Second Circuit Resolves Standard of Review on Tax Court Appeals in Transferee Liability Case (11/18/13)

Peter Reilly has posted a good discussion of the Second Circuit decision in Diebold Found. v. Commioner,  736 F.3d 172 (2d Cir. 2013), here.  Peter's blog entry Charitable Foundation Haunted by 1999 Corporate Tax Assessment (Forbes 11/17/12), here.

The substantive decision in the case deals with the application of Section 6091 transferee liability.  The case gets into some esoterica of transferee liability, so I will leaves readers of the opinion and Peter's blog to ferret that out.  I will, however, just offer a gratuitous comment that the sophisticated players in the underlying game -- generically referred to as Midco transactions -- knew that when all the shuffling was over, the IRS would be left holding the bag for a large amount of tax dollars that was due and that those tax dollars not paid would be shared among the players in various ways intended to disguise the fact that they had just participated in key steps to evade federal taxes.  Evade may be a strong word here, but for the level of sophistication -- lawyers involved -- by the players I have observed in the game, they knew -- certainly should have known -- that was the consequence of their participation in the Midco game.  I think the Second Circuit gets that point and is not too bashful to say so.  In this regard, Calvin Johnson, UT Law Professor, is quoted as saying that the shareholders (including the Diebold Foundation) [a]s a matter of economics, * * * got a price for their shares that included, by my estimates, 85 percent of the value of the tax evaded.  Andrew Velarde, Second Circuit Holding on Midco Acquisitions Seen as Big Win for Government, 2013 TNT 223-3 (11/19/13).

Moving on, at the bottom of his blog, Peter addresses the procedural issue that the Second Circuit resolves at the threshold in reaching the substantive issue it addressed.  That procedural issue is the appropriate standard of review for appeals from the Tax Court.  Section 7482(a)(1), here, confers jurisdiction upon the courts of appeals to review decisions from the Tax Court "in the same manner and to the same extent as decisions of the district courts in civil actions tried without a jury."  Pretty straight-forward.  But the Second Circuit had to correct an error of its own making in order to calibrate the right standard for "mixed questions of law and fact."  The Court's discussion of the problem and its resolution is relative short, so I just cut and paste it.  However, in order to cut out some of the "noise," I omit most of the citations and some quotations marks.

In an appeal from the Tax Court, it is without dispute in this Circuit that we review legal conclusions de novo and findings of fact for clear error. While we have previously held the standard of review for mixed questions of law and fact to be one for clear error, all Courts of Appeals are to "review the decisions of the Tax Court . . . in the same manner and to the same extent as decisions of the district courts in civil actions tried without a jury." 26 U.S.C. § 7482(a)(1). Our case law enunciating the standard of review for mixed questions of law and fact in an appeal from the Tax Court is in direct tension with this statutory mandate. Following a civil bench trial, we review a district court's findings of fact for clear error, and its conclusions of law de novo; resolutions of mixed questions of fact and law are reviewed de novo to the extent that the alleged error is based on the misunderstanding of a legal standard, and for clear error to the extent that the alleged error is based on a factual determination. Two recent panels of our Court have recognized this contradiction between our case law and 26 U.S.C. § 7482(a)(1) but did not resolve the tension, as they determined that under either standard of review the outcome in the particular case would be the same. In the instant case, the standard of review affects the outcome, so our Court can avoid the question no longer. 

The standard that mixed questions of law and fact are reviewed under a clearly erroneous standard when we review a decision of the Tax Court was established in this Circuit's jurisprudence in Bausch & Lomb Inc. v. Comm'r, 933 F.2d 1084, 1088 (2d Cir. 1991). Bausch & Lomb imported the standard from the Seventh Circuit, which, in Eli Lilly & Co. v. Comm'r, 856 F.2d 855, 861 (7th Cir. 1988), held the clearly erroneous standard to be applicable. Eli Lilly in turn relied upon another Seventh Circuit case, Standard Office Bldg. Corp. v. United States, 819 F.2d 1371, 1374 (7th Cir. 1987), a tax case on review from the district court. None of these decisions mention 26 U.S.C. § 7482(a)(1), which has been a part of the Internal Revenue Code since 1954. In Standard Office Building, the Seventh Circuit indicated that one of the open questions in the appeal was "the kind of 'mixed' question of fact and law . . . that, in this circuit at least, is governed by the clearly-erroneous standard." Id. (emphasis added). That court then cited a handful of cases from their circuit that stated this standard from cases reviewing the decision of a district court. The Seventh Circuit uses the clearly erroneous standard of review for mixed questions of law and fact when reviewing both decisions of the Tax Court and those of the district courts. Its standard is thus not in tension with 26 U.S.C. § 7482(a)(1), unlike this Court's.

Quoting Eli Lilly approvingly, in Bausch & Lomb, this Court indicated, "'We are unaware of any decision discussing the standard that governs appellate review of a Tax Court's [determination].'" Bausch & Lomb, 933 F.2d at 1088 (quoting Eli Lilly, 856 F.2d at 860-61). It was certainly the case that no decision at that time discussed the standard for such appellate review, but the statute which governs our Court's review of Tax Court decisions set out a mandatory standard, tied to the level of review in appeals on review from a district court. 26 U.S.C. § 7482(a)(1). Once imported from the Seventh Circuit, this standard for mixed questions of law and fact, which stands at odds with our standard for such review of district court decisions, was propagated again in Merrill Lynch & Co. v. Comm'r, 386 F.3d 464, 469 (2d Cir. 2004) (citing Bausch & Lomb, 933 F.2d at 1088), and again in Wright, 571 F.3d at 219 (2d Cir. 2009) (citing Merrill Lynch, 386 F.3d at 469; Bausch & Lomb, 933 F.2d at 1088). These three cases make up the bulk, if not the entirety, of the citations for this standard in subsequent decisions of this Court.  

We now conclude that this standard of review was adopted in error. n7 As all Article III courts, with the exception of the Supreme Court, are solely creatures of statute, see U.S. Const. art. III; 28 U.S.C. §§ 1-463, the statute must be determinative in this case. Moreover, there is no reason to review the Tax Court under a different standard than a district court, as "its relationship to us [is] that of a district court to a court of appeals." We hold that the Tax Court's findings of fact are reviewed for clear error, but that mixed questions of law and fact are reviewed de novo, to the extent that the alleged error is in the misunderstanding of a legal standard. See 26 U.S.C. § 7482(a)(1); Having clarified the standard of review applicable to decisions of the Tax Court, we now turn to the merits of the instant case.
   n7 We readily acknowledge that a panel of our Court is bound by the decisions of prior panels until such time as they are overruled either by an en banc panel of our Court or by the Supreme Court, and thus that it would ordinarily be neither appropriate nor possible for us to reverse an existing Circuit precedent. In this case, however, we have circulated this opinion to all active members of this Court prior to filing and have received no objection, a process we refer to "as a mini-en banc." 

In his blog, Peter quotes from  an email I sent him about this procedural holding as follows:

For appellate courts and appellate lawyers (and thus for the clients they serve), this is an important issue.  It is always uncertain whether you can separate out the pure factual findings (subject to clear error review) and those factual questions that are so intertwined with legal conclusions that they are deemed mixed questions of fact and law.  I was in DOJ Tax Appellate for a while and thus issue was a significant issue. 

I guess I am a bit surprised that the Second Circuit would have ever thought that the standard of review for Tax Court Appeals under Section 7482(a)(1) was anything other than the same standard for district court appeals.  It sounds from the opinion discussing the history of the adoption of the erroneous position that the Second Circuit just mimicked the Seventh Circuit on the standard of Tax Court review without realizing that, in the Seventh Circuit, the clear error rule applied to both Tax Court appeals and district court appeals, whereas, by adopting the Seventh Circuit rule in the Second Circuit, the earlier panels created a discontinuity between the two types of appeals in the Second Circuit.

I thought I would say a bit more in my own blog on this standard of review issue.  As indicated, I handled court of appeals work in the various federal courts of appeals while in the DOJ Tax Appellate Section.  I and others thought about this genre of issue often. The broader issue is the level of deference a court of appeals gives to the trial court's findings and conclusions (and the in-between as well).  The traditional dichotomy is between facts and law.  Facts are subject to the clearly erroneous standard in all circuits, because that is what Rule 52(a) mandates.  The problem is in the fuzzy in-between area, commonly called mixed questions of fact and law.  As noted in the Second Circuit's opinion, two different and highly respected federal courts of appeals now have different views on the subject.  Other circuits may have different views in this binary choice, but I have not tried to research that issue for these comments.

The bottom line is that a de novo review gives appellate courts more latitude to change the results of the trial court.

For readers who want to pursue these scope of appellate review issues further, I do highly recommend the following article:  Kevin Casey, Jade Camara and Nancy Wright, Standards of Appellate Review in the Federal Circuit: Substance and Semantics, 11 Fed. Cir. B.J. 279 (2001), which is available on the web in several places, including here and here.   In addition to an exhaustive analysis of the issues, the authors of that article have the following pungent quote addressing some of the jargon that has been thrown about in this context from Armour & Co. v. Wilson & Co., 274 F.2d 143  (7th Cir. 1960):

We have come to speak of questions of 'facts', 'primary facts,' 'subsidiary facts,' 'evidentiary facts,' 'ultimate facts,' 'physical facts,' 'documentary facts,' 'oral evidence,' 'inferences,' 'reasonable [**33]  inferences,' 'findings of fact,' 'conclusions,' 'conclusions of law,' 'questions of fact,' 'questions of law,' 'mixed questions of law and fact,' 'correct criteria of law,' and so on ad infinitum. The simple answer is that we are all too frequently dealing in semantics, and our choice of words does not always reflect the magic we would prefer to ascribe to them. 

So, I have no prediction as to when or if this procedural issue will be resolved by the Supreme Court.  I am a little surprised that the Supreme Court has not yet addressed it and imposed uniformity among the circuits.  I did a quick LEXIS-NEXIS key-word search of the Supreme Court opinion database.  My searching was not exhaustive but I offer as an anecdote the only relevant authority I found on the issue of the proper standard of review for mixed questions of fact and law.

In Pullman-Standard, Div. of Pullman v. Swint, 456 U.S. 273  (U.S. 1982), Title VII race discrimination case, the court of appeals had reversed a factual finding of the district court.  The Supreme Court thought the Fifth Circuit had not properly applied the Rule 52(a) standard.  Here are some quotes from the opinion:

First, from p. 286 at  fn. 16 (most case citations omitted):

fn 16 There is some indication in the opinions of the Court of Appeals for the Fifth Circuit (see n. 15, supra) that the Circuit rule with respect to "ultimate facts" is only another way of stating a standard of review with respect to mixed questions of law and fact -- the ultimate "fact" is the statutory, legally determinative consideration (here, intentional discrimination) which is or is not satisfied by subsidiary facts admitted or found by the trier of fact. As indicated in the text, however, the question of intentional discrimination under § 703(h) is a pure question of fact. Furthermore, the Court of Appeals' opinion in this case appears to address the issue as a question of fact unmixed with legal considerations. 

At the same time, this Court has on occasion itself indicated that findings on "ultimate facts" are independently reviewable. In Baumgartner v. United States, 322 U.S. 665 (1944), the issue was whether or not the findings of the two lower courts satisfied the clear-and-convincing standard of proof necessary to sustain a denaturalization decree. The Court held that the conclusion of the two lower courts that the exacting standard of proof had been satisfied was not an unreviewable finding of fact but one that a reviewing court could independently assess. The Court referred to the finding as one of "ultimate" fact, which in that case involved an appraisal of the strength of the entire body of evidence. The Court said that the significance of the clear-and-convincing proof standard "would be lost" if the ascertainment by the lower courts whether that exacting standard of proof had been satisfied on the whole record were to be deemed a "fact" of the same order as all other "facts not open to review here." Id., at 671. 

The Fifth Circuit's rule on appellate consideration of "ultimate facts" has its roots in this discussion in Baumgartner. In Galena Oaks Corp. v. Scofield, 218 F.2d 217 (CA5 1954), in which the question was whether the gain derived from the sale of a number of houses was to be treated as capital gain or ordinary income, the Court of Appeals relied directly on Baumgartner in holding that this was an issue of "ultimate fact" that an appellate court may review free of the clearly-erroneous rule. Causey v. Ford Motor Co., supra, at 421, relying on Galena Oaks Corp. v. Scofield, supra, said that "although discrimination vel non is essentially a question of fact, it is, at the same time, the ultimate issue for resolution in this case" and as such, was deemed to be independently reviewable. The passage from East v. Romine, Inc., supra, at 339, which was repeated in the case before us now, supra, at 285, rested on the opinion in Causey v. Ford Motor Co. 

Whatever Baumgartner may have meant by its discussion of "ultimate facts," it surely did not mean that whenever the result in a case turns on a factual finding, an appellate court need not remain within the constraints of Rule 52(a). Baumgartner's discussion of "ultimate facts" referred not to pure findings of fact -- as we find discriminatory intent to be in this context -- but to findings that "clearly [imply] the application of standards of law." 322 U.S., at 671.

Next from pp. 287-288 (footnotge omitted):

The Rule does not apply to conclusions of law. The Court of Appeals, therefore, was quite right in saying that if a district court's findings rest on an erroneous view of the law, they may be set aside on that basis. But here the District Court was not faulted for misunderstanding or applying an erroneous definition of intentional discrimination. n17 It was reversed for arriving at what the Court of Appeals thought was an erroneous finding as to whether the differential impact of the seniority system reflected an intent to discriminate on account of race. That question, as we see it, is a  pure question of fact, subject to Rule 52(a)'s clearly-erroneous standard. It is not a question of law and not a mixed question of law and fact.
   n17 As we noted above, the Court of Appeals did at certain points purport to correct what it viewed as legal errors on the part of the District Court. The presence of such legal errors may justify a remand by the Court of Appeals to the District Court for additional factfinding under the correct legal standard. Infra, at 291-292

Note the bold-faced part.  I infer that the Court was saying that the standard of review is different for facts and mixed questions of fact and law.  That does not resolve that the standard of review for mixed questions of law should be de novo or any other standard.  Just that they are different and certainly leans toward the Second Circuit's resolution of the issue for both district courts and the Tax Court.

Addendum 11/16/13:  The Tax Prof Blog reports today here that Professor Leandra Ledeterman, here, has a draft article that will be published in 2014.  The article is:  (Un)Appealing Deference to the Tax Court, 63 Duke L.J. ___ (2014).  I have emailed Professor Lederman a request for making a draft of it available.  If she provides such a draft or a link to it, I will update this blog entry. In the meantime, the Tax Prof Blog has this which is either a summary of the article or the opening of the article (emphasis supplied by JAT):

The U.S. Tax Court hears the vast majority of litigated federal tax cases. It occupies an unusual place in the federal government in that it is a federal court located outside the judicial branch but whose decisions are appealable to the U.S. Courts of Appeals. This unusual situation coupled with its history as an independent agency in the executive branch can give rise to important questions, such as the standard of review that should apply to its decisions. In particular, should the Courts of Appeals treat Tax Court decisions the same as those of district courts in tax cases, or should they apply a more deferential standard analogous to review of an agency decision, as the Supreme Court held in 1943 in Dobson v. Commissioner? [Dobson v. Commssioner, 320 U.S. 489 (1943), here.]

Answering the standard of review question implicates issues of both law and policy. The article argues, contrary to some scholarship, that, as a doctrinal matter, no vestige of the Dobson rule remains and that Courts of Appeals must apply the same standard of judicial review they apply to district courts in non-jury cases. The article further argues that appellate review theory supports that result, as well. As the article explains, the Dobson rule was a largely instrumental one designed by Justice Jackson to reduce the volume of tax litigation. Although tax litigation has the unique characteristics of decentralization and the expertise of the Tax Court, the article demonstrates that those differences do not support departing from the policies supporting appellate review. Appellate courts therefore should not defer to the interpretations of the Tax Court any more than they do to those of the district courts.

Focus on the highlighted question.  I thought that Section 7482(a) directly answers that question.  I haven't done any recent research on Dobson, but thought that all vestiges of it were long since past.  If that is the case, it seems to me that revisiting the prudence of Dobson is more an academic exercise than a practical one.  But, I suppose, that is what law reviews are for.  I have published some impractical articles myself, and hope at best that they were at least somewhat academic.

For further information, if not elucidation, while doing some quick research on the Dobson rule, I ran across the following article which deals with the scope of review -- the principal subject of the blog and cites Dobson in a footnote:  Christopher Pietruszkiewicz, Economic Substance and the Standard of Review, 60 Ala. L. Rev. 339, 356 (2009)).  Key (even if lengthy) excerpts are as follows with many footnotes omitted:

II. Standards of Review 

Standards of review define the scope of power between judicial actors each functioning within a statutory or rule-based scheme and carrying out specified responsibilities under that system. n55 Because the selection of a standard of review often dictates the ultimate resolution of a case, the framework for selecting the applicable standard is significant. 56 A standard of review reflects the degree to which the original decision maker must be wrong for a reviewer to reverse the original decision. Traditionally, judicial review utilizes three standards-de novo, clearly erroneous, and abuse of discretion.
   n55 See Edward H. Cooper, Civil Rule 52(a): Rationing and Rationalizing the Resources of Appellate Review, 63 NOTRE DAME L. REV. 645, 645 (1988) ("Rule 52(a) serves a vital institutional role in allocating the responsibility and the power of decision between district courts and the courts of appeals."); see also Ronald R. Hofer, Standards of Review-Looking Beyond the Labels, 74 MARQ. L. REV. 231, 232 (1991) (describing standards of review as the level of deference that appellate courts provide to trials courts); Alvin B. Rubin, The Admiralty Case on Appeal in the Fifth Circuit, 43 LA. L. REV. 869, 873 (1983) ("Standards of review . . . indicate the decibel level at which the appellate advocate must play to catch the judicial ear.").
   n56 A number of appellate courts have reversed determinations by the trial court using a de novo standard of review. Conversely, courts that applied the clearly erroneous test affirmed the decision of the trial court. See Magee & Goldman, supra note 51, at 482 n.15.  

De novo review applies to questions of law. It is the least restrictive standard of review as it calls for no degree of deference and permits a reviewing court to determine the correct resolution of an issue on its own accord. In essence, de novo review provides no deference but is a judicial determination of an issue entirely independent of the prior resolution. Under a de novo standard, a reviewing court is "willing to reverse [a prior] conclusion of law solely on the basis that it believes that conclusion to be incorrect."  

The clearly erroneous standard of review applies to questions of fact and gives a significant amount of deference to the trier of fact. Under the clearly erroneous standard of review, "a finding is 'clearly erroneous' when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." In other words, a reviewing court will not substitute its judgment for the judgment of the trier of fact notwithstanding that had the appellate court been sitting as fact finder, it might have reached a different finding. Any determination of the trial court under this standard carries with it significant weight. Consequently, the role of the appellate court under a clearly erroneous standard of review is "not to decide factual issues de novo." 

The third standard of review, abuse of discretion, provides the highest degree of deference to a determination by the trial court. Under this standard, an abuse of discretion occurs when an adjudicator fails to exercise sound, reasonable, and legal decision-making skills. This standard applies to the discretionary functions of a trial court and has been applied by appellate courts reviewing decisions such as the adequacy or excessiveness of jury verdicts, the exclusion of scientific evidence, evidentiary rulings, estoppel, sanctions, and attorneys' fees. Consequently, the abuse of discretion standard of review is not a standard of review that appellate courts would articulate in reviewing a trial court decision regarding the economic substance doctrine. 

A. Historical Perspectives 

The scope of review by appellate courts is not a new question. Beginning with the Constitution over 200 years ago, the question first surfaced regarding a provision that provides that the "supreme Court shall have appellate Jurisdiction, both as to Law and Fact, with such Exceptions, and under such Regulations as the Congress shall make." 

Historically, suits initiated under the common law were heard by juries and suits in equity were heard by judges sitting without a jury. In the equity context, because cases were presented through depositions and interrogatories completed by the parties outside of the presence of the trial court, an appellate court was just as capable of making decisions on factual issues as was the trial court. As a result, appellate courts possessed unlimited power to review the entire record of the trial court and considered factual and legal questions independently of a determination by a trial judge.

In 1912, trial courts sitting in equity began to hear oral testimony pursuant to the Federal Equity Rules, and appellate courts, recognizing that the previous de novo method of review placed them at a disadvantage compared to the trial judges who heard the testimony, abandoned the de novo standard in favor of a standard giving great weight to the findings of trial judges. This self-imposed limitation was based principally on the superior ability of trial judges to make credibility determinations. Because the trial judge heard the testimony that served as the basis for findings of fact, those findings were not disturbed unless such findings were clearly erroneous.  

Thus, the predecessor to the "clearly erroneous" rule that was later adopted in the Federal Rules of Civil Procedure applied to findings based on conflicting testimony or facts derived or inferred from such testimony. In cases of facts derived or inferred from uncontradicted evidence, documentary evidence, or deposition testimony, appellate courts did not consider themselves bound by trial court determinations because they saw themselves as just as qualified to make such determinations. 

The distinctions between law and equity were confused at best, and in 1934, a new of set of rules attempted to eliminate such distinctions. Initially, the draft version of the Federal Rules of Civil Procedure provided that facts determined by a judge (as opposed to a jury) would "'have the same effect as that heretofore given to findings in suits in equity.'" The draft rule prompted a discussion that highlighted the distinction between the heightened deference standard that was applied to findings by a jury at law and the lower standard of review given in equity practice. The rule ultimately was drafted using the clearly erroneous language, which applied to appellate review of findings at equity based on oral testimony.  

The Advisory Committee note accompanying Rule 52 provided that the clearly erroneous standard of review should apply whether the finding of fact was determined based on conflicting testimony, or "deduced or inferred from uncontradicted testimony." According to William D. Mitchell, former Attorney General and the Chairman of the Advisory Committee appointed by the Supreme Court to draft the Federal Rules of Civil Procedure, the rule that 

applies to all cases tried by a judge without a jury is practically the modern equity rule; it is not the ancient equity rule which allowed trial de novo on the facts in the appellate court-it is a limited provision-the appellate court may not set aside the findings unless they are clearly erroneous or against the clear weight of evidence.

As a result of this rule, an appellate court must examine the entire record and determine for itself the appropriate findings of fact, subject to the major restriction that the appellate court must accept as practically conclusive findings of fact by a trial judge that are based on the credibility of witnesses. Should the appellate court disagree with the findings of the trial court, it will reverse only if it is convinced that the findings of the trial court are "unquestionably wrong."

Thus, prior to the enactment of the Federal Rules of Civil Procedure, particular emphasis was given to findings based on the testimony of witnesses. In cases in which the evidence was contradictory, appellate courts did not disturb the findings of the trial judge. By contrast, if the findings were either undisputed or were based on a factual determination consisting solely of documentary evidence, the appellate court sat in a comparable position to the trial court. The Federal Rules provided that the findings of fact by a trial judge are to be accepted unless clearly erroneous. It was not explicit, however, that this standard should be applied to documentary evidence. Consequently, the only major modification to Rule 52(a) since its adoption in 1938 was adopted in 1985, adding language that the standard of review-clearly erroneous-applied to findings of fact "whether based on oral or documentary evidence." With this change, it became clear that the clearly erroneous standard of review applied not only to credibility determinations but also to facts derived from oral and documentary evidence, an obvious signal that institutional responsibilities of trial courts and appellate courts should be respected. 

B. Fact-Law Distinction 

Traditionally, the level of appellate review hinged on the distinction between law and fact. Application of this distinction to appellate review is a rather simplistic exercise-appellate courts are free to review legal conclusions de novo, and factual findings are allowed a level of deference such that those findings, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous. While seemingly straightforward, the distinction between a factual finding and a legal conclusion is often murky. n89 The systematic difficulty is in describing what constitutes fact, what constitutes law, and what constitutes both.
   n89 See Nathan Isaacs, The Law and the Facts, 22 COLUM. L. REV. 1, 1 (1922) (defining the distinction between law and fact in terms of "delusive simplicity"); see also Carissa Byrne Hessick & F. Andrew Hessick, Appellate Review of Sentencing Decisions, 60 ALA. L. REV. 1, 14-18, 25 (2008) (describing the difficult distinction in the context of criminal sentencing review).  

Just as defining the level of deference under the clearly erroneous standard to an objective certainty is impractical, n90 defining the distinction between fact and law is just as unworkable. n91 While courts often create sound bites and elaborate musings on the definition of each, the distinction cannot be articulated to an objective certainty. n92 Professor Clark noted this dilemma:
   n90 See Christopher M. Pietruszkiewicz, Conflating Standards of Review in the Tax Court: A Lesson in Ambiguity, 44 HOUS. L. REV. 1337, 1370-71 (2008) ("When Congress does not utilize terminology with which courts interact frequently, thereby providing conflicting messages, consistent application of a standard is not possible. While it is impossible to pinpoint the definition of preponderance of the evidence, clear and convincing evidence, and beyond a reasonable doubt, courts are nonetheless able to apply these standards because they are universal and commonly articulated."); id. at 1363 ("Due to the imprecision of language and the unrealistic view that infinite degrees of deference can exist, standards of review should be limited to a familiar set of gradations."); see also Clermont, supra note 57, at 1148.
   n91 See Dobson v. Comm'r, 320 U.S. 489, 500-01 (1943); [Balance of footnote omitted]
   n92 See Cooper, supra note 55, at 645; see also Tupman v. Haberkern, 280 P. 970, 973 (Cal. 1929) (stating that questions of fact are decided by the trial court and that questions of law are decided by the appellate court). Rule 52 of the Federal Rules of Civil Procedure also makes this distinction, providing that "findings of fact, whether based on oral or other evidence, must not be set aside unless clearly erroneous, and the reviewing court must give due regard to the trial court's opportunity to judge the witnesses' credibility." FED. R. CIV. P. 52(a)(6). By negative implication, Rule 52(a) promotes independent review of legal questions. See Pullman-Standard v. Swint, 456 U.S. 273, 287-88 (1982). 

There seems to be no hard and fast distinction between questions of law and questions of fact; . . . and since there are so many cases on the border line between ultimate facts and conclusions of   facts, i.e., law, the way is open to a court with understanding to accomplish substantial justice, whatever the formula. n93
   n93 Clark & Stone, supra note 72, at 211 n.93; see also Louis L. Jaffe, Judicial Review: Question of Law, 69 HARV. L. REV. 239, 239-40 (1955) (defining the fact-law distinction as imperceptibly blending into each other).  

The use of fact and law distinctions allow appellate courts to cast questions as either of fact or of law depending on whether the appellate court ultimately favors reversing, or affirming, the trial court. n94 As a result, appellate courts can produce a "correct result" by utilizing the fact-law distinction and the corresponding standard of review to achieve its design. n95 In other words, based on the less than clearly defined line between law and fact, an appellate court can easily review and decide a factual issue by simply recasting it as a question of law and applying a de novo standard of review instead of the more stringent clearly erroneous standard. 96 As Professor Wright laments:
   n94 See, e.g., Paul L. Caron, Tax Myopia Meets Tax Hyperopia: The Unproven Case of Increased Judicial Deference to Revenue Rulings, 57 OHIO ST. L.J. 637, 644-45 (1996); Russell L. Weaver, A Foolish Consistency Is the Hobgoblin of Little Minds, 44 BAYLOR L. REV. 529, 553 (1992); see also Quintin Johnstone, An Evaluation of the Rules of Statutory Interpretation, 3 U. KAN. L. REV. 1, 5 (1954) (courts "pull respectable-sounding rules to justify any possible result"); Christopher M. Pietruszkiewicz, Discarded Deference: Judicial Independence in Informal Agency Guidance, 74 TENN. L. REV. 1, 8 (2006).
   n95 Caron, supra note 94, at 644-45 ("The precise verbal formulation used by a court is mere window-dressing that does not have any effect on the ultimate resolution of the case."); Richard J. Pierce, Jr., Legislative Reform of Judicial Review of Agency Actions, 44 DUKE L.J. 1110, 1110 (1995) ("A court often can write an opinion that reverses a major agency action as easily as it can write an opinion that upholds the same action."); see also Pietruszkiewicz, supra note 94, at 8; Patricia M. Wald, The Rhetoric of Results and the Results of Rhetoric: Judicial Writings, 62 U. CHI. L. REV. 1371, 1391-92 (1995) (contrasting the tones of two opinions written in the same circuit, one of which upheld a NLRB determination and another overturning a NLRB determination).
   n96 See David Frisch, Contractual Choice of Law and the Prudential Foundations of Appellate Review, 56 VAND. L. REV. 57, 92 (2003); Paul, supra note 91, at 811-12 ("'When the courts are unwilling to review, they are tempted to explain by the easy device of calling the question one of fact; and when otherwise disposed, they say that it is a question of law.'" (internal quotation marks omitted) (quoting Dickinson, supra note 91, at 55)); Charles Alan Wright, The Doubtful Omniscience of Appellate Courts, 41 MINN. L. REV. 751, 751 (1957) (arguing that appellate courts can control outcome thereby diminishing significance of trial courts).  

The principal means by which appellate courts have obtained such complete control of litigation has been the transmutation of specific circumstances into questions of law. . . . And unless the appellate judge handling the case is a dullard, some doctrine is always at hand to achieve the ends of justice, as they appear to the appellate court. 97
   n97 Wright, supra note 96, at 751; see also Evan Tsen Lee, Principled Decision Making and the Proper Role of Federal Appellate Courts: The Mixed Questions Conflict, 64 S. CAL. L. REV. 235, 236 (1991) ("The labels 'law' and 'fact' often amount to little more than divisions of decision making authority between judges and juries or between appellate courts and trial courts."); Stephen A. Weiner, The Civil Nonjury Trial and the Law-Fact Distinction, 55 CAL. L. REV. 1020, 1022 (1967) ("Since law application cannot be meaningfully described as either lawmaking or factfinding, such terminology is not a useful analytical tool in answering the question confronting the court."). 

This is not a new problem, and the adoption of the Federal Rules of Civil Procedure had no impact on resolving the distinction, instead creating the hybrid form of a "mixed question." 98 In fact, proper application of such mixed questions was debated as early as 1786.

Courts, however, have not grasped the distinction - or more appropriately, the lack of distinction - between law and fact and continue to maintain the misguided illusion that the standard of review is dictated by whether the question presented is a determination of law or a determination of fact. n100 This perception continues notwithstanding that the Supreme Court in 1944 alluded to the lack of a clear dividing line, questioning whether review by an appellate court related to a question of fact or a question of law and stating that the distinction is "never self-executing." n101 This trend continued in Pullman-Standard v. Swint, where the Court described the distinction between fact and law as "vexing," and in Bose Corp. v. Consumers Union of United States, Inc., suggesting that the fact-law distinction "varies according to the nature of the substantive law at issue." n103
   n100 See Paul D. Carrington, The Power of District Judges and the Responsibility of Courts of Appeals, 3 GA. L. REV. 507, 518 (1969) ("Findings of fact may be defined as the class of decisions we choose to leave to the trier of fact subject only to limited review, while conclusions of law are the class of decisions which reviewers chose to make for themselves without deference to the judgment of the trial forum.").
   n101 Baumgartner v. United States, 322 U.S. 665, 671 (1944).
   n103 466 U.S. 485, 501 n.17 (1984).  

Professor Cooper sums up the application of a fact-law determination, noting that "the fundamental secret is out, and notoriously so. Characterization of an issue of law application as fact or law for purposes of identifying a formalized standard of review depends on the perceived need for review, not on the actual status of the issue." This follows a similar conclusion in 1927 that "any factual state or relation which the courts conclude to regard as sufficiently important to be made decisive for all subsequent cases of similar character becomes thereby a matter of law for formulation by the court." 

C. Division of Responsibility in Fact-Law Analysis 

If the fact-law distinction is merely a guise that permits an appellate court to create a deference standard that suits its view of a particular case, a more appropriate framework of appellate review should consist of an allocation of institutional responsibility between trial courts and appellate courts. Legal Realists and Legal Proceduralists share the jurisprudential theme of discretion in decision making. Realists recognize discretion as a key component of the decisional process and provide a "rational basis for . . . justifying decisions." Necessarily, Realists reject the notion of legal formalism in favor of judges having the ability to decide cases in contradictory ways and subsequently finding adequate grounds for reaching such a result after the fact. Conversely, Proceduralists focus attention on the institution and structure of decision making and believe that the function of discretion is to allocate responsibility among those decision makers. 108 While Realists focus on the individual thought process of reaching a result, Proceduralists focus on the relationship between the decision maker and the institution in an effort to determine the correct answer in law.  

Because of institutional factors and role of the appellate courts within the judicial hierarchy, the fact-law distinction necessarily involves discretion. The question then is how much discretion is appropriate for each decision maker in the judicial hierarchy. The answer cannot be based on a mere reflection of the current judicial practice, defined by the imprecise rules regarding what is fact and what is law that have been debated for centuries. More relevant to the role of appellate review is the function of federal courts as institutional actors. 

D. Role of Appellate Review 

Appellate courts have two primary purposes, and the standard of review to be applied by appellate courts should relate to those two purposes. Appellate courts should serve to develop the law in a particular area as guidance for future cases and to rectify egregious errors in particular cases.  

It is too restrictive to complain that the increasing volume of cases on appeal should dictate the level of review by an appellate court. n113 One possible solution to the volume problem would be to increase the capacity of the appellate courts to "handle" all appeals. This result is not feasible or desirable. Instead, appellate courts should focus their energy on getting the decision "right" in the context of those cases in which the error below rises to an egregious level. n114 A deferential standard of review-clearly erroneous-accomplishes this result. The standard of review, therefore, describes more than the level of deference: it defines the level of responsibility of appellate courts. n115
   n113 See Cooper, supra note 55, at 650 ("It would be remarkable if the actual working standard of review were not affected by the shifting functional ability of the courts of appeals to devote attention to the wisdom of specific findings in particular cases."). As the Ninth Circuit noted:
It can hardly be disputed that application of a non-deferential standard of review requires a greater investment of appellate resources than does application of the clearly erroneous standard. Appellate courts could do their work more quickly if they applied the clearly erroneous standard in most circumstances, because the courts then need only determine if the lower court's decision is a reasonable one, not substitute their own judgment for that of the trial judge.
United States v. McConney, 728 F.2d 1195, 1201 n.7 (9th Cir. 1984), overruled by Estate of Merchant v. Comm'r, 947 F.2d 1390 (9th Cir. 1991).
   A related question, which is not susceptible to measurement, is the level of judicial review afforded to district judges based on past history or reputation. District judges, and appellate judges who review their decisions, including findings of fact, do not act without influence from previous cases. In this sense, appellate review can dictate a specified level of review, but a history of an appellate court with the findings of a particular district judge may influence-subconsciously or otherwise-the level of inquiry notwithstanding the stated level of discretion. See FLEMING JAMES, JR. & GEOFFREY C. HAZARD, JR., CIVIL PROCEDURE § 12.8, at 668 (3d ed. 1985) ("An appellate court's inclination to accept a trial judge's findings depends . . . on the court's unstated degree of confidence in the trial judge's fairmindedness."); Cooper, supra note 55, at 655-56; Martin B. Louis, Allocating Adjudicative Decision Making Authority Between the Trial and Appellate Levels: A Unified View of the Scope of Review, the Judge/Jury Question, and Procedural Discretion, 64 N.C. L. REV. 993, 1015-16, 1016 n.160 (1986) (finding that, off the record, "former appellate law clerks . . . regularly attest that both important cases and the decisions of certain notorious trial judges are scrutinized more carefully than others.").
   n114 See Anderson v. City of Bessemer City, 470 U.S. 564, 575 (1985) ("Documents or objective evidence may contradict the witness' story; or the story itself may be so internally inconsistent or implausible on its face that a reasonable factfinder would not credit it. Where such factors are present, the court of appeals may well find clear error even in a finding purportedly based on a credibility determination."); United States v. U.S. Gypsum Co., 333 U.S. 364, 396 (1948) ("Where such testimony is in conflict with contemporaneous documents we can give it little weight, particularly when the crucial issues involve mixed questions of law and fact. Despite the opportunity of the trial court to appraise the credibility of the witnesses, we cannot under the circumstances of this case rule otherwise than that [this finding] is clearly erroneous.").
   n115 See Cooper, supra note 55, at 651 ("It may be wise to serve the interests of all litigants by adopting standards of review that help sift out all but the more extreme claims of error."). This division of responsibility may also rest on the belief that trial court judges and appellate court judges develop skills essential to each function and, over time, increase competency in the area with which they become most familiar. See Frisch, supra note 96, at 77.  

The effect of this division of labor, which uses responsibility as a guidepost, accomplishes both purposes. First, appellate courts can focus their capacities on developing law as opposed to focusing on factually intensive, case-specific questions with little value beyond the case at issue. Second, a clearly erroneous standard of review allows appellate courts to monitor trial courts for major errors even if a particular case does not relate to the first objective of developing the law in a particular area. Deference is appropriate in cases based primarily on "'multifarious, fleeting, special, narrow facts that utterly resist generalization'" and where the "investment of appellate energy will . . . fail to produce the normal law-clarifying benefits that come from an appellate decision on a question of law." Simply then, trial courts sort through evidence, written or oral, credibility-based or not, and make determinations of fact, and appellate courts take those facts as determined by a trial court and develop the law in a particular area. Judge Posner ascribes to this view: 

The main reason for appellate deference to the findings of fact made by the trial court is not the appellate court's lack of access to the materials for decision but that its main responsibility is to maintain the uniformity and coherence of the law, a responsibility [*362]  not engaged in if the only question is the legal significance of a particular and nonrecurring set of historical events. n121
   n121 Id.; see also Lee, supra note 97, at 240 ("The nub of Judge Posner's rationale is that if an appellate court is not going to create useful precedent with its decision [or to promote uniformity], then no rationale justifies discarding the district court's work by non-deferential review."); Wright, supra note 96, at 779 ("From the earliest times appellate courts have been empowered to reverse for errors of law, to announce the rules which are to be applied, and to ensure uniformity in the rules applied by various inferior tribunals.").  

The appellate function is not to determine whether the findings of fact as determined by the trial court are correct, but instead it is to determine whether the findings of fact as determined by the trial court are clearly wrong. n122 Under this rationale, nearly correct factual determinations are close enough that "making a finer determination" of factual issues on appeal "is either not possible or not worth the time and effort," considering the appellate courts' dual role of error correction and development of the law. In making factual determinations and drawing inferences from those determinations, the clearly erroneous standard suggests that, given the theoretical nature of fact finding, no amount of additional consideration at an appellate level is likely to produce, in theoretical terms, a more correct result.
   n122 Cooper, supra note 55, at 657; see also United States v. Aluminum Co. of Am., 148 F.2d 416, 433 (2d Cir. 1945) (Hand, J.) (Reluctance to reverse the findings of a trial judge "is true to a considerable degree even when the judge has not seen the witnesses. His duty is to sift the evidence, to put it into logical sequence and to make the proper inferences from it; and in the case of a record of over 40,000 pages like that before us, it is physically impossible for an appellate court to function at all without ascribing some prima facie validity to his conclusions."). 

Professor Maurice Rosenberg best describes the phenomenon as the difference between primary discretion and secondary discretion. 125 Although referring to legal decision making and how discretion affects those choices, the theory is equally applicable to the standard of review artificially created by legal rules and common law jurisprudence.
   n125 Maurice Rosenberg, Judicial Discretion of the Trial Court, Viewed From Above, 22 SYRACUSE L. REV. 635, 637 (1971).  

Primary discretion is grounded in the ability to act independently when rules do not exist to guide the resolution. More interesting for these purposes is secondary discretion, which concerns the relationship among judges in the hierarchical judicial system. These relationships define how much an appellate court is willing to restrain its own views in light of the responsibility of the trial court. In other words, a trial court can be wrong to a certain extent without an appellate court finding "enough" error to reverse its determination. Secondary discretion defines how wrong a trial court must be for an appellate court to substitute its views. 129 

If secondary discretion is to apply in a standard of review context, de novo review is not an appropriate standard. De novo review is appropriate for rules (i.e., law determinations). If the responsibility of an appellate court is to develop the law, then review of action by a trial court that influences the development is appropriate. Appellate courts ensure that lower courts follow the law, promote efficiency and predictability, and ultimately provide a legitimacy that is grounded in the rule of law. Legal determinations are justified by a trial court by demonstrating the application of the factual issues to the appropriate controlling rule. 

Appellate courts, by the nature of their responsibility within the judicial hierarchy and its attending institutional structure, are deemed to know more about the law and the appropriate rule that controls the legal issues. Thus, de novo review in this context fits within the system of judicial institutional goals. Considering the entire judicial system as integrated, appellate resources should be devoted to matters that are most important to the proper functioning of the judicial system-promoting uniformity and predictability in the law.  

Appellate courts, through their law-developing function, act both affirmatively and negatively in directing the trial court on the legal decision-making process. Of course, affirmative action consists of announcing a legal position or precedent to be followed by a trial court. Just as importantly, however, appellate courts restrict the choices of trial courts by ruling out particular options or raising concerns regarding particular avenues of decision making, suggesting to trial courts that particular options will be rejected on review. The law-development function promotes out comes that are based within the range of acceptable legal results and advances predictability as a welcome consequence.

Such a view comports with the notion that language is not a sufficient basis for communicating a precise rule.Because language-and therefore rules of law-cannot be defined with particularized clarity, appellate courts must constantly refine their guidance to trial courts on the basis of the law and how that law is applied to a specific set of facts as determined by a trial court. Appellate courts, through their responsibility to develop the law, should more easily reverse those decisions by the trial court that have an impact on the law's development. This is not simply done by the creation of more elaborate rules, but by the courts' willingness to reverse those determinations in which that action is warranted. 

Trial judges find facts in the context of all facts considered in arriving at a factual determination. It is because of superior institutional competence that an appellate court does not reverse a trial court even if it would have reached a contrary conclusion on the facts presented or based on inferences that can be drawn from those facts. Upholding the factual determinations of the trial judge does not demonstrate that the trial judge was correct in those determinations, but it does reflect the institutional competence of the trial court. As long as the findings do not rise to the necessary level of wrong (i.e., clearly erroneous), the factual findings are not subject to reversal.  

Others have argued that parties who know that a reversal on appeal is unlikely will devote more energy to the initial trial. While seemingly plausible, there is a significant difference in "making a record" for appeal and devoting the energies necessary to be successful at the trial stage. It is unlikely that a party would rely on an appellate court to remedy a perceived injustice at the trial stage when the party itself did not advance an effort to successfully convince a trier of fact of its basis for prevailing at trial. Although the likelihood of success on appeal under a heightened standard of review is reduced, a de novo standard should not alter the approach to the initial trial. In other words, it is implausible that a litigant would simply try a case for appeal in circumstances where additional evidence cannot be introduced for the first time at the appellate level.  

Perhaps more believable, however, is the argument that the findings of fact of a panel of three appellate judges are more accurate than the findings of one judge at the district court level. n141 As the Supreme Court suggests, "Duplication of the trial judge's efforts in the court of appeals would very likely contribute only negligibly to the accuracy of fact determination at a huge cost in diversion of judicial resources." More significantly, appellate review of factual findings made by a trial court do not present the appellate courts with an opportunity to meaningfully develop the law and are case-specific determinations lacking any appreciable impact on predictability and uniformity. In short, the division of responsibility should be the determining factor of the fact-law distinction.
   n141 See Cooper, supra note 55, at 653 ("It is easier to criticize a symphony than to write one, and much easier for one person to write a symphony than for a panel of three."); see also Judges of the Federal Courts, 901 F.2d vii-xxx (1990) (excluding the Federal Circuit, eighty-one judges of 206 federal appellate judges are former district judges).