case of an interest in property passing from the decedent, if his surviving spouse is entitled for life to all the income from the entire interest, . . . with power in the surviving spouse to appoint the entire interest, . . . (exercisable in favor of such surviving spouse, or of the estate of such surviving spouse, or in favor of either, whether or not in each case the power is exercisable in favor of others), and with no power in any other person to appoint any part of the interest, or such specific portion, to any person other than the surviving spouse — "(A) the interest . . . thereof so passing shall, for purposes of subsection (a), be considered as passing to the surviving spouse, and "(B) no part of the interest so passing shall, for purposes of paragraph (1)(A), be considered as passing to any person other than the surviving spouse. "This paragraph shall apply only if such power in the surviving spouse to appoint the entire interest, or such specific portion thereof, whether exercisable by will or during life, is exercisable by such spouse alone and in all events."
(b) No. 240, Second National Bank of New Haven, Executor v. United States.
Petitioner in this case is the executor of the will of one Brewster, a resident of Connecticut who died in September of 1958. The decedent's will, together with a codicil thereto, was admitted to probate by the Probate Court for the District of Hamden, Connecticut. The will was executed in 1958 and directed the payment "out of my estate my just debts and funeral expenses and any death taxes which may be legally assessed. . . ." It further directed that the "provisions of any statute requiring the apportionment or proration of such taxes among the beneficiaries of this will or the transferees of such property, or the ultimate payment of such taxes by them, shall be without effect in the settlement of my estate." The will also provided for certain bequests and left the residue in trust; one-third of the income from such trust was to be given to decedent's wife for life, and the other two-thirds for the benefit of his grandchildren that were living at the time of his death. In July of 1958, the decedent executed a codicil to his will, the pertinent part of which gave his wife a general testamentary power of appointment over the corpus of the trust provided for her. This qualified it for the marital deduction as provided by the Internal Revenue Code of 1954, § 2056(b)(5). In the federal estate tax return filed in 1959, the widow's trust was claimed as part of the marital deduction and that was computed as one-third of the residue of the estate before the payment of federal estate taxes. It was then deducted, along with other deductions not involved here, from the total value of the estate and the estate tax was then computed on the basis of the balance. The Commissioner disallowed the claimed deduction and levied a deficiency which was based on the denial of the widow's allowance as part of the marital deduction and the reduction of the marital deduction for the widow's trust, by requiring that the estate tax be charged to the full estate prior to the deduction of the widow's trust. After receipt of the deficiency notice, the petitioner filed an application in the state probate court to determine, under state law, the proration of the federal estate taxes paid. Notice of such proceeding was given all interested parties and the District Director of Internal Revenue. The guardian ad litem for the minor grandchildren filed a verified report stating that there was no legal objection to the proration of the federal estate tax as set out in the application of the executor. Neither the adult grandchildren nor the District Director of Internal Revenue filed or appeared in the Probate Court. The court then approved the application, found that the decedent's will did not negate the application of the state proration statute and ordered that the entire federal tax be prorated and charged against the grandchildren's trusts. This interpretation allowed the widow a marital deduction of some $3,600,000 clear of all federal estate tax. The Commissioner, however, subsequently concluded that the ruling of the Probate Court was erroneous and not binding on him, and he assessed a deficiency. After payment of the deficiency, petitioner brought this suit in the United States District Court for a refund. On petitioner's motion for summary judgment, the Government claimed that there was a genuine issue of material fact, i. e., whether the probate proceedings had been adversary in nature. The District Court held that the "decrees of the Connecticut Probate Court . . . under no circumstances can be construed as binding and conclusive upon a federal court in construing and applying the federal revenue laws." 222 F. Supp. 446, 457. The court went on to hold that under the standard applied by the state courts, there was no "clear and unambiguous direction against proration," and that therefore the state proration statute applied. Id., at 454. The Court of Appeals reversed, holding that the decedent's will "would seem to be clear and unambiguous to the effect that taxes were to come out of his residual estate and that despite any contrary statute the testator specifically wished to avoid any proration." 351 F.2d, at 491. It agreed with the District Court that, in any event, the judgment of the State Probate Court was not binding on the federal court.
Petitioner in No. 240 raises the additional point that the Court of Appeals was incorrect in holding that decedent's will clearly negated the application of the state proration statute. While we did not limit the grant of certiorari, we affirm without discussion the holding of the Court of Appeals on the point. The issue presents solely a question of state law and "[w]e ordinarily accept the determination of local law by the Court of Appeals . . . and we will not disturb it here." Ragan v. Merchants Transfer Co., 337 U.S. 530, 534 (1949); General Box Co. v. United States, 351 U.S. 159, 165 (1956); The Tungus v. Skovgaard, 358 U.S. 588, 596 (1959). The Court of Appeals did not pass on the correctness of the resolution of the state law problem involved in Bosch, No. 673, and it is remanded for that purpose.
The problem of what effect must be given a state trial court decree where the matter decided there is determinative of federal estate tax consequences has long burdened the Bar and the courts. This Court has not addressed itself to the problem for nearly a third of a century. In Freuler v. Helvering, 291 U.S. 35 (1934), this Court, declining to find collusion between the parties on the record as presented there, held that a prior in personam judgment in the state court to which the United States was not made a party, "[o]bviously . . . had not the effect of res judicata, and could not furnish the basis for invocation of the full faith and credit clause. . . ." At 43. In Freuler's wake, at least three positions have emerged among the circuits. The first of these holds that
It may be claimed that Blair v. Commissioner, 300 U.S. 5 (1937), dealt with the problem presently before us but that case involved the question of the effect of a property right determination by a state appellate court.
". . . if the question at issue is fairly presented to the state court for its independent decision and is so decided by the court the resulting judgment if binding upon the parties under the state law is conclusive as to their property rights in the federal tax case. . . ." Gallagher v. Smith, 223 F.2d 218, 225.
The opposite view is expressed in Faulkerson's Estate v. United States, 301 F.2d 231. This view seems to approach that of Erie R. Co. v. Tompkins, 304 U.S. 64 (1938), in that the federal court will consider itself bound by the state court decree only after independent examination of the state law as determined by the highest court of the State. The Government urges that an intermediate position be adopted; it suggests that a state trial court adjudication is binding in such cases only when the judgment is the result of an adversary proceeding in the state court. Pierpont v. C. I. R., 336 F.2d 277. Also see the dissent of Friendly, J., in Bosch, No. 673.
We look at the problem differently. First, the Commissioner was not made a party to either of the state proceedings here and neither had the effect of res judicata, Freuler v. Helvering, supra; nor did the principle of collateral estoppel apply. It can hardly be denied that both state proceedings were brought for the purpose of directly affecting federal estate tax liability. Next, it must be remembered that it was a federal taxing statute that the Congress enacted and upon which we are here passing. Therefore, in construing it, we must look to the legislative history surrounding it. We find that the report of the Senate Finance Committee recommending enactment of the marital deduction used very guarded language in referring to the very question involved here. It said that "proper regard," not finality, "should be given to interpretations of the will" by state courts and then only when entered by a court "in a bona fide adversary proceeding." S. Rep. No. 1013, Pt. 2, 80th Cong., 2d Sess., 4. We cannot say that the authors of this directive intended that the decrees of state trial courts were to be conclusive and binding on the computation of the federal estate tax as levied by the Congress. If the Congress had intended state trial court determinations to have that effect on the federal actions, it certainly would have said so — which it did not do. On the contrary, we believe it intended the marital deduction to be strictly construed and applied. Not only did it indicate that only "proper regard" was to be accorded state decrees but it placed specific limitations on the allowance of the deduction as set out in §§ 2056(b), (c), and (d). These restrictive limitations clearly indicate the great care that Congress exercised in the drawing of the Act and indicate also a definite concern with the elimination of loopholes and escape hatches that might jeopardize the federal revenue. This also is in keeping with the long-established policy of the Congress, as expressed in the Rules of Decision Act, 28 U.S.C. § 1652. There it is provided that in the absence of federal requirements such as the Constitution or Acts of Congress, the "laws of the several states . . . shall be regarded as rules of decision in civil actions in the courts of the United States, in cases where they apply." this Court has held that judicial decisions are "laws of the . . . state" within the section. Erie R. Co. v. Tompkins, supra; Cohen v. Beneficial Loan Corp., 337 U.S. 541 (1949); King v. Order of Travelers, 333 U.S. 153 (1948). Moreover, even in diversity cases this Court has further held that while the decrees of "lower state courts" should be "attributed some weight . . . the decision [is] not controlling . . ." where the highest court of the State has not spoken on the point. King v. Order of Travelers, supra, at 160-161. And in West v. A. T. T. Co., 311 U.S. 223 (1940), this Court further held that "an intermediate appellate state court . . . is a datum for ascertaining state law which is not to be disregarded by a federal court unless it is convinced by other persuasive data that the highest court of the state would decide otherwise." At 237. (Emphasis supplied.) Thus, under some conditions, federal authority may not be bound even by an intermediate state appellate court ruling. It follows here then, that when the application of a federal statute is involved, the decision of a state trial court as to an underlying issue of state law should a fortiori not be controlling. This is but an application of the rule of Erie R. Co. v. Tompkins, supra, where state law as announced by the highest court of the State is to be followed. This is not a diversity case but the same principle may be applied for the same reasons, viz., the underlying substantive rule involved is based on state law and the State's highest court is the best authority on its own law. If there be no decision by that court then federal authorities must apply what they find to be the state law after giving "proper regard" to relevant rulings of other courts of the State. In this respect, it may be said to be, in effect, sitting as a state court. Bernhardt v. Polygraphic Co., 350 U.S. 198 (1956).
We believe that this would avoid much of the uncertainty that would result from the "non-adversary" approach and at the same time would be fair to the taxpayer and protect the federal revenue as well.
The judgment in No. 240 is therefore affirmed while that in No. 673 is reversed and remanded for further proceedings not inconsistent with this opinion.
It is so ordered.
As the Court says, the issue in these cases is not whether the Commissioner is "bound" by the state court decrees. He was not a party to the state court proceedings and therefore cannot be bound in the sense of res judicata. The question simply is whether, absent fraud or collusion, a federal court can ignore a state court judgment when federal taxation depends upon property rights and when property rights rest on state law, as they do here.
Since our 1938 decision in Erie R. Co. v. Tompkins, 304 U.S. 64, an unbroken line of cases has held that the federal courts must look to state legislation, state decisions, state administrative practice, for the state law that is to be applied. See, e. g., Cities Service Oil Co. v. Dunlap, 308 U.S. 208; Bernhardt v. Polygraphic Co., 350 U.S. 198. Those were diversity cases; and in them we have never suggested that the federal court may ignore a relevant state court decision because it was not entered by the highest state court. Indeed, we have held that the federal court is obligated to follow the decision of a lower state court in the absence of decisions of the State Supreme Court showing that the state law is other than announced by the lower court. See, e. g., Fidelity Union Trust Co. v. Field, 311 U.S. 169; West v. A. T. T. Co., 311 U.S. 223; Six Companies of California v. Joint Highway District, 311 U.S. 180; Stoner v. New York Life Ins. Co., 311 U.S. 464.
It is true that in King v. Order of Travelers, 333 U.S. 153, we held that a federal court of appeals did not have to accept the decision of a state court of common pleas on a matter of state law. But that case was unique. The state court had relied upon the decision of a federal district court; the "Court of Common Pleas [did] not appear to have such importance and competence within [the State's] own judicial system that its decisions should be taken as authoritative expositions of that State's `law'" ( id., at 161); "the difficulty of locating Common Pleas decisions [was] a matter of great practical significance" ( ibid.); another state court had handed down an opinion rejecting the reasoning of the court of common pleas and espousing the reasoning of the Court of Appeals, illustrating "the perils of interpreting a Common Pleas decision as a definitive expression of [state law]" ( 333 U.S., at 162); and the interpretation of the Court of Appeals, which rejected the decision of the court of common pleas, was strongly supported by the decisions of the State Supreme Court. We stressed that our decision was not "to be taken as promulgating a general rule that federal courts need never abide by determinations of state law by state trial courts." Ibid.
Even before it was held that federal courts must apply state law in diversity cases, it was incumbent upon federal courts to take state law from state court decisions when federal tax consequences turned on state law. In Freuler v. Helvering, 291 U.S. 35, the trustee under a decedent's will had included in income distributed to the life beneficiaries amounts representing depreciation of the corpus. The life beneficiaries did not include the amounts constituting depreciation and the Commissioner asserted a deficiency. While the case was on appeal to the Board of Tax Appeals, the trustee filed an accounting in the state probate court, requesting its approval. The state court held that the life beneficiaries were not entitled to the distribution of depreciation of the corpus, and ordered that the life beneficiaries repay the trustee for the amount improperly distributed to them. In the tax litigation, the Court of Appeals ignored the state court determination on the ground that "no orders of the probate court, the effect of which would relate to what are deductions to be allowed under the national income taxing law, are conclusive and binding on the federal courts. . . ." 62 F.2d 733, 735. The Court reversed, holding that the probate court order was an order governing distribution within § 219 of the Revenue Act of 1921. It went on to say:
"Moreover, the decision of [the probate] court, until reversed or overruled, establishes the law of California respecting distribution of the trust estate. It is none the less a declaration of the law of the State because not based on a statute, or earlier decisions. The rights of the beneficiaries are property rights and the court has adjudicated them. What the law as announced by that court adjudges distributable is, we think, to be so considered in applying § 219 of the Act of 1921." 291 U.S., at 45.
The issue of the effect of a state court determination came up again in Blair v. Commissioner, 300 U.S. 5. The issue in that case was whether a beneficiary had effectively assigned income from a trust. In prior tax litigation, a federal court held that the trust was a spendthrift trust and that, therefore, the assignments were invalid and the income taxable to the beneficiary. The trustees then brought an action in the state court; the state courts determined that the trust was not a spendthrift trust and that the assignments were valid. The Board of Tax Appeals accepted the decision of the state court and rejected the Commissioner's claim that petitioner was liable for tax on the income. The Court rejected the Commissioner's argument that the trust was a spendthrift trust, noting that:
"The question of the validity of the assignments is a question of local law. . . . By that law the character of the trust, the nature and extent of the interest of the beneficiary, and the power of the beneficiary to assign that interest in whole or in part, are to be determined. The decision of the state court upon these questions is final. . . . It matters not that the decision was by an intermediate appellate court. . . . In this instance, it is not necessary to go beyond the obvious point that the decision was in a suit between the trustees and the beneficiary and his assignees, and the decree which was entered in pursuance of the decision determined as between these parties the validity of the particular assignments. Nor is there any basis for a charge that the suit was collusive and the decree inoperative. . . . The trustees were entitled to seek the instructions of the court having supervision of the trust. That court entertained the suit and the appellate court, with the first decision of the Circuit Court of Appeals before it, reviewed the decisions of the Supreme Court of the State and reached a deliberate conclusion. To derogate from the authority of that conclusion and of the decree it commanded, so far as the question is one of state law, would be wholly unwarranted in the exercise of federal jurisdiction.
"In the face of this ruling of the state court it is not open to the Government to argue that the trust `was, under the [state] law, a spendthrift trust.' The point of the argument is that, the trust being of that character, the state law barred the voluntary alienation by the beneficiary of his interest. The state court held precisely the contrary." Id., 9-10.
I would adhere to Freuler v. Helvering, supra, and Blair v. Commissioner, supra. There was no indication in those cases that the state court decision would not be followed if it was not from the highest state court.
The idea that these state proceedings are not to be respected reflects the premise that such proceedings are brought solely to avoid federal taxes. But there are some instances in which an adversary proceeding is impossible (see, e. g., Estate of Darlington v. Commissioner, 302 F.2d 693; Braverman Gerson, The Conclusiveness of State Court Decrees in Federal Tax Litigation, 17 Tax L. Rev. 545, 570-572 (1962)), and many instances in which the parties desire a determination of their rights for other than tax reasons.
Not giving effect to a state court determination may be unfair to the taxpayer and is contrary to the congressional purpose of making federal tax consequences depend upon rights under state law. The result will be to tax the taxpayer or his estate for benefits which he does not have under state law. This aspect is emphasized in Blair v. Commissioner, supra, where the Government attempted to tax the taxpayer for income to which he had no right under state law. In Second National Bank v. United States, the grandchildren's trusts will be assessed for the estate taxes, since the state court held that the proration statute applied; but the estate tax will be computed as if the proration statute did not apply — the marital deduction will be decreased and the tax increased. Or take the case where a state court determines that X does not own a house. After X dies, a federal court determines that the state court was wrong and that X owned the house, and it must be included in his gross estate even though it does not pass to his heirs. I cannot believe that Congress intended such unjust results.
This is not to say that a federal court is bound by all state court decrees. A federal court might not be bound by a consent decree, for it does not purport to be a declaration of state law; it may be merely a judicial stamp placed upon the parties' contractual settlement. Nor need the federal court defer to a state court decree which has been obtained by fraud or collusion. But where, absent those considerations, a state court has reached a deliberate conclusion, where it has construed state law, the federal court should consider the decision to be an exposition of the controlling state law and give it effect as such.
The central issue presented by these two cases is whether and in what circumstances a judgment of a lower state court is entitled to conclusiveness in a subsequent federal proceeding, if the state judgment establishes property rights from which stem federal tax consequences. The issue is doubly important: it is a difficult and intensely practical problem, and it involves basic questions of the proper relationship in this context between the state and federal judicial systems. For reasons which follow, I am constrained to dissent from the resolution reached by the Court in both cases.
It is useful first to summarize the legal and factual circumstances out of which these cases arose.
In No. 240, Second National Bank, the decedent's will and codicil provided that one-third of the residuary estate should be held in trust for the decedent's widow, who was given a general testamentary power of appointment over the corpus, and that the balance should be held in separate trusts for his nine grandchildren. The widow's trust was plainly within the terms of the marital deduction provided by § 2056(b)(5) of the Internal Revenue Code of 1954; the issue in this instance thus simply involves determination of the amount of this trust, and hence the amount of the marital deduction. Under Connecticut's tax-proration statute, Conn. Gen. Stat. Rev. § 12-401, a bequest exempt from estate tax, as here by reason of the federal marital deduction, is not reduced by any portion of such tax. Accordingly, if the proration statute is applicable to this decedent's will, the widow's trust would bear no part of the federal estate tax, and its entire burden would instead fall upon the grandchildren's trusts. The amount of the marital deduction would be correspondingly increased.
By its terms, the state proration statute is to be applied unless the "testator otherwise directs." Article I of the decedent's will provided, without apparent ambiguity, that the "provisions of any statute requiring the apportionment or proration of [estate] taxes . . . shall be without effect in the settlement of my estate." Nonetheless, the executor, petitioner here, contended to the Commissioner that the statute was applicable, and, upon receipt of the 30-day deficiency letter, applied to the Probate Court for the District of Hamden, Connecticut, for a determination that the estate taxes should be apportioned under the terms of the state statute. Notice of the application was given to the District Director of Internal Revenue, but, in accord with the Service's consistent position with reference to such state proceedings, Mim. 6134, Apr. 3, 1947, 1947 CCH Fed. Tax Rep. ¶ 6137, no appearance was entered in his behalf.
The deficiency was assessed at $1,333,194.35, plus interest. If the proration statute is applicable, as the executor has contended, the marital deduction attributable to the widow's trust would be approximately $3,600,000. If the statute is not applicable, as the Commissioner has held, the marital deduction would be approximately $1,700,000.
Apart from the executor's application, the probate court had the benefit only of argument from the guardian ad litem of the grandchildren; the guardian acknowledged that proration under the statute would place the burden of the estate tax entirely upon his wards' trusts, but nevertheless concluded that he had "no objection" to the executor's application. The court, filing a written opinion, determined that the decedent's disclaimer of the statute was ambiguous, and therefore concluded that the statute was applicable. Petitioner thereupon paid the assessed deficiency, and brought this suit for a refund. The District Court and the Court of Appeals both concluded that, because of the character of Connecticut's probate court system, the state judgment was not conclusive of the applicability of the proration statute. 222 F. Supp. 446; 351 F.2d 489.
The District Court concluded that Connecticut probate courts are not courts of records (but see Shelton v. Hadlock, 62 Conn. 143, 25 A. 483 , and 1 Locke Kohn, Connecticut Probate Practice 30 (1951)), that its decrees are without legal effect in the State's higher courts, and that their decrees are also subject to collateral attack even in another probate district. 222 F. Supp., at 457; see also 351 F.2d, at 494.
In No. 673, Estate of Bosch, the decedent created in 1930 a revocable inter vivos trust in favor of his wife, which also granted to her a general testamentary power of appointment over the corpus. In 1951, the decedent's wife, in order to take advantage of the Powers of Appointment Act of 1951, 65 Stat. 91, executed an instrument which purportedly converted the general power into a special power of appointment. Upon the decedent's death in 1957, his executor sought a marital deduction for the amount of the inter vivos trust; under § 2056 (b)(5), the trust would qualify for the deduction only if the decedent's wife held at his death a general power of appointment over the corpus.
The Commissioner, on the basis of the release signed in 1951 by the widow, disallowed the deduction, but the executor sought from the Tax Court a redetermination of the resulting deficiency. While the Tax Court proceeding was still pending, the executor petitioned in the New York Supreme Court for a determination under state law of the validity of the 1951 release. The Tax Court, with the Commissioner's assent, temporarily suspended its proceeding. In the state court, each of the three parties — the trustee, the widow, and the guardian ad litem of an infant who was a possible beneficiary — contended that the release was a nullity. The state court adopted their unanimous view. The Tax Court thereupon accepted the state trial court decision as an "authoritative exposition" of the requirements of state law. 43 T.C. 120. A divided Court of Appeals affirmed. 363 F.2d 1009.
The issue here, despite its importance in general, is essentially quite a narrow one. The questions of law upon which taxation turns in these cases are not among those for which federal definitions or standards have been provided; compare Burnet v. Harmel, 287 U.S. 103, 110; Heiner v. Mellon, 304 U.S. 271, 279; Lyeth v. Hoey, 305 U.S. 188, 194; it is, on the contrary, accepted that federal tax consequences have here been imposed by Congress on property rights as those rights have been defined and delimited by the pertinent state laws. The federal revenue interest thus consists entirely of the expectation that the absence or presence of the rights will be determined accurately in accordance with the prevailing state rules. The question here is, however, not how state law must in the context of federal taxation ordinarily be determined; it is instead the more narrow one of whether and under what conditions a lower state court adjudication of a taxpayer's property rights is conclusive when subsequently the federal tax consequences of those rights are at issue in a federal court.
The problem may not, as the Court properly observes, be resolved by reference to the principles of res judicata or collateral estoppel, see generally Cromwell v. County of Sac, 94 U.S. 351, 352-353; the Revenue Service has not, and properly need not have, entered an appearance in either of the state court proceedings in question here. Nor do the pertinent provisions of the revenue laws, or their legislative history, provide an adequate guide to the solution of the problem; the only direct reference in that lengthy history relevant to these questions is imprecise and equivocal. The cases in this Court are scarcely more revealing; they are, as Judge Friendly remarked below, "cryptic" and "rather dated." 363 F.2d 1009, 1015.
A supplementary report of the Senate Finance Committee, concerned with the legislation which eventually became the Revenue Act of 1948, said simply that "proper regard should be given to interpretations of the will rendered by a court in a bona fide adversary proceeding." S. Rep. No. 1013, Pt. 2, 80th Cong., 2d Sess., 4. This language is doubtless broadly consistent with virtually any resolution of these issues, but it is difficult to see the pertinence of the sentence's last four words if, as the Court suggests, conclusiveness was intended to be given to the State's highest court, but to none other.
It is, of course, plain that the Rules of Decision Act, 28 U.S.C. § 1652, is applicable here, as it is, by its terms, to any situation in which a federal court must ascertain and apply the law of any of the several States. Nor may it be doubted that the judgments of state courts must be accepted as a part of the state law to which the Act gives force in federal courts, Erie R. Co. v. Tompkins, 304 U.S. 64; it is not, for that purpose, material whether the jurisdiction of the federal court in a particular case is founded upon diversity of citizenship or involves a question arising under the laws of the United States. This need not mean, however, that every state judgment must be accepted by federal courts as conclusive of state law. The Court has, for example, never held, even in diversity cases, where the federal interest consists at most in affording a "neutral" forum, that the judgments of state trial courts must in all cases be taken as conclusive statements of state law; apart from a series of cases decided at the 1940 Term, the Court has consistently acknowledged that the character both of the state proceeding and of the state court itself may be relevant in determining a judgment's conclusiveness as a statement of state law. This same result must surely follow a fortiori in cases in which the application of a federal statute is at issue.
See, e. g., Maternally Yours, Inc. v. Your Maternity Shop, Inc., 234 F.2d 538; Friendly, In Praise of Erie — and of the New Federal Common Law, 39 N.Y. U. L. Rev. 383, 408, n. 122; Note, The Competence of Federal Courts to Formulate Rules of Decision, 77 Harv. L. Rev. 1084, 1087.
Fidelity Union Trust Co. v. Field, 311 U.S. 169; Six Companies of California v. Joint Highway District, 311 U.S. 180; West v. A. T. T. Co., 311 U.S. 223; and Stoner v. New York Life Ins. Co., 311 U.S. 464. See also Vandenbark v. Owens-Illinois Glass Co., 311 U.S. 538. All these cases, with the possible exception of Field, and apart from the rather different issue in Vandenbark, concerned intermediate state courts. They have been strongly and repeatedly criticized by commentators. Judge Friendly, for example, described them as "outrages," supra, at 401. See also Corbin, The Laws of the Several States, 50 Yale L. J. 762, 766-768; Clark, State Law in the Federal Courts, 55 Yale L. J. 267, 290-292; and 2 Crosskey, Politics and the Constitution 922-927 (1953). It may also be wondered whether these cases have any vitality left after King and Bernhardt, supra.
Freuler v. Helvering, 291 U.S. 35; King v. Order of Travelers, supra; Bernhardt v. Polygraphic Co., supra.
Similarly, it is difficult to see why the formula now ordinarily employed to determine state law in diversity cases — essentially that, absent a recent judgment of the State's highest court, state cases are only data from which the law must be derived — is necessarily applicable without modification in all situations in which federal courts must ascertain state law. The relationship between the state and federal judicial systems is simply too delicate and important to be reduced to any single standard. See Hill, The Erie Doctrine in Bankruptcy, 66 Harv. L. Rev. 1013; Note, The Competence of Federal Courts to Formulate Rules of Decision, 77 Harv. L. Rev. 1084. Compare, e. g., Morgan v. Commissioner, 309 U.S. 78, 80-81; Cardozo, Federal Taxes and the Radiating Potencies of State Court Decisions, 51 Yale L. J. 783. The inadequacy of this formula is particularly patent here, where, unlike the cases in which it was derived, the federal court is confronted by precisely the legal and factual circumstances upon which the state court has already passed.
Accordingly, although the Rules of Decision Act and the Erie doctrine plainly offer relevant guidance to the appropriate result here, they can scarcely be said to demand any single conclusion.
Given the inconclusiveness of these sources, it is essential to approach these questions in terms of the various state and federal interests fundamentally at stake. It suffices for present purposes simply to indicate the pertinent factors. On one side are certain of the principles which ultimately are the wellsprings both of the Rules of Decision Act and of the Erie doctrine. First among those is the expectation that scrupulous adherence by federal courts to the provisions of state law, as reflected both in local statutes and in state court decisions, will promote an appropriate uniformity in the administration of law within each of the States. Uniformity will, in turn, assure proper regard in the federal courts for the areas of law left by the Constitution to state discretion and administration, and, in addition, will prevent the incongruity that stems from dissimilar treatment by state and federal courts of the same or similar factual situations. Finally, it must be acknowledged that state courts are unquestionably better positioned to measure the requirements of their own laws; even the lowest state court possesses the tangible advantage of a close familiarity with the meaning and purposes of its local rules of law.
On the other side are important obligations which spring from the practical exigencies of the administration of federal revenue statutes. It can scarcely be doubted that if conclusiveness for federal tax purposes were attributed to any lower state court decree, whether the product of genuinely adversary litigation or not, there would be many occasions on which taxpayers might readily obtain favorable, but entirely inaccurate, determinations of state law from unsuspecting state courts. One need not, to envision this hazard, assume either fraud by the parties or any lack of competence or disinterestedness among state judges; no more would be needed than a complex issue of law, a crowded calendar, and the presentation to a busy judge of but essentially a single viewpoint. The consequence of any such occurrence would be an explication of state law that would not necessarily be either a reasoned adjudication of the issues or a consistent application of the rules adopted by the State's appellate courts.
It is difficult to suppose that adherence by federal courts to such judgments would contribute materially to the uniformity of the administration of state law, or that the taxpayer would be unfairly treated if he were obliged to act, for purposes of federal taxation, as if he were governed by a more accurate statement of the requirements of state law. Certainly it would contribute nothing to the uniformity or accuracy of the administration of the federal revenue statutes if federal courts were compelled to adhere in all cases to such judgments.
See, on the importance of uniformity in federal taxation, Hylton v. United States, 3 Dall. 171, 180; Cahn, Local Law in Federal Taxation, 52 Yale L. J. 799.
The foregoing factors might, of course, be thought consistent with a variety of disparate resolutions of the questions these two cases present. If emphasis is placed principally upon the importance of uniformity in the application of law within each of the several States, and thereby upon the apparent unfairness to an individual taxpayer if an issue of state law were differently decided by state and federal courts, it might seem appropriate to accept, in all but the most exceptional of circumstances, the judgment of any state court that has addressed the question at issue. This is the viewpoint identified with the opinion of the Court of Appeals for the Third Circuit in Gallagher v. Smith, 223 F.2d 218; it is, in addition, apparently the rule adopted today by my Brother DOUGLAS. Conversely, if emphasis is placed principally upon the hazards to the federal fisc from dubious decisions of lower state courts, it might be thought necessary to require federal courts to examine for themselves, absent a judgment by the State's highest court, the content in each case of the pertinent state law. This, as I understand it, is the rule adopted by a majority of the Court today.
In my opinion, neither of these positions satisfactorily reconciles the relevant factors involved. The former would create excessive risks that federal taxation will be evaded through the acquisition of inadequately considered judgments from lower state courts, resulting from proceedings brought, in reality, not to resolve truly conflicting interests among the parties but rather as a predicate for gaining foreseeable tax advantages, and in which the point of view of the United States had never been presented or considered. The judgment resulting from such a proceeding might well differ only in form from a consent decree. The United States would be compelled either to accept as binding upon its interests such a judgment, or to participate in every state court proceeding, brought at the taxpayer's pleasure, which might establish state property rights with federal tax consequences.
The second position, on the other hand, would require federal intervention into the administration of state law far more frequently than the federal interests here demand; absent a judgment of the State's highest court, federal courts must under this rule re-examine and, if they deem it appropriate, disregard the previous judgment of a state court on precisely the identical question of state law. The result might be widely destructive both of the proper relationship between state and federal law and of the uniformity of the administration of law within a State.
The interests of the federal treasury are essentially narrow here; they are entirely satisfied if a considered judgment is obtained from either a state or a federal court, after consideration of the pertinent materials, of the requirements of state law. For this purpose, the Commissioner need not have, and does not now ask, an opportunity to relitigate in federal courts every issue of state law that may involve federal tax consequences; the federal interest requires only that the Commissioner be permitted to obtain from the federal courts a considered adjudication of the relevant state law issues in cases in which, for whatever reason, the state courts have not already provided such an adjudication. In turn, it may properly be assumed that the state court has had an opportunity to make, and has made, such an adjudication if, in a proceeding untainted by fraud, it has had the benefit of reasoned argument from parties holding genuinely inconsistent interests.
I would therefore hold that in cases in which state-adjudicated property rights are contended to have federal tax consequences, federal courts must attribute conclusiveness to the judgment of a state court, of whatever level in the state procedural system, unless the litigation from which the judgment resulted does not bear the indicia of a genuinely adversary proceeding. I need not undertake to define with any particularity the weight I should give to the various possible factors involved in such an assessment; it suffices to illustrate the more important of the questions which I believe to be pertinent. The principal distinguishing characteristic of a state proceeding to which, in my view, conclusiveness should be attributed is less the number of parties represented before the state court than it is the actual adversity of their financial and other interests. It would certainly be pertinent if it appeared that all the parties had instituted the state proceeding solely for the purpose of defeating the federal revenue. The taking of an appeal would be significant, although scarcely determinative. The burden would be upon the taxpayer, in any case brought either for a redetermination of a deficiency or for a refund, to overturn the presumption, Welch v. Helvering, 290 U.S. 111, 115, that the Commissioner had correctly assessed the necessary tax by establishing that the state court had had an opportunity to make, and had made, a reasoned resolution of the state law issues, after a proceeding in which the pertinent viewpoints had been presented. Proceedings in which one or more of the parties had been guilty of fraud in the presentation of the issues to the state court would, of course, ordinarily be entitled to little or no weight in the federal court's determination of state law.
I recognize, of course, that this approach lacks the precision of both the contrasting yardsticks suggested by the Court and by my Brother DOUGLAS. Yet I believe that it reflects more faithfully than either of those resolutions the demands of our federal system and of the competing interests involved.
It may be doubted, however, whether this approach would actually produce serious practical disadvantages. It is essentially the standard which has been embodied in the Treasury Regulations since 1919, see now Treas. Reg. §§ 20.2053-1(b)(2), 20.2056(e)-2(d)(2), and which was urged before this Court in these cases by counsel for the United States. It is, moreover, similar to the standards employed in various opinions by a number of the courts of appeals. See, e. g., Saulsbury v. United States, 199 F.2d 578; Brodrick v. Gore, 224 F.2d 892; In re Sweet's Estate, 234 F.2d 401; Old Kent Bank Trust Co. v. United States, 362 F.2d 444. See also Cahn, supra, at 818-819; Braverman Gerson, The Conclusiveness of State Court Decrees in Federal Tax Litigation, 17 Tax L. Rev. 545. If any practical difficulties actually attend this standard, they have apparently not, despite its wide use, yet appeared.
I would apply these general principles to the present cases in the following manner. In No. 240, the Court of Appeals agreed with the District Court that "it was unnecessary" to make a finding on whether the proceedings in the Connecticut probate court were collusive or "nonadversary," since the decrees of the probate court could "`under no circumstances'" be considered binding. 351 F.2d 489, 494. I would therefore vacate the judgment of the Court of Appeals and remand the cause for further proceedings in accordance with the views expressed herein.
In No. 673, the Court of Appeals apparently concluded that, absent fraud or collusion, any state court proceeding which terminates in a judgment binding on the parties as to their rights under state law is also conclusive for purposes of federal taxation. 363 F.2d 1009, 1014. I would therefore reverse the judgment of the Court of Appeals, and again would remand the cause for further proceedings consistent with the views expressed in this opinion.
While I join the dissenting opinion of my Brother HARLAN, I believe it appropriate to add these few comments. As my Brother HARLAN states, in a case in which federal tax consequences depend upon state property interests, a federal court should accept the final conclusion of a competent state court, assuming that such a conclusion is an adjudication of substance arrived at after adversary litigation and on the basis of the same careful consideration that state courts normally accord cases involving the determination of state property interests. The touchstone of whether the state proceeding was "adversary" is not alone entirely satisfactory. I think that this concept has been helpfully embellished by Judge Raum of the United States Tax Court in the Bosch case, 43 T.C. 120, 123-124. Judge Raum suggests that among the factors to be considered in determining whether the decision of the state court is to be accepted as final for federal tax purposes are the following: whether the state court had jurisdiction, and whether its determination is fully binding on the parties; whether, in practice, the decisions of the state court have precedential value throughout the State; whether the Commissioner was aware of the state proceedings and had an opportunity to participate; whether the state court "rendered a reasoned opinion and reached a `deliberate conclusion', Blair v. Commissioner, 300 U.S.  at p. 10"; whether the state decision has potentially offsetting tax consequences in respect of the state court litigant's federal taxes; and, in general, whether the state court decision "authoritatively determined" future property rights, and thus, as Judge Raum stated, "provided more than a label for past events. . . ."