Monday, November 25, 2019
Lesson From The Tax Court: The Scope And Standard Of Review In CDP Cases
Many Tax Court cases teach lessons about Collection Due Process (CDP). The case of Norman Hinerfeld v. Commissioner, T.C. Memo. 2019-47 (May 2, 2019) (Judge Halpern), teaches a nice lesson about how the Tax Court reviews IRS CDP decisions. It illustrates the difference between the concepts of “scope” of review and “standard” of review. And it introduces readers to the wacky world of tax administrative law which, must to the consternation of those academics who like their law neat and tidy, is anything but neat and tidy. More below the fold.
To understand today’s lesson, one needs just a smidgen of background on the difference between scope of review and standard of review.
General Administrative Law
“Scope of review” refers to the appropriate set of information that the court may consider in conducting its review of an agency’s action. In general administrative law, the appropriate information set is, basically, the same information set that the agency had before it, commonly called the administrative record. Courts are supposed to use that same record to review the agency decision and “not some new record made initially in the review court.” Camp v. Pitts, 411 U.S. 138, 142 (1973). Limiting the information set to the administrative record is often called the “administrative record rule.”
The opposite of an administrative record rule would be a rule that permitted all parties to introduce anew (i.e. de novo) evidence before the reviewing court, whether or not that evidence had been available to the parties during the administrative process or put before the agency decision-makers. You could call this a de novo record rule.
“Standard of review” refers the degree of scrutiny a court gives the agency’s decision, based on the information set before it. A typical way to think about standard of review is to think about how much deference a reviewing court gives the agency decision. When no deference is required, the court is free to substitute its judgment for that of the agency. That is called de novo review. When, however, the court is reviewing an agency’s decision simply to ensure the agency has not abused the discretion given it by statute, then the court may not substitute its judgment for that of the agency. Instead, the court must uphold the agency’s decision unless the decision (1) rests on an error of law or (2) is “not sustainable on the administrative record made.” Camp v. Pitts, 411 U.S. at 143.
The general rule in administrative law is the abuse of discretion standard. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402 (1971). The Supreme Court has said that the Administrative Procedure Act (AP) authorizes de novo review “...in only two circumstances. First, such de novo review is authorized when the action is adjudicatory in nature and the agency factfinding procedures are inadequate. And [second,] there may be independent judicial factfinding when issues that were not before the agency are raised in a proceeding to enforce nonadjudicatory agency action.” Id. at 415 (1971).
Tax Court Exceptionalism
While the APA sets out general rules governing judicial review of agency actions, Congress is free to create special rules for court review of particular agencies. And boy has Congress done that in tax administration. It has created a special relationship between the Tax Court and the IRS. That special relationship has a long history, dating back to 1924, some 24 years before Congress even enacted the APA. For those interested, the go-to source for Tax Court history is "The United States Tax Court: An Historical Analysis," originally written by Harold Dubroff and expertly updated and revised by Brant Hellwig.
The Tax Court occupies a unique position in the U.S. Court system. That is partly because when it started out in 1924, it was an administrative office. And it is partly because the Tax Court's job description is inexorably tied to a single agency, the IRS. And, finally, it is partly because over time Congress has given the Tax Court new jobs to do.
At the beginning, the Tax Court (then called the Board of Tax Appeals) had the job of reviewing only one type of determination made by the IRS (then known as Bureau of Internal Revenue), a determination about the proper tax liability. Congress created the deficiency procedures to permit taxpayers to obtain pre-payment review of the IRS determination to assess additional taxes. See my post here for more on that process and on the role of the Notice of Deficiency (NOD) in that process. When Congress gave the Tax Court power to review NODs, it expressed the power this way: “the Tax Court shall have jurisdiction to redetermine the correct amount of the deficiency.” That is still the operative language, codified in §6214.
From the get-go, the Tax Court has interpreted the word “redetermine” to mean it should apply both de novo standard and de novo scope of review in deficiency proceedings. Barry v. Commissioner, 1 B.T.A. 156 (1924). Sure, the NOD gets a presumption of correctness, and the taxpayer bears the burden to prove entitlements to either exclusions or deductions, but the taxpayer is free to introduce any evidence it wants to convince the Tax Court that the NOD was erroneous. And what the taxpayer must persuade the court is not that the IRS abused its discretion but that the IRS was simply wrong: the Tax Court will substitute its judgment for that of the agency and will reverse the agency if it is convinced the IRS was simply incorrect. Substitution of judgment is the key distinction between a de novo standard of review and an abuse of discretion standard of review.
Over the years, Congress added more and more types of agency decisions to the Tax Court’s review portfolio. Each addition comes with its own statutory directive, which has created multiple issues on what scope and standard of review the Tax Court should use for each type of IRS decision it reviews. The Tax Court now has power to review IRS decisions about collection actions, spousal relief, and whistleblower awards. Today’s case involves the first.
In 1998 Congress gave the Tax Court the power to review certain IRS collection decisions through the CDP process. Sections §6320 and §6330 require the IRS to give taxpayers a CDP hearing before the IRS cranks up its collection machine and starts using its lien and levy powers. The CDP hearing gives taxpayers an opportunity to talk to a live IRS employee, called a Settlement Officer (SO), located in the Office of Appeals. The purpose of the hearing is for the SO to review the case and make sure that further administrative collection actions are appropriate. The taxpayer gets the chance to resolve outstanding liabilities through the use of “collection alternatives” such as installment plans, compromises of liability, or spousal relief.
Section 6330(c) lists all the “matters” that the SO is supposed to consider when deciding whether the IRS can continue collection action. One of the “matters” may be the validity of the underlying tax liability. Section 6330(c)(2)(B) permits taxpayers to sometime “raise at the hearing challenges to the existence or amount of the underlying tax liability....” The other matters go to the efficacy and fairness of the collection decision.
Thus, a CDP hearing might involve both matters about the appropriate amount of tax to collect as well as matters about the appropriate means of collection. Section 6330(d) permits taxpayers unhappy with the SO’s determination as to any matter properly before the SO to “petition the Tax Court for review of such determination” and provides that “the Tax Court shall have jurisdiction with respect to such matter.”
Currently, the Tax Court applies the same de novo record rule to all aspects of CDP cases, at least when those cases are not going up to the 1st, 8th, 9th Circuits. Robinette v. Commissioner, 123 T.C. 85, 95 (2004) rev’d 439 F.3d 455 (8th Cir. 2006). Those three circuits say the Tax Court must use the administrative record rule. You can find the cites in Judge Halpern's opinion. The Tax Court’s position here appears to be driven more by analogy to its traditional role in deficiency proceedings rather than the particular statutory language in §6330(d). That is, the Tax Court currently sees the additional CDP task Congress has given it as extensions of its task in deficiency proceedings. Since the Court easily employs a de novo record rule in the traditional task, there is no reason to think Congress meant anything different in assigning the additional task. However, more recent cases suggest that the Tax Court may be wobbling on this rationale. See Kaspar v. Commissioner, 150 T.C. 8 (2018)(deciding to use administrative record review when reviewing whistleblower awards).
In sharp contrast to the Tax Court's use of the same record rule to all aspects of CDP review, it applies two different standards of review, depending on what “matter” is under review. When the matter under review is the validity of the tax liability, the Court will use a de novo standard of review. For all other matters, the Court uses an abuse of discretion standard of review. Goza v. Commissioner, 114 T.C. 176 (2000).
You may be scratching your head. How can the Tax Court give an abuse of discretion review to an IRS decision when it may be looking at new evidence? It would seem that an abuse of discretion standard of review must be tied to an administrative record scope of review. After all, how can an agency abuse its discretion when it did not have a chance to look at the very evidence the taxpayer claims shows the abuse?
The answer is today's lesson. Hinerfeld suggests how a de novo scope of review might conceptually fit with an abuse of discretion standard of review.
Facts of the Case
In 2006 the IRS assessed the §6672 Trust Fund Recovery Penalty (TFRP) against Mr. Hinerfeld. For those who need background on the TFRP, see my prior Lessons: “The TFRP Trap for Accommodating Payroll Service Providers,” and “The Misunderstood Trust Fund Recovery Penalty.”
Here, after assessing Mr. Hinerfeld, the IRS sought to collect the assessed TFRP. In 2006 it sent him a CDP notice that it was going to start levying. He received a §6330 CDP hearing, lost it, and then lost a subsequent Tax Court review. Hinerfeld v. Commissioner, 139 T.C. 277 (2012).
In 2013 the IRS filed a Notice of Federal Tax Lien (NFTL) and, per §6320, sent Mr. Hinerfeld another CDP notice. The amount of TFRP liabilities listed on the NFTL totaled over $550,000. He again asked for, and received, a second CDP hearing and, again, asked for Tax Court review when the SO sustained the NFTL filing and rejected his OIC of $12,720.
Mr. Hinerfeld argued in Tax Court that the SO had abused her discretion in rejecting his OIC. The SO had rejected the OIC because she determined that Mr. Hinerfeld had sufficient assets with which to pay the liability. In particular, the SO counted as one of those assets 50% of the value of Mr. Hinerfeld’s principal residence in Larchmont, NY, a location where the median home value is over $1 million, at least according to this website.
Mr. Hinerfeld said the SO was simply wrong about the 50% interest because he did not legally own the primary residence he lived in. He said he sold it to his wife in 2003.
The SO (eventually) concluded that Mr. Hinerfeld’s 50% interest in the house should indeed count because while Mrs. Hinerfeld indeed had legal title, she held title as Mr. Hinerfeld’s nominee.
Nominee status results from a relationship between the taxpayer and property transferred to a third person. If that third person holds the property as the taxpayer’s nominee, any federal tax liens arising after the transfer will attach to the property because it is still “property belonging to the taxpayer” under section 6321. Therefore, the theory allows the IRS to vindicate tax liens that may arise even after property has been transferred, the if the transferee holds the property as the nominee of the taxpayer. You can find a much fuller discussion of nominee liability---as well as the related concepts of transferee and alter ego liabilities---in my article "Collecting Liabilities From Third Parties," 152 Tax Notes 1549 (September 12, 2016).
Whether a person has the legal status of a nominee depends on a number of factors. The IRM lists many of the factors various courts use to determine when one person holds property as the nominee of another person. The SO did not give any analysis of those factors but simply wrote: “Most of all of the factors listed in [IRM 5.17.2.5.7.2]...are present in this case.”
Mr. Hinerfeld argued that the SO’s failure to actually analyze the factors constituted an abuse of discretion and required, at the very least, a remand. Both he and his wife testified before Judge Halpern on the circumstances of the 2003 transfer of title. After all, the Tax Court was using the de novo record rule.
Lesson: De novo scope of review can fit an abuse of discretion standard of review.
Judge Halpern rejected Mr. Hinerfeld’s argument that the SO abused his discretion by failing to analyze the nominee factors. He allowed Mr. Hinerfeld to introduce new evidence (including oral testimony) and then walked through the various factors that a New York court (which is where the property was located) would use to decide the legal issue of Mrs. Hinerfeld’s nominee status. It was a quick walk because the factors pointed strongly to one conclusion: she was a nominee. In his walk, Judge Halpern was aided by the new evidence he was able to consider. So while he still performed an abuse of discretion review, the de novo scope of review helped him do that. The reason goes back to the two ways that an agency can abuse discretion: (1) its decision rests on an error of law; or (2) its decision is not supported by the record.
To me, it is key that the nominee determination is a mixed question of law and fact. That is, the legal status depends on a fact-specific, indeterminate, multi-factor test. The underlying factual determinations are the province of the SO but the application of those facts to the test is a question of law. Sometimes, a factual finding is so important to the legal rule that we say the result is “a matter of law.” For example, the fact that a person is over 25 years old and makes more than the §151 personal exemption amount means that person cannot be your dependent “as a matter of law” regardless of other factors. Generally, however, when the legal determination is bound up with factual determinations, the reviewing court will give it a “light appellate touch” much like an abuse of discretion standard of review. See e.g. the great discussion between Judges Posner and Wood in United States v. Frederick, 182 F.3d 496 (7th Cir. 1999).
In this case, Mr. Hinerfeld essentially argued that the SO’s decision rested on an error of law: Mrs. Hinerfeld was not, legally, Mr. Hinerfeld’s nominee. But that legal status depended on facts. The taxpayer attempted to adduce additional facts at trial to show why his wife should not, as a matter of law, be classified as his nominee. It did not work. Judge Halpern carefully reviewed all the evidence, both the evidence in the administrative record, and the new evidence produced at trial. He concluded that the SO’s decision did not rest on an error of law, summing it up this way:
“because...all of the...factors other than the recording of the deed weigh in favor of treating Mrs. Hinerfeld as petitioner's nominee in holding title to the Larchmont residence, SO Matthews' determination to that effect was not only reasonable but correct. *** Because Mrs. Hinerfeld can appropriately be treated as petitioner's nominee in holding title to the Larchmont residence, SO Matthews did not abuse her discretion in rejecting an OIC that did not reflect the value of that property.”
Comment
The distinction between de novo review and abuse of discretion review is not as neat and tidy as academics would like. Recall that the Tax Court performs a de novo review in deficiency proceedings. But it also uses a very strong presumption that the NOD is correct and it is up to the taxpayer to disprove that correctness. To me, de novo review with that strong presumption looks a lot like an abuse of discretion review. I think one slides into the other as a practical matter.
When I think of pairing a de novo scope of review with an abuse of discretion standard of review, I think of football. I think of NCAA replay review. Sometimes a field referee makes a call that, from the information available to the ref in real time is a correct call. However, when a reviewing official is able to acquire more information (different camera angles, different speeds) that reviewing official may disagree with the call. The reviewing official may only overturn a field call, however, when there is indisputable visual evidence that the original call was incorrect. To me that is like an abuse of discretion review because it does not allow the reviewing official to simply substitute his or her judgment for the field ref. Instead, only if the new evidence makes it indisputable that the field ref was wrong can the judgment of the field ref be overturned.
Just as the line between de novo and abuse of discretion review is not always clear, the distinction between the de novo record rule and the administrative record rule is much fuzzier than some might have you believe. Camp v. Pitts itself illustrates that the line is not a bright one. There, the federal Comptroller of the Currency rejected a bank’s application for a charter with only a very short, conclusory, explanation. The District Court sustained the rejection. The Court of Appeals reversed and remanded because it found that the Comptroller’s decision provided no basis for judicial review. There was plenty of information in the record. The problem, however, was the agency decision failed, like so many of my first year law students, to apply the law to the facts. The Court of Appeals ordered the District Court to hold a “trial de novo.”
The Supreme Court reversed the Court of Appeals, saying that the Court of Appeals was right about the problem but wrong in the remedy. Said the Supreme Court: “if there was such failure to explain administrative action as to frustrate effective judicial review, the remedy was not to hold a de novo hearing but, as contemplated by Overton Park, to obtain from the agency, either through affidavits or testimony, such additional explanation of the reasons for the agency decision as may prove necessary.” Id. at 143.
Thus, in this week's case, even if the Tax Court had applied the administrative record rule, that would not necessarily require a remand. The Court could have asked the IRS to supply affidavits from the RO in which she could give “such additional explanation” as would help the court properly review whether her decision rested on either an erroneous understanding of nominee law or else lacked such factual basis in the administrative record as to make the decision arbitrary.
Coda: The case shows again the downside to Collection Delay Process. Notice that it has now been over 13 years since the IRS assessed the TFRP liability against Mr. Hinerfeld. During almost that entire time the taxpayers have been using CDP to contest collection. Perhaps some of the liability has been paid because Judge Halpern notes the outstanding balance as of trial was down to $300,000. But the fact that the parties are still arguing over the NFTL strongly suggests that the IRS is still seeking payment. To me, this case is just a sad illustration of how Congress has knee-capped the IRS’s enforced collection function. Worse, the taxpayers are still on the hook with most if not all of collection statute of limitations still to go: 13 years later and the clock has probably not moved one day. So the taxpayers are likely another 10 years away from SOL relief. What good is done by keeping taxpayers under a Sword of Damocles for over 23 years? This is not good tax administration. It’s not fair to the government, to honest taxpayers, or to these taxpayers.
Bryan Camp is the George H. Mahon Professor of Law at Texas Tech University School of Law
https://taxprof.typepad.com/taxprof_blog/2019/11/lesson-from-the-tax-court-the-scope-and-standard-of-review-in-cdp-cases.html
Comments
Hi Jack, thanks for those thoughtful comments. I am still processing them, but one quick idea I would throw back is that the presumption of correctness also shields the NOD from arguments that the process which produced it was faulty in some respect. That is, the Court frequently says it will not look "behind" the NOD, which takes off the table a number of potential taxpayer attacks. So in that sense I do not see the presumption duplicative of the burden of persuasion.
Posted by: bryan | Nov 25, 2019 12:47:20 PM
Bryan, good post as I expect from you. A lot of administrative law to unpack there.
I have only one comment. You say in the opening paragraph under Comment:
Recall that the Tax Court performs a de novo review in deficiency proceedings. But it also uses a very strong presumption that the NOD is correct and it is up to the taxpayer to disprove that correctness. To me, de novo review with that strong presumption looks a lot like an abuse of discretion review. I think one slides into the other as a practical matter.
Do I read correctly that you are saying that, as a practical matter, a deficiency proceeding is an abuse of discretion review because of the combination of the presumption of correctness (the “strong presumption”) and the taxpayer burden to disprove?
I would not state it that way. I think by using the conjunctive for the presumption of correctness and the taxpayer burden of proof (persuasion in this case), you seem to say that they are mutually enhancing and the combination is stronger for the IRS than the components of the conjunction. I don’t think that is right.
In traditional burden of proof theory, a party with the burden of persuasion bears, for that reason alone, a production burden. In tax deficiency cases (and refund suits), since the taxpayer bears the burden of persuasion, the taxpayer bears the burden of production. In traditional burden of proof theory, presumptions merely impose upon a production burden and thus helps only where the party upon whom it is imposed would not otherwise have a production burden. Since the taxpayer already bears the burden of production by bearing the burden of persuasion, saying that the IRS determination carries a presumption of correctness is meaningless. As one court pungently noted, such a presumption covers with a handkerchief something already covered by a blanket. Chicago Stock Yards Co. v. Commissioner, 129 F.2d 937, 948 (1st Cir. 1942) (involving a IRC statutory presumption against the taxpayer, but noting that presumptions just “assist the party having the burden of proof” and that there is not “no practical effect” to a presumption where the party already has the burden of proof and, thus, the burden of production, analogizing the situation as being “like a handkerchief thrown over something also covered by a blanket”).
When I was with DOJ Tax Appellate, because of the common need for discussion of burden of proof principles in briefs, I prepared a memorandum on burden of proof that, when finalized, was circulated to guide DOJ Tax attorneys discussions of burden of proof in their briefs and arguments. One of the points as I recall I made (I have lost my copy of the memo) was that DOJ Tax attorneys should avoid discussion of the presumption of correctness where the taxpayer has the burden of persuasion because it added nothing except the erroneous implication that the taxpayer’s burden was somehow greater than it was or should be.
Your thoughts?
Posted by: Jack Townsend | Nov 25, 2019 11:04:45 AM
Good point on the process issue. However, even without the presumption, a de novo review with the taxpayer bearing the burden of persuasion (and thus burden of production) would mean that the process is irrelevant. So, the question I have not researched is whether the courts have justified the failure to look behind the NOD because of the presumption of correctness. So, if the presumption serves that function, then it is important. My point was that to the extent that the presumption serves the traditional procedural role of imposing a burden of production, that role is redundant with the normal burden of persuasion and resulting burden of production.
Posted by: Jack Townsend | Nov 27, 2019 9:57:28 AM