Can a state apportion the gain when a parent sells a subsidiary it runs separately?

A statutory test and a constitutional limit

Every subsidiary sale is tested twice. The first test is the state's own definition of apportionable income. In California, the functional test reaches gain on property integral to the taxpayer's regular operations: in Hoechst Celanese Corp. v. Franchise Tax Board (2001) 25 Cal.4th 508, the California Supreme Court held "that a separate functional test exists," under which income is business income if "the taxpayer's acquisition, control and use of the property" "contribute materially to the taxpayer's production of business income," and "the extraordinary nature or infrequency of the income-producing transaction is irrelevant." A sale of subsidiary stock does not become nonbusiness income because it happens once.

The second test is constitutional, and the same court said so: income inside the statutory definition "must still pass constitutional muster," and only income from assets that "serve an operational rather than an investment function" may be apportioned (Hoechst Celanese, citing Allied-Signal). Illinois folds the two tests into one by defining business income as "all income that may be treated as apportionable business income under the Constitution of the United States," with nonbusiness income meaning "all income other than business income or compensation" (35 ILCS 5/1501(a)(1), (a)(13)). For a subsidiary sale, the constitutional question usually decides the case in either kind of state.

The constitutional rule comes from two divestiture cases, which the library's article on the unitary business principle introduces. In Allied-Signal, where the parties stipulated that the seller and the company whose stock it sold were unrelated, the Court held that New Jersey could not apportion the gain. It stated that "[t]he hallmarks of an acquisition that is part of the taxpayer's unitary business continue to be functional integration, centralization of management, and economies of scale" (504 U.S. at 789), and rejected the state's argument that a planned acquisition program changed the answer: "the mere fact that an intangible asset was acquired pursuant to a long-term corporate strategy" of acquisitions and dispositions does "not convert an otherwise passive investment into an integral operational one" (504 U.S. at 788). It also allowed apportionment without a unitary relationship between payor and payee where a capital transaction "serve[s] an operational rather than an investment function" (504 U.S. at 787), such as working capital held on deposit.

MeadWestvaco closed the gap that language had opened. The Illinois Appellate Court had upheld apportionment of a parent's gain on selling a division because the division served an "operational function," without deciding whether the two were unitary. The Supreme Court vacated, holding that its references to operational function "were not intended to modify the unitary business principle by adding a new ground for apportionment" (553 U.S. 16, Part II-C). The concept recognizes that "an asset can be a part of a taxpayer's unitary business even if" no unitary relationship exists between payor and payee; but "[w]here, as here, the asset in question is another business," the hallmarks of a unitary relationship govern (Part II-C). Usefulness to the parent is not enough; the two must have been one business.

The burden runs against the taxpayer that claims separation: Mobil Oil Corp. v. Commissioner of Taxes, 445 U.S. 425, 439 to 440 (1980), looks for "proof of discrete business enterprise," and Container Corp. of America v. Franchise Tax Board, 463 U.S. 159 (1983), asks only whether the state court's conclusion was "within the realm of permissible judgment." The record made at the first level of review usually decides the case.

What running a subsidiary separately looks like on the record

Two decisions show the facts on which a nonunitary finding has been built. The record in MeadWestvaco described a parent headquartered in one state while "a separate management team ran" the division "out of its headquarters" in another; the two "maintained separate manufacturing, sales, and distribution facilities, as well as separate accounting, legal, human resources, credit and collections, purchasing, and marketing departments"; "[n]either business was required to purchase goods or services from the other"; and the parent's involvement "was generally limited to approving" the division's annual business plan (Part I-B). The same record showed that the parent "managed" the division's "free cash, which was swept nightly" from its bank accounts. The Illinois trial court found the hallmarks lacking and concluded the two were not unitary. The Supreme Court did not review that finding; it remanded and expressed "no opinion" on unity. The facts describe what a trial court accepted as separation, not a Supreme Court holding that those facts suffice.

California's Office of Tax Appeals reached a nonunitary result on a thinner separation in Appeals of Western Distributing Company, et al., 2026-OTA-467P, an opinion OTA designates pending precedential. An S corporation with fourteen divisions sold the assets of one of its two largest divisions, a distribution business that operated in a single state other than California. The divisions shared one insurance policy, centralized human resources, payroll and a retirement plan, and a single checking account on which the controlling shareholder signed most checks. Division managers made their own purchasing, advertising and sales decisions, and the selling division's payments to the company's trucking division were under one percent of that division's revenue, at arm's length rates. OTA applied the three unities test of Butler Bros. v. McColgan (1941) 17 Cal.2d 664 (ownership, operation and "unity of use in its centralized executive force and general system of operation") and the contribution or dependency test of Edison California Stores, Inc. v. McColgan (1947) 30 Cal.2d 472. It started from the rule that the Franchise Tax Board's unitary determination "is presumptively correct" and "the taxpayer bears the burden of proving otherwise." It found unity of ownership, left the FTB's finding of unity of operation undisturbed, and found no unity of use, because its two top officers "exercised the type of oversight that an investor might exercise" rather than control over major policy. It held that the division "did not operate as a unitary trade or business" with the others and noted that because it "operated a separate trade or business, none of its business income is taxable by California."

The result turned on management, not on the back office. Shared insurance, payroll and banking did not make the division unitary where the people running it made the operating decisions and the top officers acted as investors. The opposite record appears in Appeal of NextEra Energy Capital Holdings, Inc. and Affiliates, 2026-OTA-311P (precedential), where OTA found a regulated utility with no California presence unitary with an affiliated energy business on findings of controlling ownership, strong central management with centralized departments, transfers of key executives, shared services and intercompany transactions. Where a subsidiary is unitary, the sale of its stock is business income: in Times Mirror Co. v. Franchise Tax Board (1980) 102 Cal.App.3d 872, the Court of Appeal held that the proceeds of selling a unitary subsidiary's stock "constitute business income, as a matter of law."

Intent to integrate, and the moment the decision is made

A plan to integrate a subsidiary is not integration. The Franchise Tax Board's Legal Ruling 2012-01, Business/Nonbusiness Characterization on Sale of Stock (2012), works through three minority stock purchases. A corporation that bought 20 percent of a company in its own line of business "merely as part of its plan" to acquire control and integrate it, with no "preexisting or ongoing business relationship or operational involvement," had nonbusiness income when the plan failed and it sold the stock. A corporation that bought stock to learn the issuer's technology, and did, had integrated the stock into its unitary business and had business income. A corporation that was a significant distributor of the issuer's products had business income because the operating relationship continued after its takeover attempt failed. The ruling quotes the State Board of Equalization's decision in Appeal of Occidental Petroleum (83-SBE-119) that "mere potential is insufficient to support a finding that the gains on these sales were business income under the functional test," and states the rule: "it is the actual operational ties and the significance of such ties that are most important." Allied-Signal reasoned the same way, observing that a 20.6 percent holder "did not have the potential to operate" the issuer "as an integrated division of a single unitary business" (504 U.S. at 788).

Timing matters as much as intent. The FTB's Multistate Audit Technique Manual, summarizing Occidental Petroleum, states that "the moment of judgment will generally be when the decision to sell is made," and that if the stock "is an integral part of the taxpayer's unitary business at that moment," the gain or loss "will be business income" (MATM 4030). A subsidiary separated shortly before a sale, or drawn into the parent's operations in its last year, is tested on its condition when the sale decision is made. No California authority located states how long a separation must last before it is credited, and the history of the relationship remains evidence of what the relationship was at the decision date.

Where a nonapportionable gain goes

California and Illinois allocate to commercial domicile. Under Rev. and Tax. Code section 25125, "nonbusiness capital gains and losses from sales of stock and other intangible property are allocated to the state in which the taxpayer is commercially domiciled" (MATM 4030). Illinois provides that "[c]apital gains and losses from sales or exchanges of intangible personal property are allocable to this State if the taxpayer had its commercial domicile in this State at the time of such sale or exchange" (35 ILCS 5/303(b)(3)). A parent commercially domiciled in either state that proves its subsidiary was separate is taxed there on the entire gain, while a parent domiciled elsewhere removes the gain from that state's base.

An invented example shows the reversal. A manufacturer of laboratory instruments, commercially domiciled in California, owns a subsidiary that operates veterinary clinics in four other states under its own officers, lenders and systems; the parent's role is two board seats and an annual budget approval. If the subsidiary is not unitary with the parent, the parent's gain on selling it is allocated to California in full. If it is unitary, the gain is business income apportioned by the combined group's factors, and the share assigned to California may be a fraction of the gain. Here the taxpayer, not the state, has the incentive to argue unity, and the evidence that would defeat a unitary finding in another state's audit becomes the auditor's evidence in California. A manufacturer domiciled outside California with the same subsidiary has the opposite incentive in California, and a position taken in one state is read in the others.

A separately run division sold as assets is treated differently from a subsidiary's stock. Western Distributing involved a division's assets, the taxpayer did not dispute that the gain was business income, and the result was that the division was a separate trade or business. California apportions each separate business by its own factors: where a corporation has two or more nonunitary businesses, "the business income from each trade or business must be separately apportioned" (FTB, 2025 Instructions for Form 100 Schedule R, Section K). The gain then follows the selling business's own factors, which in that case placed none of it in California.

New York reaches a similar result through a different structure. Its tax base is "the portion of the taxpayer's business income apportioned within the state" (N.Y. Tax Law section 210(1)(a)), and "business income" means "entire net income minus investment income and other exempt income" (section 208(8)). Investment income includes capital gains from investment capital (section 208(6)(a)), subject to the statute's interest attribution adjustments and its cap at entire net income. Investment capital is, among other requirements, stock held for investment for more than one year, whose disposition is treated as producing long-term capital gain or loss under the Internal Revenue Code, and that is "clearly identified in the taxpayer's records" as held for investment before the close of the day it is acquired, with a different deadline for stock acquired before October 1, 2015 (section 208(5)(a)). "Stock in a corporation that is conducting a unitary business with the taxpayer" is excluded, and a corporation in which the taxpayer holds "less than twenty percent of the voting power" is "presumed to be conducting a business that is not unitary with the business of the taxpayer" (section 208(5)(a)). A qualifying gain on stock of a separately run subsidiary falls outside the business income base. A gain on stock that fails the identification or holding period rules does not, even if the subsidiary was separate, so the identification is made at acquisition.

The open question: taxing by the subsidiary's own factors

MeadWestvaco left one route undecided. Illinois argued in the alternative that the division's own contacts with the state justified apportioning the parent's gain, whether or not the two were unitary. The Court declined to reach the argument because it "was neither raised nor passed upon in the state courts," noted that "[t]he States of Ohio and New York, for example, have both adopted the rationale," citing their statutes then in force, and concluded that "the question is best left for another day" (Part III).

Massachusetts has addressed the point for pass-through interests. In VAS Holdings & Investments LLC v. Commissioner of Revenue, 489 Mass. 669 (2022), a nondomiciliary S corporation sold a 50 percent interest in a Massachusetts limited liability company taxed as a partnership, the parties agreed the two were not unitary, and the Supreme Judicial Court, as the Department of Revenue's TIR 22-14 (November 30, 2022) describes the decision, found "no U.S. constitutional prohibition" on taxing the apportioned gain but held that the Massachusetts statutes "do not authorize taxation in these circumstances." The TIR addresses pass-through entity interests only. Illinois writes the investee approach into its statute for the same kind of interest, allocating nonbusiness gain on S corporation shares or a partnership interest by the average of the entity's Illinois apportionment factor for the year of sale and the two preceding years (35 ILCS 5/303(b)(4)); a corporate parent's gain on C corporation stock stays under the commercial domicile rule.

No authority located settles whether a state where a nonunitary corporate subsidiary operated may tax the parent's gain on its stock through the subsidiary's own factors. The Supreme Court has neither approved nor rejected that route.

When the gain is apportioned: the sales factor

Sales of intangible property are assigned "to the extent the property is used in this state" (Cal. Code Regs., tit. 18, section 25136-2(d)), and for "the complete transfer of all property rights in stock of a corporation," the location of use "is the location of the use of the underlying assets of the corporation" (section 25136-2(b)(12)). For taxable years beginning on or after January 1, 2026, the receipts from stock other than marketable securities are assigned by the subsidiary's own California factors: by the average of its California payroll and property factors if 50 percent or more of its assets, at original cost and excluding cash, receivables and prepaid items, are real and tangible personal property, and by its California sales factor if more than 50 percent are intangible property, using the most recent 12 month year or, for a sale more than six months into the year, the current year (section 25136-2(d)(1)(A)1.a, b). Without the information for either rule, the receipts follow the subsidiary's commercial domicile (section 25136-2(d)(1)(A)1.e).

Many subsidiary sales never reach that assignment. The occasional sale rule provides that "[w]here substantial amounts of gross receipts arise from an occasional sale of a fixed asset or other property held or used in the regular course of the taxpayer's trade or business," the receipts "shall be excluded from the sales factor," and gives as an example that "gross receipts from the sale of a factory, patent, or affiliate's stock will be excluded if substantial," with sales to the same purchaser in a single year aggregated (Cal. Code Regs., tit. 18, section 25137(c)(1)(A), as quoted in Appeals of Amarr Company, 2022-OTA-041P, precedential). The exclusion removes the receipts from the factor, not the gain from the base; in Amarr the gain stayed in apportionable income. Where the parties elect to treat a stock sale as a deemed asset sale, the gain is the subsidiary's own and the analysis changes, as the library's article on Section 338(h)(10) elections in California explains.

Practice notes

The classification of a subsidiary gain is decided by the relationship, and the relationship is shown by records created long before the sale: board composition and who sets the subsidiary's strategy and capital budget, the subsidiary's own officers and their reporting lines, intercompany agreements and the volume and pricing of trade between the companies, guarantees and financing, cash management, shared services and how they are charged, and the parent's public description of its segments. Shared back office functions alone did not establish unity in Western Distributing; central management and transferred executives did in NextEra. The sound sequence before a sale is to fix the decision date, test the relationship as of that date under the hallmarks and the state's own formulation, and then run the result through each filing state's allocation rule, because the same nonunitary finding that removes the gain from one state's base places all of it in the domicile state's base. For New York, the investment capital identification is made when the stock is acquired, not when it is sold. Where the gain is apportioned in California, the occasional sale rule is tested before the look-through sourcing rule is applied. G&G State Tax Group provides unitary and business income analysis of subsidiary dispositions, multistate allocation and apportionment review, and audit representation alongside its state and local tax practice.