Washington-based businesses tend to focus on Washington’s business taxes, because those taxes are typically the most relevant. However, as businesses grow into interstate business activities, they can become subject to another state’s income tax jurisdiction … like California. If a business is taxable in another state, then that state will typically apportion the interstate revenue to itself in a proportion that is determined by a formula. This formula is a proxy for your activities in that other state that contributed to your interstate income. The U. S. Constitution allows states to demand a tax that is fairly related to a taxpayer’s business activity in that state.
For many years, states have used a three-factor formula that consisted of evenly weighted sales, payroll and property factors to approximate the business activity occurring in a particular jurisdiction. This formula known as the Uniform Division of Income for Tax Purposes Act (UDITPA) is part of the Multistate Tax Compact (“Compact”). Many states, including California, Oregon and Washington, have adopted the Compact.
As years have passed, states have determined that destination states (i.e., where the buyer is located) would do better if the sales factor had greater weight. So, states like California double weighted the sales factor, making it a four-factor formula, causing more income to be apportioned into California from out-of-state businesses. Cal. Rev. & Tax Code (“RTC”) §25128. Oregon and Washington have gone one step further by abandoning costs to generate interstate income, and using only a single sales factor formula (in Washington, it is limited to businesses reporting as service, royalty or financial businesses). Such an extreme approach may well have constitutional problems, but that is another topic for another post on another day. It also has some practical problems.
The Gillette Case
Along came The Gillette Company in 2010, questioning RTC §25128’s double-weighted sales factor as the only apportionment method for the tax year 2006, filed a claim for refund. The taxpayer read the Compact and found that if a state deviates from UDITPA for income tax purposes, then the taxpayer has the election to report tax under UDITPA’s three-factor formula.  California, as one could predict, did not like that argument because it would diminish the tax burden on the out-of-state businesses while instate businesses would continue to export their income under the formula … no instate business would deliberately elect to use the UDITPA formula to increase its tax obligations. This would appear to be big trouble for a state that is already struggling with significant budget problems. So, California argued that the Compact is not a real compact even though it statutorily adopted the Compact into law in 1974. In California’s eyes, the Compact is merely a model law, affording no substantive rights to a taxpayer.
Did the law permit an election to a more favorable formula?
California Courts of Appeal, First Appellate District heard and decided the case, The Gillette Company & Subsidiaries v. California Franchise Tax Board (“Gillette”).  It agreed with the taxpayer that the Compact allowed for an election.  California, as a sovereign signatory member of the Compact, could not unilaterally modify  the Compact’s provisions. Although the court has granted a rehearing and vacated its opinion on its own motion on August 9, 2012,  the case continues to have significant importance, because even if this court retreats, this issue remains alive and well in many other states where the courts have not yet addressed the question.
The California legislature was not sure how the case would come down after it was argued, so it decided that California would withdraw from the Multistate Tax Commission. In June, the legislature repealed the Compact (RTC § 38001 et. seq.) and withdrew from the commission, hopefully eliminating the issue for both past and future periods. The Governor signed the bill an June 27, 2012. See SB 1015.
Can the California legislature retroactively eliminate the refund petitions with new filing standards?
If you are one of the out-of-state taxpayers who paid income tax under California’s four-factor formula (double-weighted sales factor), you would probably like to consider a refund. However, there is a slight problem. Apparently, the legislature was concerned about a flood of refund requests, and on June 27, 2012, when SB 1015, § 4 became law, the refund possibility seems to be gone. The law now reads:
- The doctrine of election (see generally Pacific Nat. Co. v. Welch (1938) 304 U.S. 191), provides that an election affecting the computation of tax must be made on an original timely filed return for the taxable period for which the election is to apply and once made is binding.
- The doctrine of election described in subdivision (a) applies to any election that affects the computation of tax under Part 10 (commencing with Section 17001), Part 10.2 (commencing with Section 18401), and Part 11 (commencing with Section 23001) of Division 2 of the Revenue and Taxation Code, unless otherwise provided.
- Subdivision (b) does not constitute a change in, but is declaratory of, existing law.
This means that if the taxpayer did not originally elect to use the UDIPTA equally-weighted three-factor formula when it filed its original return, then it cannot now file an amended return to make the election. The statute asserts that this position is not a change in law, but a declaration of existing law. Of course, this sounds a bit disingenuous because the state maintained throughout the litigation that no election ever existed.
Naturally, the taxpayers did not believe that they had an election, because RTC § 25128 said: “Notwithstanding Section 38006 [the section adopting the UDIPTA formula], all business income shall be apportioned to this state by” using the double weighted sales factor formula. The word “shall” is not permissive, it is mandatory. Thus, the state statutorily led taxpayers to believe that there was no election possible. It is not surprising that taxpayers did not make an election that the statute said did not exist. Let’s step back to think about what the California legislature was effectively saying on the eve of the Gillette decision: “Even if we said there was no election available in 1993, there actually was an election and by this legislation, we are clarifying that the right to that election that we previously said you did not have. And, by the way, under the doctrine of election, you needed to make the election that we said did not exist … at least not until June 27, 2012 … on your original return.”
There appear to be a few problems with SB 1015. First, many observers believe that SB 1015 violates the California Constitution that requires a 2/3 approval in the legislature for tax increases. See California Constitution, Art. XIII A, §3 (also known as Prop. 26), a constitutional provision, that requires a 2/3 approval in each house of the legislature to increase taxes. SB 1015 did not pass out of the legislature with the 2/3 approval requirement. So, there may be litigation surrounding that issue.
Another issue is SB 1015’s reliance on Pacific National Co. v. Welch (“Pacific”).  That case is factually distinguishable from the facts to which SB 1015 applies. In Pacific, the taxpayer had an election to report tax under one of two methods. Clearly, there was an election to be made. After the facts later developed, it was apparent that the taxpayer made the wrong election. The taxpayer would have fared better with the other method. So, the taxpayer then amended prior years’ returns to elect the method that produced a lesser tax liability. The U. S. Supreme Court rejected the taxpayer’s attempt to use 20-20 hindsight. It justified the denial in this way:
The parties agree that, if allowed to change to the installment method, petitioner would be entitled to a refund in some amount. But that fact has no tendency to discredit the deferred payment method as inapplicable. The amount of the tax for the year in question is only one of many considerations that may be taken into account by the taxpayer when deciding which method to employ. The one that will produce a higher tax may be preferable because of probable effect on amount of taxes in later years. In case of overstatement and overpayment, the taxpayer may obtain refund calculated according to the method on which the return was made. Change from one method to the other, as petitioner seeks, would require recomputation and readjustment of tax liability for subsequent years and impose burdensome uncertainties upon the administration of the revenue laws. It would operate to enlarge the statutory period for filing returns (§ 53(a)) to include the period allowed for recovering may be taken into account by the taxpayer when deciding which method to employ. The one that will produce a higher tax may be preferable because of probable effect on amount of taxes in later years. In case of overstatement and overpayment, the taxpayer may obtain refund calculated according to the method on which the return was made. Change from one method to the other, as petitioner seeks, would require recomputation and readjustment of tax liability for subsequent years and impose burdensome uncertainties upon the administration of the revenue laws. It would operate to enlarge the statutory period for filing returns (§ 53(a)) to include the period allowed for recovering overpayments (§ 322(b)). There is nothing to suggest that Congress intended to permit a taxpayer, after expiration of the time within which return is to be made, to have his tax liability computed and settled according to the other method. By reporting income from the sales in question according to the deferred payment method, petitioner made an election that is binding upon it and the commissioner.
The application for refund in the Gillette case did not challenge the applicability of the less favorable of two methods, making Gillette is arguably different. In Gillette, the statute said that there was only one statutory method, a theme consistently argued by the state in the litigation. In Pacific, there were two statutory methods, allowing a taxpayer to choose between them. According to the Court, allowing a taxpayer to revoke one election in favor of a different election would require “recomputation and readjustment of tax liability” in future years. Certainly, that is the case if a taxpayer flip flops between available elections because it made a mistake in the initially chosen election. In Gillette, as mentioned above, the taxpayer never had a choice to make. This is not a situation where the taxpayer chose the wrong election and then wanted to correct it. This is a case where the state unintentionally misled the taxpayer into believing there was no choice.
Keep in mind that the taxpayer could face interest and potential penalties if it elected not to follow the state statute and file a return on an apportioned basis that the state adamantly stated did not exist. Remember, the statute, notwithstanding the Compact, said there was no election.
Also, Schedule R of the corporate income tax return did not provide for an election to UDIPTA, explaining only that the double-weighted single-sales factor formula applied, to wit: “For taxable years beginning on or after January 1, 1993, the sales factor is double weighted for most corporations.” The exceptions were for agriculture, savings and loan activities and banking businesses. The instructions went on to say “If the entire group satisfies one of the exceptions, all members of the group must use a single-weighted sales factor. If not, all members of the group must use a double weighted sales factor.” See page 3 of 2008 instructions.
Further, according to the Gillette opinion, “the FTB contends that the plain language of section 25128 mandates the exclusive use of the double-weighted sales apportionment formula, thereby eliminating use of the equally weighted three-factor apportionment formula … ‘. Gillette, slip opinion at page 8. The state has consistently taken the position that there was no election since 1993 through the day it made it arguments in Gillette. Thus, in the minds of the FTB, there was no election to be made. This was not just the FTB incorrectly interpreting the law, it was also the legislature incorrectly interpreting the law under the Compact as well (i.e. it was bound to the terms of the Compact and could not unilaterally alter it).
So, the question is whether Pacific, stands for the position taken in SB 1015. In a case when the taxpayer has two choices and makes a bad choice in hindsight, Pacific says that the taxpayer’s initial choice is binding. However, should the courts allow the FTB and the state legislature to push principles of the Pacific case beyond its actual facts? In other words, if the state consistently denied an election by statute, by taxpayer instructions and by oral and written arguments to the court, can it now be heard to claim that such an election existed all along and the taxpayer’s failure to disbelieve the statute, reporting instructions and the state written and oral arguments to the court works to the taxpayer’s disadvantage?
This case screams for a remedy in the nature of estoppel, barring the state to take a position inconsistent with its previous representations when there is reliance and irreparable harm (denial of a refund in this case). While it is true that estoppel is rarely applied to tax collection, this is a case that insists on its application. The conditions that the taxpayer must prove are:
- the government agency must be shown to have been aware of the actual facts;
- the government agency must be shown to have made an incorrect or inaccurate representation to the relying party and intended that its incorrect or inaccurate representation would be acted upon by the relying party or have acted in such a way that the relying party had a right to believe that the representation was so intended;
- the relying party must be shown to have been ignorant of the actual facts; and
- the relying party must be shown to have detrimentally relied upon the representations or conduct of the government agency.
This doctrine applies to public agencies when injustice would result from a failure to estop in a “sufficient dimension to justify any effect upon public interest and policy which would result from estoppel.”  Such injustice must be “grave”  or “manifest.”  Even sufficient reliance does not permit estoppel if it will defeat the “effective operation of a policy adopted to protect the public.”  If there is a bilateral mistake of law, reliance on that mutual mistake does not create estoppel.  “Estoppel is available only in the ‘unusual case’ in which its justification is ‘clear and the injustice great.’ The failure to collect the tax authorized by a statute is insufficient to justify estoppel, even if the taxpayer relies on erroneous construction of a statute by an official.” 
For taxpayers seeking the same relief that the taxpayer, Gillette, might receive, they would need to do much to make an estoppel case. Nevertheless, it does seem that a plausible case can be made. Was the government aware of the facts? There was a statute that provided for an election and a statute that that said only one form of apportionment was permitted. Clearly, the state was aware of the facts.
Second, did the government make an incorrect or inaccurate representation and intended the taxpayer to rely on the representation? It’s hard to imagine that state did not intend taxpayers to rely and act in accordance with RTC § 25128.
Third, can it be shown that taxpayers were ignorant of the actual facts? This might be a bit more difficult. At least one taxpayer, Gillette, was not ignorant that the Compact allowed for an election. However, taxpayers reading only RTC § 25128 and the reporting instructions were ignorant, but perhaps not permissively ignorant under the law.
Fourth, the taxpayer must show that it relied on RTC § 25128 to its detriment so as to leave the taxpayer suffering a manifest injustice. Clearly, there is an injustice if the taxpayer cannot use an election that it was misled into believing did not exist and is now barred from making the election.
There are certainly many hurdles to overcome to establish estoppel but one might ask in connection with those hurdles: Is it good tax policy to encourage taxpayers to question the face value of statutory and agency reporting instructions? It would seem not. Ironically, such a policy would actually result in the intended evil averted in Pacific, because taxpayers would no longer take statutes or agency instructions at face value, but instead they would be encouraged to use methods not statutorily authorized. It would result in burdensome uncertainties upon the administration of the revenue laws. That policy could actually encourage abusive tax shelters. The policy also could invite much unnecessary litigation, as taxpayers would not know if the statutes really meant what they said and could only find out after a judicial review.