dissenting:
Today the majority announces that it is overruling nearly 16 years of precedent in this Court. See Yusko v. Commissioner, 89 T.C. 806, 808 (1987); Pyo v. Commissioner, 83 T.C. 626, 634 (1984); Weinroth v. Commissioner, 74 T.C. 430, 436-437 (1980); Alta Sierra Vista, Inc. v. Commissioner, 62 T.C. 367, 376-377 (1974), affd. without published opinion 538 F.2d 334 (9th Cir. 1976); Lifter v. Commissioner, 59 T.C. 818, 822 (1973); Budlong v. Commissioner, 58 T.C. 850, 852-853 (1972). The principles of these prior opinions still appear to be generally supported by a majority of the courts of appeals. See King v. Commissioner, 857 F.2d 676, 679 n. 5 (9th Cir. 1988) (noting that the Tax Court and a majority of the courts of appeals have ruled that a subsequent return is not wholly determinative of a taxpayer’s last known address). The majority, while continuing to adhere to a rationale requiring respondent to “exercise reasonable diligence in ascertaining the taxpayer’s correct address” (see Alta Sierra Vista, Inc. v. Commissioner, supra at 374), nevertheless finds that respondent was unreasonable on November 30, 1982 (the date the notice of deficiency was mailed) in determining petitioner’s last known address in conformity with the very rules we have prescribed prior to today. Exacerbating this situation is the fact that both the old and new rules for determining a taxpayer’s last known address are largely procedural in nature, making it of utmost importance that the controlling rules provide a “bright line” so that respondent has “clear guidance as to what information it must examine in determining a taxpayer’s last known address.” King v. Commissioner, supra at 680; United States v. Zolla, 724 F.2d 808, 811 (9th Cir. 1984).
I do not disagree with the trial judge’s fact finding regarding the technological advances in respondent’s computer capabilities nor do I advocate unswerving adherence to past precedent for its own sake. Although the majority’s new rule will require respondent to increase significantly his use of computer resources, I do not find anything in the record indicating that these new requirements are unreasonable if they are prospective requirements. However, invalidation of the notice of deficiency in issue, which was mailed 6 years ago in accordance with procedures we upheld until today, is inequitable and inconsistent with any rule whose foundation is based upon the required exercise of reasonable diligence.
Proper resolution of this case and implementation of the majority’s new rule requires a consideration of competing objectives. On the one hand, the rules for determining a taxpayer’s last known address are designed to provide “petitioner with prompt and fair notice of the deficiencies.” Alta Sierra Vista, Inc. v. Commissioner, supra at 377. The best methods for achieving this objective will necessarily change with technological advancements. Competing with the objective of giving actual notice is the recognition that respondent should not be required to exhaust all possibilities in ascertaining a last known address. The very structure of section 6212(b)(1) recognizes that actual notice is not always reasonably possible and that actual receipt of the notice is not a prerequisite to validity. King v. Commissioner, supra at 679. Finally, given the numerous possible methods that might be used to determine a last known address, the rules for ascertaining a taxpayer’s last known address should be clear and serve as a rehable guide to the respondent. King v. Commissioner, supra at 680. The application of new and different rules to notices of deficiency mailed in prior years, in accordance with our then-outstanding precedents, totally ignores these last two considerations and makes our opinions an unreliable source of guidance. Following the majority’s approach, it is possible that 6 years from now we might invalidate a notice issued this year in conformity with the majority’s new rule by finding that there existed even better methods for determining a last known address.
A solution is to apply the majority’s new rules for determining a taxpayer’s last known address prospectively only. The primary authority for limiting the retroactive application of judicial decisions is Chevron Oil Co. v. Huson, 404 U.S. 97 (1971).1 In Chevron, the plaintiff commenced an action for injuries he sustained while working on an offshore oil rig. The action was filed several years after the alleged injury. Subsequent to filing the suit in Chevron, the Supreme Court held in Rodrique v. Aetna Casualty & Surety Co., 395 U.S. 352 (1969), that the Federal law, upon which the claim in Chevron was based, adopted the law of the adjacent State instead of general admiralty principles. The Court in Chevron held that the law of the adjacent State included the State statute of limitations, which in Chevron ran 1 year after the date of injury. Application of the State statute of limitations would have totally barred plaintiffs claim. Noting that the plaintiff filed his action relying on pre-Rodrique precedent under which his suit would have been timely, the Court stated:
When the respondent was injured, for the next two years until he instituted his lawsuit, and for the ensuing year of pretrial proceedings, these Court of Appeals decisions represented the law governing his case. It cannot be assumed that he did or could foresee that this consistent interpretation of the Lands Act would be overturned. The most he could do was rely on the law as it then was. “We should not indulge in the fiction that the law now announced has always been the law and, therefore, that those who did not avail themselves of it waived their rights.” Griffin v. Illinois, 351 U.S. 12, 26 (Frankfurter, J., concurring in judgment). [Chevron Oil Co. v. Huson, 404 U.S. at 107.]
Chevron sets forth three separate factors to be considered when evaluating the extent to which a new rule will not be applied retroactively. “First, the decision to be applied nonretroactively must establish a new principle of law, either by overruling clear past precedent on which litigants may have relied, * * * or by deciding an issue of first impression whose resolution was not clearly foreshadowed.” Chevron Oil Co. v. Huson, supra at 106. There is no doubt that we are overruling “clear past precedent” in this Court. Presently, most courts have said that respondent is required to consult computer records for the address on the taxpayer’s most recent return only when he knows, or should know, based on all the facts and circumstances, that the taxpayer has moved. See Mulder v. Commissioner, 855 F.2d 208, 212 (5th Cir. 1988), revg. a Memorandum Opinion of this Court; Wallin v. Commissioner, 744 F.2d 674, 677 (9th Cir. 1984), revg. a Memorandum Opinion of this Court; McPartlin v. Commissioner, 653 F.2d 1185, 1190 (7th Cir. 1981); Crum v. Commissioner, 635 F.2d 895, 899-900 (D.C. Cir. 1980). While the majority cites cases indicating that we might at some point consider new requirements for determining a taxpayer’s last known address (majority opinion at pp. 1033-1034), none of our opinions clearly foreshadow the specific rules laid down in the majority opinion.
The majority holds that the address appearing on a taxpayer’s most recently filed return is his or her last known address and that this address “is available to the agent issuing a notice of deficiency with respect to a previously filed return, if such address could be obtained by a computer generation of an IRS computer transcript using the taxpayer’s TIN [social security number] in the case of a separately filed return, or both taxpayers’ TINS in the case of a previously filed joint return.” (Majority opinion at p. 1035.) The majority does not cite, and I am unaware of, any cases from other courts requiring respondent to ascertain joint filers’ last known addresses by not only searching for subsequently filed returns using the primary taxpayer’s social security number, but also by conducting an independent search for subsequent separate returns that might possibly have been filed by the spouse even though the spouse gave no other indication that he/she was maintaining a separate residence. Indeed, the majority’s new rule also goes beyond any decision previously rendered by the one court that has consistently differed with the Tax Court in this area, the Ninth Circuit Court of Appeals.
In Wallin v. Commissioner, 744 F.2d 674 (9th Cir. 1984), the Ninth Circuit was faced with a taxpayer who had filed an individual return for 1977, after which she married, changed her name, moved to another address, and filed joint returns with her new husband for 1978 and .1979 in which her social security number and new last name appeared in the spousal box while her husband’s appeared in the primary box. Following an unsuccessful effort by respondent to locate the taxpayer, respondent mailed a notice of deficiency for 1977 to the address listed on the 1977 return. The notice was returned as “unclaimed.” The taxpayer argued that the subsequent filing of joint returns was sufficient, standing alone, to notify the IRS of her new address in light of the Ninth Circuit’s rule that a new address on a subsequent return is the taxpayer’s last known address. The Court, while finding that “The IRS has the ability to perform a computer search of both sets of social security numbers in order to discover a subsequent return filed under a different name or in a different state” nevertheless found that the taxpayer’s “subsequent return bore a name different from that contained in the 1977 tax return and offered almost no notice to the IRS.” Wallin v. Commissioner, supra at 677. In Wallin, the Ninth Circuit clearly indicated that it was unnecessary to conduct a search for a taxpayer’s last known address using his or her social security number to see if it may have been listed as a spousal social security number, unless the IRS was otherwise notified of a change of address. The majority’s ruling requires respondent to always check his computer files for spousal social security numbers in situations where the filing sequence is the same as that presented in Wallin, even though respondent has no notice of a change of address. While the facts in Wallin are distinguishable from the instant case, that opinion also raises questions about how the Circuit Court would deal with the instant case.
Aside from overruling clear past precedent, the majority opinion also decides an issue that is before this Court for the first time. Respondent argued that a literal interpretation of section 6212(b)(2) supports his decision in this case to not conduct an independent search absent specific notice to him “by either spouse that separate residences have been established.” Sec. 6212(b)(2); emphasis added. Petitioner’s most recent return shows the address of her accountants and does not purport to indicate that it was her residential address. The majority opinion cites no precedent rejecting (or even discussing) respondent’s literal application of the statute nor any opinion “clearly foreshadowing” the result reached by the majority. Indeed, I think even the majority would concede respondent is “literally” correct. Under these circumstances, the majority’s ruling was not clearly foreseeable nor could respondent’s approach to the “last known address” determination be branded unreasonable.
The second factor considered in Chevron requires “looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation.” Chevron Oil Co. v. Huson, supra at 107. The purpose of the rule announced by the majority is to enhance the likelihood of actual notice by requiring respondent to use his present computer capabilities when determining the taxpayer’s last known address. At the same time, one of the concepts behind the statutory provision is that actual notice is not required so long as respondent’s method for determining a taxpayer’s last known address is reasonable. Nothing can now be done to improve past determinations of last known addresses. The issue is whether respondent was reasonable. Retroactive application as a sanction is inappropriate where respondent adhered to procedures which we had previously found sufficient. The legislative purpose of prohibiting the assessment of a deficiency prior to giving notice of a deficiency could arguably be served by retroactive application, but then we would have to ignore the legislative intent, that it is not actual notice, but only the exercise of reasonable diligence by respondent that is required.
The third factor to be considered in a retroactivity determination is the weighing of the inequities that would result. Chevron Oil Co. v. Huson, supra at 107. The majority’s opinion will invalidate an unknown number of notices of deficiency issued in conformity with our prior opinions and will result in the wholesale dismissal of an unknown number of cases on jurisdictional grounds without ever considering the underlying facts and law determinative of the actual tax liability. It is also likely that the statute of limitations will prevent respondent from reissuing notices in the vast majority of such cases.2 Respondent conformed his conduct to the requirements set forth in our prior opinions, and it is inequitable to conclude that he should be barred from the opportunity of ever obtaining a correct determination of tax based upon the individual merits of these cases.3 This result is particularly inequitable considering the procedural nature of the majority’s new rule. It is true that taxpayers such as petitioner also suffer a disadvantage if the rule is not applied retroactively because nonretroactive application will result in loss of a prepayment forum. Taxpayers, however, may still obtain a resolution of their cases on the merits, in that they still have an opportunity for an administrative disposition with respondent and, if that proves unsuccessful, the merits can be adjudicated in a refund suit.
Another equitable factor to consider is the significant increase in the burdens being placed on respondent. The majority’s broad based rule will require computer searches using social security numbers of all taxpayers in all situations. The majority and the Ninth Circuit in Wallin v. Commissioner, supra, recognized that it was respondent’s long-standing practice normally to restrict his search for later filed returns to one looking for primary taxpayers’ social security numbers. If the taxpayer appeared as the spousal taxpayer on a subsequent return, this type of search would not reveal it. This general practice has never previously been criticized in the absence of other factors putting respondent on notice that the taxpayer has moved. Under the new rule, it seems clear that whenever a notice of deficiency is to be mailed regarding a joint liability, the respondent will be required, in all cases, to search for subsequently filed returns using both the primary and spousal social security number. It will also be necessary for the search to cover the possibility that either of the prior spouses might have remarried and been listed as either primary or spousal taxpayers on their subsequent joint returns. Notices of deficiency regarding individual returns will also require similar searches for subsequently filed joint returns where the individual’s name may have changed and/or the individual is listed as the spousal taxpayer, rather than the primary taxpayer. It is likely that all of the possible search variations will frequently be necessary since taxpayers can, but frequently do not, file early. For example, a computer search for a primary taxpayer made on April 1, that fails to show a return for the prior year, may mean that no return has yet been filed or processed, but may also mean that the taxpayer is listed as a spouse on a recently filed joint return. Considering that over one hundred million individual income tax returns are filed each year (see Internal Revenue Service Annual Report 1987), the increased obligations imposed by the majority’s rule, and the possibility that significant system changes are necessary, it is inequitable to invalidate notices of deficiency issued prior to today’s ruling in conformity with our prior opinions.
A final equitable consideration that applies to all taxpayers is that retroactive application of the majority’s new rule will result in disparate treatment of similarly situated taxpayers depending upon whether their case was disposed of before or after the announcement of our new standards.
Regarding the equities in this particular case, the facts do not support special consideration for petitioner. Petitioner apparently relied totally upon her ex-husband to handle her tax matters, even to the extent of being ignorant about whether returns were ever filed. When petitioner separated from her husband, she apparently made no provisions for ascertaining the status of her tax obligations for prior years and failed to notify respondent that she had changed residences. The fact that she filed a subsequent return using the filing status of married filing separately and giving her address as “do: Segal, Goldman & Macnow, Inc.,” her accountants, gave no notice that petitioner was separated or had changed her residence, nor, in my opinion did it constitute “clear and concise notification from the taxpayer directing the Commissioner to use a different address” with respect to correspondence involving prior years. See Alta Sierra Vista, Inc. v. Commissioner, 62 T.C. 367, 376-377 (1974), affd. without published opinion 538 F.2d 334 (9th Cir. 1976); see Wallin v. Commissioner, supra at 676. It is only after the majority’s establishment of a “bright line” test that petitioner’s subsequent return in this case can be said to constitute sufficient notification of a change in her last known address for the years in issue.
For the reasons set forth above, I dissent from that portion of the majority’s opinion which would apply its newly announced rules for determining a taxpayer’s last known address to notices of deficiency mailed prior to the date of the opinion in this case.4
Parker, Gerber, and Williams, JJ., agree with this dissent.A very recent application of the Chevron case by the Court which has appellate jurisdiction over this case is contained in Austin v. City of Bisbee, Arizona, 855 F.2d 1429 (9th Cir. 1988).
A notice of deficiency incorrectly addressed and not received by a taxpayer prior to the 90th day after its mailing does not toll the running of the statute of limitations. Boren v. Riddell, 241 F.2d 670, 671 (9th Cir. 1957); Welch v. Schweitzer, 106 F.2d 885 (9th Cir. 1939); Reddock v. Commissioner, 72 T.C. 21, 26 (1979); Rodgers v. Commissioner, 57 T.C. 711, 713 (1972).
As previously noted, it is far from clear how the Ninth Circuit will rule if this case is appealed. In any event, the opinions definitively setting forth the Ninth Circuit rule were decided subsequent to issuance of the notice of deficiency in this case. Wallin v. Commissioner, 744 F.2d 674 (9th Cir. 1984), revg. a Memorandum Opinion of this Court; United States v. Zolla, 724 F.2d 808 (9th Cir. 1984).
There is no inherent impediment to the announcement of a new rule, while at the same time refusing to apply it to the parties in that case. See Great Northern Railway Co. v. Sunburst Oil & Refining Co., 287 U.S. 358 (1932). See also Chevron Oil Co. v. Huson, 404 U.S. 97 (1971).