The Internal Revenue Service determined that Leon Malachinski was deficient in the payment of his federal income taxes for the year 1980. Dr. Malachinski contested the assessment in the United States Tax Court. He contended (1) that his signature on a consent form to extend the statute of limitations on assessment had been forged and (2) that if any deficiency were assessed, he was entitled to credit against it the $20,400 he previously had remitted to the IRS in 1984. The tax court ruled in favor of the IRS; it found that Dr. Malachinski's signature on the consent form was genuine. It further concluded that it lacked jurisdiction to determine whether Dr. Malachinski was entitled to a credit for the $20,400 remittance. For the reasons set forth in the following opinion, we affirm the decision of the tax court.

* BACKGROUND

Dr. Malachinski and his now ex-wife Wynne Superson were married in May 1980. The couple filed a joint federal income tax return for the year 1980 on April 15, 1981. In late 1982 or early 1983, the IRS contacted the Malachinskis regarding the 1980 tax return, and Dr. Malachinski sought the advice of Vincent Arnone, his tax return preparer. The Malachinskis granted power of attorney to Arnone and to attorney Eugene LaPorte in March 1983.

Superson filed for divorce in March 1983 but did not tell Dr. Malachinski until a month later. On May 23, 1983, IRS agent Alan Neubauer sent a letter addressed to the Malachinskis that asked them to sign and return Form 872-A, "Special Consent to Extend the Time to Assess Tax," with respect to their 1980 income taxes. The form permitted the IRS to extend the time period in which it had to assess any deficiencies. On or about June 8, 1983, the IRS received the form, which had been dated June 1, 1983, and bore what purported to be the signatures of both Dr. Malachinski and Superson. An IRS official countersigned the form on June 8, 1983.

Three days later, on June 11, 1983, Dr. Malachinski and Superson granted another power of attorney to John Ostrand and Stephen Mack, certified public accountants. The power of attorney was originally prepared on May 23, 1983. Superson, however, issued a separate power of attorney on September 1, 1983, naming as her representatives two attorneys, Anthony Scariano and Justino Petrarca, and a certified public accountant, Michael Moxley.

The Malachinskis were divorced in March 1984. Also in March 1984, Dr. Malachinski filed a separate tax return for his taxable year 1982. In April, acting on advice from one of his advisors, he sent to the IRS a $20,400 remittance to reduce any potential tax obligations for 1980. The IRS received the funds and prepared a voucher; the voucher contained a handwritten entry that described the payment as a "cash bond." Further, a section of the voucher was checked, indicating that the amount was an "Advance payment on deficiency." The voucher did not have a separate section to indicate that the remittance was in the nature of a cash bond or deposit.

Mack, Dr. Malachinski's representative, met with Neubauer on June 29, 1984. The IRS subsequently issued an examination report to Dr. Malachinski and Superson that indicated, in relevant part, that the two had a deficiency in income tax of $91,086 for the year 1980. Dr. Malachinski then executed a third power of attorney, naming Glenn Acquino, a certified public accountant, as his representative. Acquino duly filed a protest with the IRS on the Malachinskis' behalf.

In April 1988, the IRS transferred the $20,400 remittance from the 1980 joint account of Dr. Malachinski and Superson to Dr. Malachinski's 1982 individual tax account. In October 1988, for reasons not indicated in the record, the IRS refunded the $20,400 plus $902 in interest; Dr. Malachinski claims, however, that he never received the money. The IRS does not have a copy of the refund check nor does it have any supporting documentation. It is the IRS' policy to destroy such records after six years.

Dr. Malachinski executed a fourth power of attorney in December 1988, authorizing attorney Michael Boylan to represent him before the IRS. Boylan wrote to the IRS in August 1999, informing the agency that Dr. Malachinski had not been able to secure Superson's approval of a proposed settlement.

In late 1989, Superson began to complain bitterly about Dr. Malachinski to a friend and co-worker, Claudine Mellerke.1 Mellerke became concerned that Superson might carry out the various threats she had made against her ex- husband's life. These concerns came to the attention of local police, and Superson ultimately was convicted of violating 18 U.S.C. sec. 1958, use of interstate commerce facilities in the commission of murder for hire. (Dr. Malachinski was the intended victim of Superson's scheme.) At the time of trial in this tax matter, Superson was incarcerated in Chicago and awaiting sentencing.

On May 3, 1994, the IRS issued a notice of deficiency to Dr. Malachinski and Superson regarding their 1980 taxes, and Dr. Malachinski timely filed a petition in the tax court. He later filed an amended petition that asserted that the statute of limitations on assessment had expired before the notice of deficiency was mailed. He maintained that his signature on the form that purported to extend the limitations period had been forged, presumably by Superson. Dr. Malachinski's counsel deposed Superson in this case, but she refused to answer any questions, citing her privilege against self-incrimination.

At trial, Dr. Malachinski testified that he had not signed the consent form nor had he seen the form until 1995. He believed that Superson either had signed it or had someone else sign his name to it, but he never had authorized her to do so nor had he given her a power of attorney to act for him. To bolster these arguments, Dr. Malachinski offered the testimony of Superson's former friend and co-worker, Mellerke, who testified that Superson had told her that "she had done something to get even" with Dr. Malachinski; she had "contacted the IRS and had reported Dr. Malachinski for several things that he had not done, and also she provided a document to the IRS shortly before they were divorced that he was not aware of." R.48 at 116. According to Mellerke, Superson claimed that the document was so damaging that Dr. Malachinski would be "feeling the effects of it for the rest of his life." Id. at 116-17. Although the Commissioner objected to Mellerke's testimony as hearsay, the court reserved ruling on its admissibility.

Dr. Malachinski also introduced the expert report of Diana Marsh, a board- certified forensic document examiner. After examining twenty exemplars of Dr. Malachinski's signature, Marsh made the following conclusion in a written report:

After an examination of all the documents submitted, it is my opinion that Dr. Malachinski did not write the name "L.S. Malachinski" on the document at issue, the IRS Consent Form dated June 1, 1983. In addition, it is my opinion that the same individual wrote both dates of "6/1/83."

Ex.24.

Dr. Malachinski attempted at trial to elicit from Marsh the factual basis for her conclusion that he had not signed the form, but the IRS objected that the questions solicited information beyond the scope of the report. The court sustained the IRS' objection and restricted Marsh's testimony to the material contained in the written documentation. The court explained that Tax Court Rule 143(f) provides that the facts, data, and analysis that form the basis for an expert's conclusion are admissible only to the extent that they are set forth in a written report.

The court nonetheless permitted Dr. Malachinski to proffer additional testimony from Marsh regarding the basis and reasons for her conclusion. In the proffer, Marsh related her concerns about the letter formations in the questioned signature. Specifically, she noted characteristics of the loop on the "L" of Dr. Malachinski's first name, the angle of the bottom of the "S" in his middle name, and the absence of a loop at the end of his signature.

In response, the IRS offered the report and testimony of James Davidson, also a board-certified forensic document examiner and chief of the Questioned- Document Section of the IRS Criminal Investigation Division's National Forensic Laboratory in Chicago. Davidson compared the signature on the consent form to 26 known exemplars of Dr. Malachinski's signature. He noted that the exemplars fell into categories: 13 exemplars were "quickly written abbreviated last names;" 10 were "formally written with distinct letter definition;" and three were "shortened, quickly written signatures." Ex.AA at 2- 3. Davidson found that the exemplars from the first two categories were of "minimal value" in making the comparison and that the signature on the consent form was most similar to the third category. Id. at 3. He was, however, unable to determine whether the signature on the form was genuine; in his opinion, the three signatures in the third category were "not enough of a representative sample of the writer for [him] to make any determinations." Id.

Regarding the $20,400 remittance, Dr. Malachinski testified that he had never requested the payment to be transferred to his account for 1982 or refunded. He also denied having received a refund of that payment in 1988. On cross- examination, he indicated that he had never checked his bank records to determine whether he had deposited a similar amount at the time because he had not understood that it was the IRS' contention that the payment had been refunded. Dr. Malachinski's witness, Mellerke, testified that Superson had told her that she had received a refund from the IRS that was intended for her ex-husband but that she had cashed it and kept the money. The Commissioner objected to this testimony as hearsay, but the court again reserved ruling on its admissibility.

In a memorandum opinion filed after trial, the tax court determined that the notice of deficiency was not barred by the statute of limitations. Because the notice had been mailed more than 10 years after the 1980 return was filed-- and the statute of limitations on assessment is typically three years from the filing of the return--the court concluded that Dr. Malachinski had made a prima facie showing that the notice was untimely. The burden, therefore, fell on the IRS to establish that the bar was inapplicable. In the court's view, the IRS had met that burden by producing a facially valid Form 872-A indicating a waiver of the limitations period. Dr. Malachinski then was required to show that the agreement was invalid, a difficult task because, as the court noted, an individual's signature on a document is prima facie evidence that the individual actually signed it.

Although Dr. Malachinski had attempted to overcome this statutory presumption by presenting Marsh's testimony that his signature was forged, the court found that Marsh's report did not adequately set forth the facts and reasons supporting Marsh's conclusions. The court also noted that no showing had been made that the failure to include these materials in the report was due to good cause; it concluded, therefore, that allowing Marsh to provide additional direct testimony would have unduly prejudiced the IRS' ability to cross- examine her. It accordingly sustained the IRS' objection and restricted Marsh's direct testimony to the material set forth in her written report.

The tax court also explained that Davidson, the Commissioner's expert, had determined that the exemplars were not sufficiently representative to permit him to render an opinion regarding the authenticity of the signature. After considering both expert opinions and scrutinizing the documents itself, the court found that the signature was not a forgery. The court thus concluded that Dr. Malachinski had not overcome the statutory presumption that he signed the consent form.

In making its finding on the genuineness of the signature, the court also based its conclusion on various pieces of circumstantial evidence. For example, the court noted that Dr. Malachinski had first argued that his signature had been forged in 1995, twelve years after the form was signed. The court did not believe that, during the twelve years that followed the alleged waiver of the statute of limitations, none of Dr. Malachinski's four sets of professional advisors had consulted with him about the validity of the waiver. Rather, the court believed that Dr. Malachinski's advisors had done so and had learned from him that he had signed the consent form; otherwise, they would have sought to terminate the proceedings immediately. In the court's view, the "conduct of [Dr. Malachinski] and his advisors is thus plainly inconsistent with the claim that the consent was forged." R.56 at 13.

Moreover, typewritten portions of a power of attorney to advisors Ostrand and Mack indicated that Dr. Malachinski and Superson had contacted the accountants by May 23, 1983. The consent was executed on June 1, 1983, and, ten days later, Dr. Malachinski and Superson signed the power of attorney. The court deemed it "unlikely" that Superson would forge her husband's signature on a consent form days before cooperating with him in hiring advisors to represent them; her "duplicity would have been too easily uncovered." Id.

The court also noted that Dr. Malachinski, at the urging of his advisors, made a $20,400 payment with respect to 1980 liability in April 1984. He made the payment after the period of limitations for 1980 would have expired unless a valid waiver were in effect; competent professionals, opined the court, "would not advise a client to make a payment with respect to a tax liability which could not be collected." Id. The court further observed that Dr. Malachinski's advisors had conducted additional discussions with the Commissioner's agents in the latter part of 1984, and, even after the Commissioner sent Dr. Malachinski and Superson a letter proposing a liability of more than $90,000, the advisors made no suggestion that the collection of the taxes was time-barred. Finally, the court noted that almost seven years passed between the time Dr. Malachinski granted a power of attorney to Boylan, his attorney of record in the tax court, and the time the question of the validity of the signature was first raised.

Although Dr. Malachinski argued that Superson possessed enough animosity toward him to motivate the forgery of the consent form--a hatred evidenced in large part by her involvement in a murder-for- hire scheme against him--the court concluded that such animosity was "too attenuated and too remote in time" to support a finding that she forged Dr. Malachinski's name to the consent document. Id. at 14-15. Evidence of Superson's criminal involvement arose in 1989, six years after the form was executed and after a bitter custody battle had taken place. The court noted, too, that it would have reached the same conclusion even if it had considered Mellerke's proffered testimony in this regard. The court did not reach the IRS' hearsay objection because Mellerke's testimony was "too vague and inconclusive to be of evidentiary value in determining whether [Dr. Malachinski's] signature on the consent was forged." Id. at 15.

The tax court also held that it lacked jurisdiction to determine whether Dr. Malachinski was entitled to a credit for the $20,400 payment. It explained that its jurisdiction is limited to the determination of deficiencies and certain overpayments, and Dr. Malachinski's remittance did not fall into either category. The payment was better characterized as a deposit in the nature of a cash bond, the court concluded, a payment over which the tax court does not have jurisdiction.2

Dr. Malachinski argues on appeal that the tax court erred in determining (1) that he consented to extend the statute of limitations for assessment and (2) that the court did not have jurisdiction to determine whether he was entitled to a credit for the $20,400 remittance.

Section 6501(a) of the Internal Revenue Code ("the Code") provides that the IRS has three years from the date a return is filed to assess deficiencies. This period can be extended, however, if both the "Secretary and the taxpayer have consented in writing to its assessment after such time." I.R.C. sec. 6501(c)(4).

In asserting a statute of limitations defense, a taxpayer makes a prima facie case by showing that the notice of deficiency was not mailed within the three-year time period. See Adler v. Commissioner, 85 T.C. 535, 540 (1985). If the IRS produces a consent to extend the limitations period that is valid on its face,3 the burden then shifts back to the taxpayer to show that the consent is invalid. See id. at 540-41. The ultimate burden of proof on the limitations defense always rests on the taxpayer. See id. at 540; see also Feldman v. Commissioner, 20 F.3d 1128, 1132 (11th Cir. 1994).

In this case, Dr. Malachinski filed his tax return for the year 1980 on April 15, 1981. On June 8, 1983, prior to the expiration of the three-year statute of limitations, a representative of the Commissioner executed a consent form, purportedly signed by Dr. Malachinski and Superson, that agreed to extend indefinitely the statute of limitations on assessment for 1980. The IRS, relying on that consent, issued the notice of deficiency at issue here in May 1994.

Dr. Malachinski argues that the agreement to extend the statute of limitations is invalid because he never signed the consent form. He speculates, instead, that his ex-wife forged his signature. To overcome the presumption that a signature on a document is authentic, see I.R.C. sec. 6064, Dr. Malachinski presented at trial the expert report and testimony of Diane Marsh, a board-certified forensic document examiner. He now maintains on appeal that the tax court abused its discretion in curtailing Marsh's testimony.

Tax Court Rule 143(f)(1) requires that an expert's written opinion be submitted at least 30 days in advance of trial. The report must not only set forth the witness' opinions but also include the "facts or data on which that opinion is based" and "reasons for the conclusion." Testimony is "excluded altogether" for:

failure to comply with the provisions of this paragraph, unless the failure is shown to be due to good cause and unless the failure does not unduly prejudice the opposing party, such as by significantly impairing the opposing party's ability to cross-examine the expert witness or by denying the opposing party the reasonable opportunity to obtain evidence in rebuttal to the expert witness' testimony.

Tax Court Rule 143(f)(1).

We agree with the tax court that Marsh's written report does not satisfy Rule 143(f)(1). The report is terse and conclusory, indicating only Marsh's opinion that Dr. Malachinski had not signed his name to the consent form and that the same individual had written the date on both signature lines. Marsh does not explain her conclusions nor does she denote the facts supporting it. Moreover, Dr. Malachinski has not shown that the failure to include this information was due to good cause. In our view, the tax court was on solid ground in concluding that permitting the additional testimony would have unduly prejudiced the IRS' ability to cross-examine Marsh. See Diego Investors IV v. Commissioner, 58 T.C.M. 753, 763 (1989) (noting that "[o]ne purpose of exchanging expert reports is to facilitate effective cross-examination by the other party"); see also Estate of Friedberg v. Commissioner, 63 T.C.M. 3080, 3081-23 (1992) ("[W]hen one party seeks to introduce expert testimony, which may be exceedingly complex and difficult for the layperson to readily understand, the other party must be given sufficient time to overcome those obstacles."). The use of expert testimony is within the discretion of the trial judge, see Diego Investors, 58 T.C.M. at 764, and we see no evidence that the tax court abused its discretion in curtailing Dr. Malachinski's direct examination of Marsh.

In rebuttal to Marsh's report, the Commissioner presented the report and testimony of James Davidson, also a certified document examiner. Davidson examined various exemplars of Dr. Malachinski's signature and concluded that the samples were not representative enough for him to find that the signature was a forgery. Taking into account the conclusions of the two experts, the tax court examined the exemplars itself to reach its own conclusion that the signature was genuine. Dr. Malachinski now argues that the court committed clear error in making this determination.

As an initial matter, we note that Dr. Malachinski faces a difficult burden in contesting the tax court's factual findings. Under the clearly erroneous standard of review, a finding of fact is reversed "only when the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." Coleman v. Commissioner, 16 F.3d 821, 825 (7th Cir. 1994) (internal quotation marks omitted). We "must view the evidence in the entire record in the light which is most favorable to the finding." Pittman v. Commissioner, 100 F.3d 1308, 1313 (7th Cir. 1996) (quoting Tripp v. Commissioner, 337 F.2d 432, 434 (7th Cir. 1964)).

Although Dr. Malachinski objects to the court's substitution of its own judgment for that of his expert, the tax court is permitted to do just that. The court has broad discretion to "evaluate the overall cogency of each expert's analysis." Estate of Davis v. Commissioner, 110 T.C. 530, 538 (1998) (citation and quotation marks omitted). It can reject, in whole or in part, reports and testimony of expert witnesses in favor of its own independent evaluation of the evidence in the record. See Helvering v. National Grocery Co., 304 U.S. 282, 294-95 (1938); see also Shepherd v. Commissioner, 115 T.C. 376, 390 (2000) ("We may be selective in our use of any part of an expert's opinion."). Thus, the tax court apparently undertook an independent assessment of the signature. See, e.g., Bybee v. Commissioner, 72 T.C.M. 607, 608 (1996) (after studying handwriting exemplars, tax court concluded that the "handwriting reflected by petitioners' signatures on their 1982, 1983, and 1984 joint Federal income tax returns appears to be identical to the handwriting reflected by petitioners' signatures on the Form 872, and no credible evidence indicates to the contrary").

Dr. Malachinski also maintains that the inferences the court drew from the circumstantial evidence were not reasonable. For example, the tax court in part based its determination on the failure of Dr. Malachinski and his tax professionals to raise the issue of consent until 12 years after the form was signed.4 Dr. Malachinski argues that he received a letter from the IRS in 1990 advising him that the 1980 tax dispute had been resolved; he had no reason, therefore, to pursue the issue of consent. He also asserts that he relied on his lawyers to defend the tax return and did not even know that the forged waiver existed.

Again, however, we are not persuaded that the court clearly erred in making its factual findings. As we have noted previously, the existence of evidence to support an inference contrary to that drawn by the trier of fact does not mean that the findings were clearly erroneous. See Harper v. City of Chicago Heights, 223 F.3d 593, 600 (7th Cir. 2000), cert. denied, 121 S. Ct. 883 (2001); see also Malkin v. United States, 243 F.3d 120, 123-24 (2d Cir. 2001) (rejecting petitioner's contention that the evidence was insufficient to support the district court's finding that he signed an agreement to extend the statute of limitations on assessment). A fact finder's choice between two permissible inferences from the evidence cannot be clearly erroneous. See Anderson v. Bessemer City, 470 U.S. 564, 574 (1985).

Dr. Malachinski also takes issue with the tax court's determination that his ex-wife did not forge his signature. We conclude, however, that the court was well within its province when it found that Superson's hostility toward Dr. Malachinski--as manifested in her attempt to hire someone to kill him--was too far removed in time (six years after the execution of the consent form) to support the forgery theory. Although Dr. Malachinski marshals evidence to the contrary (Superson secretly had filed for divorce; the two had been embroiled in a heated custody battle; Mellerke testified that Superson had made several suspicious statements; Superson was convicted of attempting to hire someone to kill Dr. Malachinski; and, in a deposition related to this case, Superson refused to answer, on self-incrimination grounds, questions relating to whether she forged Dr. Malachinski's signature), we review the record in the light most favorable to the tax court's findings and, as indicated previously, defer to those findings when there are two permissible views of the evidence. See United States v. Hardamon, 188 F.3d 843, 848 (7th Cir. 1999).

Dr. Malachinski sent the IRS a remittance of $20,400 in April 1984 in anticipation of income tax liability for 1980. Notably, at the time Dr. Malachinski made this payment, the IRS had not defined any tax liability for 1980. Although the taxes were under audit, no examination report proposing a deficiency had been prepared. Indeed, the deficiency was not determined until ten years later. In 1988, the IRS transferred the payment to Dr. Malachinski's individual 1982 income tax account. The money, plus $902 in interest, was refunded by the IRS six months later.

Dr. Malachinski argues that he did not request either the credit of the funds to his 1982 taxable year or the refund of that amount. He denies, moreover, that he ever received the refund; he speculates that Superson intercepted the check and forged his endorsement. Thus, even if he does not prevail on the statute of limitations defense, Dr. Malachinski asserts that he is entitled to a credit of $20,400 against the deficiency he now owes on his 1980 tax return.

The tax court determined that it was without jurisdiction to determine whether Dr. Malachinski was entitled to a credit against the 1980 deficiency. The court reasoned that, when the payment was initially made, it was not a payment toward a 1980 liability and therefore ought to be considered a deposit rather than a payment to satisfy a particular tax liability. Later, the IRS applied the funds to Dr. Malachinski's 1982 tax liability and, in the course of determining the extent of that 1982 liability, returned these funds as an overpayment. Because the tax court did not have jurisdiction over Dr. Malachinski's 1982 tax liability in this action, the court could not review the correctness of the IRS' disposition of the funds with respect to that year. Therefore, at the time it was asked to apply the funds to the 1980 tax liability, there were no funds within the court's jurisdiction to apply to that liability.

In reviewing the determination of the tax court, we turn first to its initial determination that, at the time of their receipt by the IRS, the funds were a deposit rather than a payment on the 1980 taxes. In Moran v. United States, 63 F.3d 663 (7th Cir. 1995), this circuit joined a significant number of other circuits in concluding that a court must look to the facts and circumstances of an individual case to determine whether a remittance is a deposit or a payment. See id. at 667-68. We held that "for the purposes of deciding whether a remittance was a payment of tax, formal assessment is only one factor to be considered." Id. at 668 (quoting Ewing v. United States, 914 F.2d 499 (4th Cir. 1990), cert. denied, 500 U.S. 905 (1991)). Other factors to be considered include the taxpayer's intent upon making the remittance, how the IRS treats the remittance upon receipt, and when the tax liability is defined. See Moran, 63 F.3d at 668.

In Moran, we also took the view, without extended discussion, that we ought to assess these factors under a de novo standard of review. Further reflection has convinced us that a deferential standard is more appropriate. Recent years have seen a growth in our national jurisprudence on the appropriate standard of review for the application of legal principles to fact-specific questions. The Supreme Court has made it clear that plenary review is appropriate when a relevant legal principle can be given meaning and clarified only through the application of that rule to the circumstances of a particular case and when there are particularly important reasons for the courts to craft a defined set of rules that ensure that future determinations involving important rights are made accurately. See Ornelas v. United States, 517 U.S. 690, 697-98 (1996); see also Cooper Indus., Inc. v. Leatherman Tool Group, Inc., 121 S. Ct. 1678, 1685-88 (2001). However, absent such special considerations, a proper allocation of the roles of trial and appellate courts counsels that fact- specific questions not easily susceptible of useful generalization be reviewed deferentially by appellate courts. See Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 399-405 (1990); Pierce v. Underwood, 487 U.S. 552, 557-63 (1988). The application of the "facts and circumstances" test that we embraced in Moran indicated strongly that we ought to review the determination of the tax court on this question deferentially. On this point--the applicable standard of review--our language to the contrary in Moran is overruled.5

Applying a deferential standard of review to the determination of the tax court in this case, we see no reason to disturb that court's conclusion. The record adequately supports the conclusion that the tax court correctly determined the payment in this case, at the time it was made, was a deposit. As we already have noted, Dr. Malachinski's payment was made in April 1984, well before any liability was defined. At that time, his 1980 tax return was being audited, but no report proposing a deficiency was prepared until 15 months later, and the deficiency was not determined until 10 years later. The amount of the remittance bore, in the words of the tax court, "no perceptible relationship" to the amount of the deficiency proposed in the examination report (more than $90,000). R.56 at 20. Dr. Malachinski, additionally, did not indicate that the remittance constituted a payment; rather, he recalled making it to halt the accrual of interest.6

Further, the IRS apparently treated the remittance as a cash bond. When it refunded the remittance, it included $902 in interest--a reflection of the accrual of interest for the six months the money was in the 1982 account and not for the total four and-a-half years the IRS had the money. The payment of interest only for the time that the money was posted to the 1982 account indicates that the IRS treated the payment as a cash bond for the year 1980.7

Although the tax court has jurisdiction to determine whether a deposit payment is applicable to a particular deficiency, see Hays v. Commissioner, 71 T.C.M. 1754, 1757-58 (1996), here, the court was never asked to determine whether the deposit in question could be attributed to the 1980 tax year at any time prior to the IRS' application of those funds to the 1982 tax account in 1988. Whether those funds were properly allocated to the 1982 tax account was a matter not properly before the tax court in this proceeding because the court did not have jurisdiction over the 1982 tax account. See I.R.C. sec. 6214(b). At first glance, section 6512(b) seemingly might provide a jurisdictional hook to permit the tax court to reach this question. That section states:

[I]f the Tax Court finds that there is no deficiency and further finds that the taxpayer has made an overpayment of income tax for the same taxable year . . . in respect of which the Secretary determined the deficiency, or finds that there is a deficiency but that the taxpayer has made an overpayment of such tax, the Tax Court shall have jurisdiction to determine the amount of such overpayment, and such amount shall, when the decision of the Tax Court has become final, be credited or refunded to the taxpayer.

We cannot read this provision as permitting the tax court to determine whether the $20,400 deposit ought to be attributed to Dr. Malachinski's 1980 tax deficiency. First, sec. 6512(b) refers to overpayments, and, as we just held, with respect to the 1980 tax year, the remittance at issue here is a deposit and not a payment. Although the term "overpayment" is nowhere statutorily defined, see Estate of Baumgardner v. Commissioner, 85 T.C. 445, 449 (1985), the Supreme Court has defined overpayment as "any payment in excess of that which is properly due." Jones v. Liberty Glass Co., 332 U.S. 524, 531 (1947) (defining overpayment as used in the statutory predecessor to sec. 6512(b), sec. 322 of the 1939 Code). For there to be an overpayment, the taxpayer first must have made a payment. See Bachner v. Commissioner, 109 T.C. 125, 129 (1997), aff'd without published opinion, 172 F.3d 859 (3d Cir. 1998). Because Dr. Malachinski has not made a payment for the 1980 tax year, it follows that he could not have made an overpayment that would have brought him within the purview of sec. 6512(b).

Even if Dr. Malachinski's remittance could be considered a payment rather than a deposit when it was transferred and credited to his 1982 account, sec. 6512(b)(4) nevertheless prevents the tax court from exercising jurisdiction over the remittance. This recently added subsection of I.R.C. sec. 65128 reads as follows:

(4) Denial of Jurisdiction Regarding Certain Credits and Reductions.--The Tax Court shall have no jurisdiction under this subsection to restrain or review any credit or reduction made by the Secretary under section 6402.9

The Senate Report explains that the addition "clarifies that the Tax Court does not have jurisdiction over the validity or merits of the credits or offsets that reduce or eliminate the refund to which the taxpayer was otherwise entitled." S. Rep. 105-33, 105th Cong., 1st Sess. 302-303 (1997).Although the tax court has jurisdiction to determine the amount of an overpayment, see I.R.C. sec. 6512(b)(1), it does not have jurisdiction to direct the disposition of an overpayment when that payment has been credited against another year's assessment pursuant to sec. 6402(a) prior to the commencement of a tax court proceeding. See I.R.C. sec. 6512(b)(4); Savage v. Commissioner, 112 T.C. 46, 48- 51 (1999).

The tax court did not clearly err when it determined that Dr. Malachinski's signature on the consent form was genuine. We also agree that the tax court lacked jurisdiction to determine whether Dr. Malachinski was entitled to a credit for the $20,400 remittance. Accordingly, we affirm the decision of the tax court.

Posner, Circuit Judge, concurring in part and dissenting in part.

I agree with the majority's analysis and disposition of the taxpayer's first ground of appeal and with its conclusion anent the second ground that the remittance to the IRS was a deposit and not a payment of taxes. I also agree that the Tax Court's determination that it was a deposit should not be reviewed "de novo," that is, without deference to the Tax Court's view of the matter, though I think we could say a bit more on that subject. But I disagree with the majority's holding that the Tax Court lacked jurisdiction to apply the taxpayer's deposit toward his tax deficiency--and let me begin with this issue.

The majority points out that the Tax Court's jurisdiction is limited to the resolution of disputes over deficiencies, 26 U.S.C. sec.sec. 6213(a), 6214--that is, underpayments--but that its power to resolve such disputes includes the power to determine whether a taxpayer is entitled to a refund of payments made in excess of a deficiency, 26 U.S.C. sec. 6512(b), that is, to a refund of an overpayment. So far, so good. But when a taxpayer makes a deposit against a possible liability, so that there is only a potential deficiency, the majority holds that the Tax Court has no jurisdiction to apply the deposit against the deficiency even if the court later determines, as it did in this case, that, yes, there is a deficiency against which the deposit could be credited to reduce the taxpayer's net deficiency. The consequence of my colleagues' position is that any time the IRS refuses to acknowledge that a taxpayer against whom it has assessed a deficiency had made a deposit against that deficiency, the taxpayer must, if he wishes to challenge both the deficiency and the refusal to credit his deposit against it, file two separate suits: one in the Tax Court to litigate his deficiency and a second in a federal district court or the U.S. Court of Federal Claims (28 U.S.C. sec.sec. 1346(a)(2), 1491(a)(1); see Tosello v. United States, 210 F.3d 1125, 1128 (9th Cir. 2000)) to force the IRS to return his deposit. See Rosenman v. United States, 323 U.S. 658 (1945); Ertman v. United States, 165 F.3d 204, 206 (2d Cir. 1999); New York Life Ins. Co. v. United States, 118 F.3d 1553, 1555-58 (Fed. Cir. 1997).

Nothing compels so inefficient a procedure for litigating disputes over deposits against potential deficiencies. The motive for a taxpayer's depositing money with the IRS is his realization that there is a potential tax deficiency (in other words, that he may have underpaid his taxes) coupled with a desire to stop interest and penalties from accruing. See VanCanagan v. United States, 231 F.3d 1349, 1353 (Fed. Cir. 2000); Callaway v. Commissioner, 231 F.3d 106, 113 n. 11 (2d Cir. 2000); IRS Rev. Proc. 84-58 sec. 5.01, 1984-2 C.B. 501, sec. 5.01. Should a deficiency eventually be assessed, the deposit is then used to pay it off. See id. sec.sec. 4.02(2), (3); Michael I. Saltzman, IRS Practice and Procedure sec. 6.02[3] [a][ii], p. 6-15 (2d ed. 1991). In determining a taxpayer's net deficiency, there- fore, the Tax Court, as a matter of elementary judicial economy, ought to be able to determine the status of a deposit. That has been the court's steady practice, see, e.g., Hays v. Commissioner, T.C. Memo 1996-18, 1996 WL 20554 (U.S. Tax Court Jan. 22, 1996), from which it unaccountably departed in this case with an unelaborated citation to an irrelevant case, Savage v. Commissioner, 112 T.C. 46 (1999). The power for which I am arguing is entirely consistent with the limitation of the Tax Court's jurisdiction to disputes over deficiencies (and overpayments in connection therewith), with the case law, and with common sense.

The question is not whether the Tax Court has "jurisdiction over a deposit." It has jurisdiction to determine the amount of deficiencies. This can be done only by taking the amount of taxes owed and subtracting any payments toward them. No one would say that the Tax Court has "jurisdiction over a tax payment"; it just uses the payment to compute the amount owed, and if there are disputes about how much has been paid the Tax Court must resolve them. The same is true if the remittance is not a payment of tax but instead a deposit against a potential tax liability.

Against this my colleagues make two points. The first is that a deposit cannot be an overpayment because, "for there to be an overpayment, the taxpayer first must have made a payment." That isn't true as a matter of ordinary usage, and my colleagues' opinion gives no reason why it should be interpreted this way. If you make a deposit on the purchase of a lamp, not knowing the price, and the price turns out to be less than the deposit, the seller will refund the difference-- you've "overpaid." The second point made by my colleagues relates to 26 U.S.C. sec. 6512(4)(b), which forbids the Tax Court "to restrain or review any credit or reduction made by the Secretary [of the Treasury, i.e., the IRS] under section 6402." Section 6402(a) authorizes the IRS to credit an overpayment of taxes for one tax year against a deficiency in the taxes paid by the taxpayer in another tax year, and so is inapplicable to this case because the deposit (overpayment) was not made on account of a different year from the year for which the deficiency was assessed; the deposit was against taxes assessed for that year. It is not surprising that the government didn't even bother to cite section 6402 in its original brief, or for that matter section 6512(4)(b). What is more, as is apparent from a comparison of section 6512(4)(b) with sections 6402(b) and (c), and from the legislative history of section 6512(4)(b), see S. Rep. No. 33, 105th Cong., 1st Sess. 302-03 (1997); H.R. Rep. No. 48, 105th Cong., 1st Sess. 637-38 (1997), the section was enacted in order to prevent the Tax Court from second-guessing the IRS's determination to reduce a refund by the amount of child-support payments or payments due on government loans. It has nothing to do with deposits.

Turning now to the proper standard of appellate review of a finding that a remittance was a deposit rather than a payment of taxes, I agree, as I said at the outset, that it should be the clearly-erroneous standard and not the de novo (no deference) standard of Moran v. United States, 63 F.3d 663, 666 (7th Cir. 1995). It is worth noting that the court adopted that position in Moran without elaboration and without explanation beyond a bare citation to a case that says something entirely different and completely irrelevant, which is that review of summary judgment is de novo. Hefti v. IRS, 8 F.3d 1169, 1171 (7th Cir. 1993). Moran was also a summary judgment case, and it is possible that the reference in it to review de novo was intended to say nothing more than what Hefti had (correctly) said, though the specific language of Moran is against this interpretation: "All these contentions essentially hinge on one issue: whether the remittances made by the Morans in 1985 and 1986 were payments of tax or deposits that the IRS converted into payments in September, 1991. We review the district court's decision on this legal question de novo." However all this may be, the Sixth Circuit, in a case decided after Moran, reached the opposite conclusion, Gabelman v. Commissioner, 86 F.3d 609, 611 (6th Cir. 1996), and so we have a circuit split, which behooves us to revisit Moran. See United States v. Carlos- Colmenares, 253 F.3d 276, 277-78 (7th Cir. 2001).

The usual significance of the distinction between the payment of taxes and a deposit against a potential deficiency is, as in Moran itself, 63 F.3d at 666-67, 670; Ertman v. United States, supra, 165 F.3d at 206; and Blatt v. United States, 34 F.3d 252, 253-54 (4th Cir. 1994), that to get back an overpayment of taxes you must file a refund suit, and there are limitations (including how much time one has to sue), see 26 U.S.C. sec. 6511, that are not applicable to claims for the return (or crediting) of deposits. So the question of classification is significant even if, as I believe, my colleagues are wrong to think it has jurisdictional significance in this case.

As the majority points out, which animal--tax payment or deposit--a particular remittance is depends on "the facts and circumstances of an individual case," including such facts as "the taxpayer's intent upon making the remittance, how the IRS treats the remittance upon receipt, and when the tax liability is defined." (This has been dubbed "the 'facts and circumstances' test." Ertman v. United States, supra, 165 F.3d at 207; see also Gabelman v. Commissioner, supra, 86 F.3d at 613.) When a legal conclusion is based on a kaleidoscope of factual determinations, appellate courts generally review it under the clearly- erroneous standard, recognizing that the trier of fact is in a better position than the appellate judges to weigh up the various facts and that the primary appellate role is to assure legal uniformity, an unattainable goal when the dispositive issue is case-specific. See, e.g., Buford v. United States, 121 S. Ct. 1276 (2001); Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 401-04 (1990); Icicle Seafoods, Inc. v. Worthington, 475 U.S. 709, 711-14 (1986); United States v. Hill, 196 F.3d 806, 808 (7th Cir. 1999); Cook v. City of Chicago, 192 F.3d 693, 696-97 (7th Cir. 1999); United States v. Frederick, 182 F.3d 496, 499-500 (7th Cir. 1999); United States v. Rule Indus- tries, Inc., 878 F.2d 535, 538, 541-42 (1st Cir. 1989), and with reference to review of such determinations in tax cases, Williams v. Commissioner, 1 F.3d 502, 505 (7th Cir. 1993), where we nted that because "a question about the proper application of a legal standard to the (real) facts . . . requires a judgment based on the idiosyncratic facts of a particular case rather than the formulation of a general rule, it is one best confided to the first-line judicial officer subject to only light appellate review." A ruling in this case that Malachinski's payment was a deposit has no precedential significance, because there will never be a case with the same facts. The rule that overpayments and deposits are treated differently in tax law is a genuine rule of law; the "rule" that Malachinski loses this case is not.

There is an analogy to a jury's or a trial judge's finding of negligence. Even if the facts are undisputed, the inference from them that the defendant was or was not negligent is drawn by the trier of fact and reviewed deferentially. The Supreme Court so held in McAllister v. United States, 348 U.S. 19 (1954); see 9A Charles A. Wright & Arthur R. Miller, Federal Practice and Procedure sec. 2590, pp. 622-24 (2d ed. 1995); see also Cooter & Gell v. Hartmarx Corp., supra, 496 U.S. at 402; Halek v. United States, 178 F.3d 481, 485 (7th Cir. 1999); Mucha v. King, 792 F.2d 602, 605 (7th Cir. 1986); Ogden v. Commissioner, 244 F.3d 970 (5th Cir. 2001) (per curiam). The inference that Malachinski's remittance was a deposit rather than a tax payment would likewise be for the Tax Court to draw subject to light review by us even if the underlying facts--Malachinski's intent, what the IRS did with his check, and so forth--were stipulated.

There are, it is true, exceptions to the principle that legal inferences from facts, or what are more commonly called rulings on "mixed questions of law and fact" or "ultimate questions of fact," are to be reviewed for clear error. The exceptions have mainly to do with sensitive issues of constitutional law or jurisdiction, see, e.g., Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 121 S. Ct. 1678, 1685-86 (2001); Ornelas v. United States, 517 U.S. 690, 696-99 (1996), and so are inapplicable here. Cf. United States v. Frederick, supra, 182 F.3d at 499-500; Door Systems, Inc. v. Pro-Line Door Systems, Inc., 126 F.3d 1028, 1031 (7th Cir. 1997). Al- though there is a regrettable lack of uniformity in the practice of appellate courts, as pointed out in United States v. Frederick, supra, 182 F.3d at 503-04 (concurring opinion); see also Witkowski v. Welch, 173 F.3d 192, 198 n. 7 (3d Cir. 1999), this circuit has adhered to the principle quite steadily--see, e.g., besides the cases cited earlier, In re Rovell, 194 F.3d 867, 870-71 (7th Cir. 1999); United States v. Frederick, supra, 182 F.3d at 499; and Mars Steel Corp. v. Continental Bank N.A., 880 F.2d 928, 933-36 (7th Cir. 1989) (en banc)--and it is fully applicable to the present case.

The government has conceded that if the Tax Court did have jurisdiction to apply the taxpayer's deposit to his deficiency, the case must be remanded. The IRS contends that it returned the deposit to the taxpayer, and if that is right then of course he is not entitled to credit it against his deficiency. But that is a factual question for the Tax Court to answer. I would therefore remand the case to the Tax Court.