Estate of McLendon v. Commissioner

E. GRADY JOLLY, Circuit Judge: The only question remaining in this appeal 1 is whether Gordon B. McLendon was sufficiently close to death on March 5, 1986, to require him to depart from the actuarial tables published by the Commissioner of Internal Revenue (the “Commissioner”) in valuing a remainder interest and related annuity. The Tax Court determined that he was, from which final decision McLendon’s Estate appeals. We reverse. I Although this ease raises several contentious legal questions, the underlying facts are not in serious dispute. Through various partnership interests, McLendon was the principal owner and director of a vast broadcasting and entertainment empire. His interests ranged from the 458-station Liberty Broadcasting System to numerous individual radio stations, television stations, and movie theaters. Over his life time, McLendon became a very wealthy man. Mortality hovers over the castle as well as the cottage, however, and in May 1985 McLendon was diagnosed with esophageal cancer. Although his condition initially improved following radiation therapy, the cancer recurred in September. At this point, McLendon’s cancer was categorized as “systemic” — the most severe of three types of cancer growth. There is no dispute that the cancer was very likely terminal from this point forward, with a 2-3% overall survival rate. In particular, any remissions achieved after this point were generally expected by McLendon’s doctors to be temporary. Nonetheless, from October 1985 through March 1986, McLendon received six courses of chemotherapy at M.D. Anderson’s world-renowned cancer treatment facility in Houston, Texas. On December 3, 1985, after three courses of chemotherapy, McLendon’s doctor wrote on his discharge summary: The patient had an esophagogastroduode-noscopy on November 26, 1985, and it showed complete endoscopic remission *1019 confirmed by multiple biopsies of the affected area. Despite this upbeat news, on December 5, 1985, McLendon attempted suicide by shooting himself in the head with a handgun. A suicide note reflected his belief that he would eventually succumb to the cancer and his desire not to prolong the suffering of his family. After being hospitalized for over a month for treatment of injuries from the failed suicide, McLendon began a fourth course of chemotherapy. He returned home in late January 1986 and began to receive periodic in-home examinations and treatment from a Dr. Gruebel. Her impression at the time was that he was doing well. In early February, McLendon fell at home and was admitted to the hospital for treatment of his injuries. On February 14, while hospitalized, McLendon purportedly dictated 2 a letter to Dr. Freireich, his oncologist, which evidenced a renewed sense of confidence. McLendon stated that he was feeling much better even though the chemotherapy was “very, very debilitating.” Stating that he was “beginning to make plans for the rest of [his] life,” McLendon inquired specifically about his “total remission” and prognosis for the future, and asked whether he could “make long term plans.” Dr. Freireich responded on February 19. Advising against further surgical procedures, he noted: The objective evidence that we have has failed to demonstrate any residual disease. This includes endoscopy with biopsies of the esophagus which have proven to be negative on several occasions and the repeated x-ray examinations by CT scan which fail to reveal any evidence of residual malignancy. [By] clinical and laboratory objective criteria, the present condition of your illness must be characterized as “complete remission.” The word remission is used advisedly, because the risk of recurrence is still much in the picture. On the other hand patients who are cured of their disease are exclusively drawn from the population of patients who have a “complete remission.” To state that positively, you are certainly a candidate for long term control which fulfills medical and lay criteria for curability. Unfortunately the maturity and quantity of our clinical data does not permit good estimates of the risk of recurrence in your specific instance. It is therefore necessary for me as a physician, to advise you of the risk that the disease might recur, but to state frankly and without hesitation that the possibility that your disease has been permanently eradicated is definite and significant and in my professional opinion, should form the basis for your planning for the future. At the end of February, McLendon returned home under twenty-four hour care from a staff of private duty nurses. Notes taken by these nurses show that during the period from March 2 through March 5, McLendon was able to take short walks and perform minor tasks, but was at times sick to his stomach, was in constant need of pain medication, and was receiving artificial sustenance to ensure proper caloric intake. McLendon was examined at home on March 5 by the optimistic Dr. Gruebel. It was her impression at that time that McLendon was “markedly improved” and in the best condition since he had come into her care in January. The Commissioner subsequently presented undisputed expert testimony, however, that McLendon’s chances of surviving for more than one year from this date were approximately 10 percent. This estimate was based principally on the likelihood of recurrence in a case like McLendon’s. On March 5, McLendon entered into a private annuity transaction with his son and the newly minted McLendon Family Trust. This transaction involved the transfer of remainder interests in McLendon’s partnership holdings to his son and the Trust in exchange for $250,000 and an annuity to be paid to McLendon for life. The amount of the annuity was set such that its aggregate present value would equal the present value of the remainder interests. In valuing the remainder interests and the annuity, the parties referred to the Commissioner’s actuarial ta *1020 bles for life expectancy then contained in Treas. Reg. § 25.2512-5(f). McLendon was sixty-five years old on March 5, 1986, resulting in an actuarial life expectancy of fifteen years from that date. Based on this figure, the parties ultimately determined that the remainder interests had a value of $5,881,-695, 3 and that the annuity would need to be $865,332 in order to match. In late March, McLendon completed his final course of chemotherapy. In May, tests revealed a major recurrence of the cancer. Treatments were discontinued within a few weeks, and McLendon died at home on September 14. From the time that he was first admitted to M.D. Anderson in October 1985 until his death, McLendon survived longer than 75% of patients diagnosed with esophageal cancer. II McLendon’s estate tax return relied on a presumption that he had received an adequate and full consideration for the assets transferred in the private annuity transaction. The Commissioner disagreed with this presumption, taking issue with both the use of the actuarial tables and certain substantive aspects of the valuation of the partnership interests. With regard to the actuarial tables, the Commissioner took the position that McLen-don’s life expectancy was sufficiently predictable on March 5, 1986, to make their use unnecessary and erroneous. Based on the medical evidence, the Commissioner further found that McLendon’s actual life expectancy on this date was less than one year. Because this was significantly less than the fifteen-year figure used by the parties, the Commissioner concluded that the remainder interests had been so undervalued, and the annuity so overvalued, that the March 5 transfer had not been for an adequate and full consideration. As such, the Commissioner declared several million dollars in gift and estate tax deficiencies based on McLendon’s erroneous use of the actuarial tables. Additional deficiencies were declared based on the substantive valuation issues. McLendon’s Estate took this dispute to the Tax Court, where the issues were reduced by joint stipulation to six discrete questions. One of these questions was whether it was proper for McLendon to apply the actuarial tables to determine his life expectancy in valuing the remainder interests and annuity. The rest of the questions concerned the substantive aspects of the valuation of the partnership interests. On September 30, 1993, the Tax Court issued its first opinion in this case, generally agreeing with the Commissioner and imposing $12.5 million in additional gift and estate taxes. Of significance to the instant appeal, the Tax Court held that use of the actuarial tables was improper because McLendon’s life expectancy was reasonably predictable at the time the private annuity transaction occurred, being approximately one year. McLendon’s Estate appealed the Tax Court’s ruling to this court. Estate of McLendon v. Commissioner of Internal Revenue, No. 94-40584, 77 F.3d 477 (5th Cir. 1995). In an unpublished opinion, the panel reversed the Tax Court on the substantive valuation questions, but remanded as to use of the actuarial tables. Writing for the court, Judge Jones stated that: [W]e are unable to discern whether the Tax Court followed Revenue Ruling 80-80 or found reason to depart from it [in resolving the actuarial table question]. The Tax Court’s opinion is both ambiguous and ambivalent regarding the revenue ruling, as it holds that Gordon had a life expectancy of one year, a finding that would suggest to us under the express language of the revenue ruling that death was not clearly imminent. We must remand for the court to clarify its position with regard to the applicability of Revenue Ruling 80-80 so that we will have a sounder basis for appellate review. On July 8, 1996, the Tax Court issued its second opinion. It held that, although neither party had argued a position inconsistent with Rev. Rui. 80-80, the court had not felt obliged to follow that ruling, and had instead applied a standard gleaned from prior case *1021 law. It noted, however, that the result would have been the same under the ruling anyway. In the light of this clarification, McLendon’s Estate now continues its appeal of the Tax Court’s determination that use of the actuarial tables was improper. III We review a decision of the Tax Court applying the same standards used in reviewing a decision of the district court: Questions of law are reviewed de novo; findings of fact are reviewed for clear error. Estate of Hudgins v. Commissioner of Internal Revenue, 57 F.3d 1393, 1396 (5th Cir.1995). IV As the prior panel foresaw, the remainder of this case turns on the applicability of Rev. Rui. 80-80. Because we hold that the ruling provides the legal test applicable to McLen-don’s situation, we find that his use of the actuarial tables was proper. A The controversy in this case ultimately stems from 26 U.S.C. §§ 2036(a) and 2512(b). Under § 2036(a), a decedent’s gross estate for estate tax purposes is defined to include any property transferred by him in which he retained a life estate, “except in ease of a bona fide sale for an adequate and full consideration in money or money’s worth.” Similarly, under § 2512(b), a taxable gift is defined as a transfer of property “for less than an adequate and full consideration.” Here, the transfer in question was the March 5 exchange of the partnership remainder interests for the cash and annuity. As the parties concede, the question whether that transfer was for “an adequate and full consideration” turns on the proper valuation of the remainder interests and the annuity. At the time of the events in this case, Treas. Reg. § 25.2512-5 provided that “the fair market value of annuities, life estates, terms for years, remainders, and reversions transferred after November 30,1983, is their present value determined in this section.” Because the economic present value of these assets is dependent upon the predicted length of a measuring life, the regulation goes on to provide actuarial tables for life expectancy and instructions for using them to arrive at valuations of the assets in question. There is no dispute in this case over the valuation formulas contained in the regulation. The parties concede that the only question is whether the circumstances of McLendon’s case allowed him to use a life expectancy figure derived from the tables of § 25.2512-5, or instead required him to use some other method to determine his “actual” life expectancy. If use of the actuarial tables was proper, then the parties agree that the values calculated by McLendon were correct, and that the cash and annuity were adequate consideration for the remainder interests. If, on the other hand, use of the actuarial tables was improper, then the parties agree that the remainder interests had a much higher value, and the annuity a much smaller value, such that the cash and annuity were not adequate consideration. The sole question before this court, then, is whether McLendon was allowed to follow the express language of Treas. Reg. § 25.2512-5 and use its actuarial tables in valuing the remainder interests and annuity. B This question is less straightforward than it might seem. Despite their apparently clear command, Treas. Reg. § 25.2512-5 and its predecessors have not always been vigorously enforced by the courts. In particular, in Miami Beach First National Bank v. United States, 443 F.2d 116, 119-20 (5th Cir. 1971), this court held that “where there is sufficient evidence regarding the actual life expectancy of a life tenant, the presumptive correctness of the Treasury tables will be overcome.” 4 *1022 Based on Miami Beach First National Bank and other cases of departure, 5 the Commissioner issued various revenue rulings over the years 6 that attempted to clarify his position with regard to use of the tables. At the time of the events in this case, the effective ruling was Rev. Rul. 80-80. It provides, in relevant part: The actuarial tables in the regulations are provided as an administrative necessity, and their general use has been readily approved by the courts. The actuarial tables are not based on data that exclusively involve persons of “good” or “normal” health. They reflect the incidence of death by disease and illness as well as by accident. The actuarial tables are properly applicable to the vast majority of individual life interests, even though the health of a particular individual is obviously better or worse than that of the “average” person of the same age and sex. Occasionally, however, the actual facts of an individual’s condition are so exceptional as to justify departure from the actuarial tables. In view of recent case law, the resulting principle is as follows: the current actuarial tables in the regulations shall be applied if valuation of an individual’s life interest is required for purposes of the federal estate or gift taxes unless the individual is known to have been afflicted, at the time of the transfer, with an incurable physical condition that is in such an advanced stage that death is clearly imminent. Death is not clearly imminent if there is a reasonable possibility of survival for more than a very brief period. For example, death is not clearly imminent if the individual may survive for a year or more and if such a possibility is not so remote as to be negligible. Rev. Rul. 80-80, 1980-1 C.B. 194 (emphasis added, citations omitted). McLendon’s Estate argues that Rev. Rul. 80-80 clearly allows his use of the tables. In this regard, the Estate notes that the undisputed testimony of the Commissioner’s own expert was that McLendon had a 10 percent chance of surviving for a year or more on March 5, 1986. As such, the Estate concludes that McLendon’s possibility of surviving for a year or more from that date was not so remote as to be negligible, and that he therefore was permitted and required to use the tables under the clear terms of the ruling. Although the Commissioner maintains that this court is not bound to follow Rev. Rul. 80-80, he also purports to take the position that the ruling does not mandate the result indicated by the Estate. The Commissioner *1023 argues that, although the ruling is a correct statement of the law, it cannot be taken at face value, and must be interpreted in the light of Miami Beach First National Bank The Commissioner contends that under this reading McLendon’s use of the tables was inappropriate since there was “sufficient evidence” of his actual life expectancy on March 5,1986. If Rev. Rul. 80-80 does govern this case, we, like the earlier panel of this court, find it undeniable that it supports the Estate’s position. The ruling states a clear standard, expressed in language and example unneedful of further interpretation, and we are convinced that the 10 percent figure is sufficient to satisfy it. Whatever “negligible” might mean in a closer ease, we are certain that it does not refer to a one-in-ten chance. 7 As such, McLendon’s use of the tables was clearly proper under the ruling. The question, then, is whether Rev. Rul. 80-80 states the legal test applicable to McLendon’s situation. If it does, then it is clear that McLendon’s use of the actuarial tables was proper. The Tax Court ultimately chose not to apply Rev. Rul. 80-80 to this case. 8 This choice was a purely legal decision, and is thus reviewed de novo. C We note at the outset that the Tax Court has long been fighting a losing battle with the various courts of appeals over the proper deference to which revenue rulings are due. 9 Whereas virtually every circuit recognizes some form of deference, 10 the Tax Court stands firm in its own position that revenue rulings are nothing more than the legal contentions of a frequent litigant, undeserving of any more or less consideration than the conclusory statements in a party’s brief. 11 Although the Supreme Court has not spoken definitively on the subject, its recent jurisprudence tends to support the view that the courts owe revenue rulings a bit more deference than the Tax Court would have us believe. 12 Still, revenue rulings are *1024 odd creatures unconducive to precise categorization in the hierarchy of legal authorities. They are clearly less binding on the courts than treasury regulations or Code provisions, but probably (and in this circuit certainly) more so than the mere legal conclusions of the parties. Apart from that, little can be said with any certainty, and in the absence of a definitive statement from on high, the Tax Court continues its crusade to ignore them in toto. This bit of background explains a great deal with regard to the posture of this case. In support of its general position on deference, the Tax Court went to great lengths to avoid applying Rev. Rui. 80-80 to McLen-don’s situation. The earlier panel of this court noticed this slight, and asked the Tax Court if it really wanted an open confrontation on the issue. Sticking to its guns, the Tax Court replied that it did. The result was the instant appeal. As it turns out, however, this case does not require us to step squarely into the fray. Most questions of deference to a revenue ruling involve an argument by the taxpayer that a particular ruling is contrary to law. Here, however, the argument to ignore or minimize the effect of Rev. Rui. 80-80 comes from the Commissioner, the very party who issued the ruling in the first place. 13 In such a situation, this circuit has a well established rule that is sufficient to resolve this case without probing the penumbrae of the general deference question. In Silco, Inc. v. United States, 779 F.2d 282, 286 (5th Cir.1986), we held that a taxpayer was entitled to rely on the legal standard implied by two revenue rulings extant at the time of his transaction, even though they had been subsequently abrogated. In reaching this conclusion, we noted that: Treas. Reg. § 601.601(e) provides that taxpayers may generally rely on published revenue rulings in determining the tax treatment of their own transactions, if the facts and circumstances of their transactions are substantially the same as those that prompted the ruling. Id. at 286. 14 Because the statute, regulations, and case law were less than clear at the time of the taxpayer’s transaction, we found that the rulings “provide[d] the only insight available to [the] taxpayer at the time of [his] transaction as to the conceptual approach the [Commissioner] would use,” and that the Commissioner acted improperly in subsequently applying a different test to that taxpayer. Id. at 287. Silco stands for the proposition that the Commissioner will be held to his published rulings in areas where the law is unclear, and may not depart from them in individual cases. Furthermore, under Silco the Commissioner may not retroactively abrogate a ruling in an unclear area with respect to any taxpayer who has relied on it. 15 *1025 Applying Silco to this ease, it quickly becomes clear that Rev. Rul. 80-80 must govern our decision. McLendon went to great lengths to structure his transaction to comply with applicable law, 16 and the Commissioner does not dispute that in so doing McLendon expressly relied on Rev. Rul. 80-80’s clarification of the admittedly murky area of future and dependent interest valuation. The Commissioner ignored the clear language of his own ruling in declaring deficiencies, and it is precisely this kind of tactic that Silco declares to be intolerable. Because McLendon was entitled to rely on Rev. Rul. 80-80, the Tax Court was not at liberty to disregard it. Its decision to do so was error, and we reverse on that basis. Furthermore, because the application of Rev. Rul. 80-80 clearly sustains the Estate’s position, we need not remand yet again for further proceedings. Consistent with our discussion of the application of the ruling above, we render for McLendon’s Estate. V Where the Commissioner has specifically approved a valuation methodology, like the actuarial tables, in his own revenue ruling, he will not be heard to fault a taxpayer for taking advantage of the tax minimization opportunities inherent therein. Here, the Commissioner had no right to ignore Rev. Rui. 80-80 and the Tax Court was bound to apply it consistent with McLendon’s right of reliance. The Tax Court’s manifest failure to apply the ruling was clearly wrong, and, accordingly, we REVERSE its judgment and RENDER for the Estate. REVERSED and RENDERED. Footnotes [1] Earlier, another panel of this court decided all of the other issues and remanded as to this question. See Estate of McLendon v. Commissioner of Internal Revenue, No. 94-40584, 77 F.3d 477 (5th Cir.1995)(unpublished). [2] This letter was not signed by McLendon, but did carry the initials of Billie P. Odom, his personal secretary. [3] Based on a value of $18,363,970 for the partnership interests themselves. [4] This tendency of courts to ignore the regulations in certain cases stems from the somewhat precarious position of the tables in the statutory framework. In this context, the Internal Revenue Code seeks only to assign "value” to various things for tax purposes. As even the Commissioner concedes, the use of actuarial tables does not result in particularly accurate measurements of actual value in individual cases. Because of the difficulty of computing the actual values of *1022 future and dependent interests, however, the Supreme Court recognized long ago that the use of actuarial tables was a necessary compromise. The Court noted that inaccuracies would prevail in individual cases, but concluded that they would cancel out in the aggregate, and that the tables were simply an administrative necessity. See Ithaca Trust Co. v. United States, 279 U.S. 151, 155, 49 S.Ct. 291, 291-92, 73 L.Ed. 647 (1929); Simpson v. United States, 252 U.S. 547, 550, 40 S.Ct. 367, 367-68, 64 L.Ed. 709 (1920). For this reason, the actuarial tables are tolerated. Implicit in this toleration, however, is the idea that the tables need not be resorted to where the "administrative necessity” does not exist. In particular, where the facts and circumstances are such that an actual value can be calculated in a suitably reliable way, use of the tables would seem to not be required. This is ultimately the logic underlying Miami Beach First National Bank and its progeny. [5] Cases, it might be noted, where the courts’ decisions to allow departure from the harshness of the tables were without exception favorable to the taxpayers in ultimate result. In addition to Miami Beach First National Bank, see, e.g., Estate of Jennings v. Commissioner of Internal Revenue, 10 T.C. 323, 1948 WL 147 (1948) (larger deduction allowed for the charitable gift of a remainder interest because the remainder was properly valued according to the shorter actual expected length of the measuring life rather than the longer length derived from the tables). Here, of course, the Commissioner seeks departure at the taxpayer’s expense. This distinction is not wholly irrelevant, as we shall see. [6] See, e.g., Rev. Rul. 80-80, 1980-1 C.B. 194; Rev. Rul. 66-307, 1966-2 C.B. 429. Subsequent to the events in this case, the Commissioner abandoned this effort and created regulations to govern the applicability of the actuarial tables. See Treas. Reg. §§ 1.7520-3(b)(3) (income tax), 20.7520-3(b)(3) (estate tax), and 25.7520-3(b)(3) (gift tax). Rev. Rul. 80-80, the last of the old line, was revoked in favor of the new regulations, but only for tax years after 1995. See Rev. Rul. 96-3, 1996-1 C.B. 348. [7] Indeed, faced with the clear text of the ruling at oral argument, the Commissioner was unable to come up with any definition of "negligible” that would embrace McLendon’s situation. This silent exclamation underscores the meritless nature of the Commissioner’s argument under the ruling. The Commissioner's silence may have been prompted by the fact that he has consistently defined the phrase "not so remote as to be negligible” to mean “less than 5 percent” in other areas of estate tax. See, e.g., Treas. Reg. §§ 26.2612-1 (b)(l)(iii), 26.2612-1 (d)(2)(ii) & 26.2632-1 (c)(2)(b) (generation-skipping transfer tax); Rev. Rul. 85-23, 1985-1 C.B. 327 (charitable deduction); Rev. Rul. 78-255, 1978-1 C.B. 294 (same); Rev. Rul. 77-374, 1977-2 C.B. 329 (same); Rev. Rul. 70-452, 1970-2 C.B. 199 (same). Although this might not be enough to establish that 5 percent is indeed the correct figure, it would be sufficient to estop the Commissioner from arguing otherwise in this case, as we shall soon see. [8] The Tax Court’s completely unpersuasive alternative holding that Rev. Rul. 80-80 also supports its result founders for the same reasons as the Commissioner’s arguments, and would be reversible error under any standard of review. For purposes of this opinion, we need only address the core of the Tax Court’s reasoning, i.e., that Miami Beach First National Bank, not Rev. Rul. 80-80, provides the legal test applicable to McLendon's case. [9] See generally Linda Galler, Judicial Deference to Revenue Rulings: Reconciling Divergent Standards, 56 Ohio St. L.J. 1037, 1059-74 (1995). [10] See, e.g., Amato v. Western Union International, Inc., 773 F.2d 1402, 1411-12 (2d Cir.1985); Gillis v. Hoechst Celanese Corp., 4 F.3d 1137, 1145 (3d Cir. 1993); Foil v. Commissioner of Internal Revenue, 920 F.2d 1196, 1201 (5th Cir. 1990); Threlkeld v. Commissioner of Internal Revenue, 848 F.2d 81, 84 (6th Cir.1988); Walt Disney Inc. v. Commissioner of Internal Revenue, 4 F.3d 735, 740 (9th Cir. 1993). In this circuit, revenue rulings are generally " 'given weight as expressing the studied view of the agency whose duty it is to carry out the statute.’ ” Foil, 920 F.2d at 1201 (quoting United States Trust Co. v. Internal Revenue Service, 803 F.2d 1363, 1370 n. 9 (5th Cir.1986)). Of course, any deference extended to a revenue ruling evaporates in the face of clear and contrary statutory language. Foil, 920 F.2d at 1201. [11] See, e.g., Pasqualini v. Commissioner of Internal Revenue, 103 T.C. 1, 8 n. 8, 1994 WL 371893 (1994); Exxon Corp. v. Commissioner of Internal Revenue, 102 T.C. 721, 726 n. 8, 1994 WL 243435 (1994); Spiegelman v. Commissioner of Internal Revenue, 102 T.C. 394, 405, 1994 WL 67785 (1994); Rath v. Commissioner of Internal Revenue, 101 T.C. 196, 205 n. 10, 1993 WL 338664 (1993). [12] See, e.g., Davis v. United States, 495 U.S. 472, 110 S.Ct. 2014, 109 L.Ed.2d 457 (1990). Note, *1024 however, that the Court in Davis was conspicuously silent as to the applicability of Chevron deference to revenue rulings. See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Given the context, this omission cautions against placing too much reliance on revenue rulings as administrative legal interpretations, and the position of the Tax Court is not without some merit. [13] The Commissioner’s position is not entirely clear in this case. He purports to maintain that Rev. Rui. 80-80 is an accurate statement of the law, yet would prefer the court decide the case based on the rule of Miami Beach First National Bank. This rule, he implies, is the same as that of Rev. Rui. 80-80. The Commissioner cannot eat his cake and have it too. As we explained above, Rev. Rui. 80-80 is unambiguous in its support for the Estate’s position. To the extent that he argues for a rule inconsistent with the ruling’s clear language, we construe the Commissioner's position to be that the ruling should not apply. [14] Treas. Reg. § 601.601(e) states: “Taxpayers generally may rely upon Revenue Rulings published in the Bulletin in determining the tax treatment of their own transactions_” Although not cited therein, Silco also finds support in Treas. Reg. § 601.601(d), which provides that revenue rulings "are published to provide precedents to be used in the disposition of other cases.” [15] This latter portion of Silco might be read to be in conflict with the Supreme Court’s well established rule that the Commissioner may retroactively revoke certain revenue rulings, even where taxpayers may have relied on them to their detriment. See Automobile Club of Michigan v. Commissioner of Internal Revenue, 353 U.S. 180, 183— 84, 77 S.Ct. 707, 709-10, 1 L.Ed.2d 746 (1957) (Brennan, J.); Dixon v. United States, 381 U.S. *1025 68, 72-73, 85 S.Ct. 1301, 1304-05, 14 L.Ed.2d 223 (1965) (Brennan, J.). For a number of reasons, however, we perceive no conflict. First, the Automobile Club rule applies only where the Commissioner revokes a prior ruling that is contrary to the Internal Revenue Code. This was not the case in Silco, nor is it the case here. The Silco rule is expressly limited to areas where the Code does not provide a clear answer. Second, Silco is grounded on the Commissioner’s invitation to taxpayers to rely on his revenue rulings as set out in Treas. Reg. § 601.601(e), a factor not present in the Automobile Club or Dixon cases. The essence of the Silco rule is that traditional notions of equity and fair play prevent the Commissioner from changing his position after inviting reliance with his own regulations. Finally, even if there were some tension between Silco and Automobile Club, we would be bound in this case by our past circuit precedent. "One panel of this Court may not overrule another (absent an intervening decision to the contrary by the Supreme Court or the en banc court ... ).” Hogue v. Johnson, 131 F.3d 466, 491 (5th Cir. 1997) (Garwood, J.) (emphasis added). See also United States v. McPhail, 119 F.3d 326, 327 (5th Cir. 1997) (Smith, J., dissenting), and cases cited therein. Supposed conflicts with prior Supreme Court precedent are grist for the en banc mill, but not for ad hoc panel revision. See 5th Cir. IOP to Fed. R.App. P. 35. For all of these reasons, we are content that Silco continues to be good law. [16] As indeed he should have. " 'Over and over again courts have said that there is nothing sinister in so arranging one's affairs as to keep taxes as low as possible. Everybody does so, rich or poor; and all do right, for nobody owes any public duty to pay more than the law demands: taxes are enforced exactions, not voluntary contributions. To demand more in the name of morals is mere cant.’ ’’ Commissioner of Internal Revenue v. First Security Bank of Utah, 405 U.S. 394, 398 n. 4, 92 S.Ct. 1085, 1089 n. 4, 31 L.Ed.2d 318 (1972) (quoting Learned Hand's celebrated dissent in Commissioner of Internal Revenue v. Newman, 159 F.2d 848, 850-51 (2d Cir. 1947)).