Nicholas Hallock1Nicholas Hallock is a staff member of The University of Chicago Law Review and a J.D. candidate in the University of Chicago Law School class of 2022. He received his B.A. from Columbia University in 2017. Thanks to the editors of The University of Chicago Law Review for their help with this piece.
During the COVID-19 pandemic, many businesses transitioned to remote work for some or all of their employees, relying on videoconference platforms like Zoom and Microsoft Teams for communication. In some cases, remote work increased productivity, and some companies, like Twitter, have offered their staff the option to work from home permanently. For these reasons, businesses with only remote employees—known as “distributed companies”—will likely become more common. A few have existed for years.
The trend toward distributed companies will have numerous legal consequences. One element of the federal diversity-jurisdiction statute is a particularly poor fit for distributed companies that choose to incorporate. Under 28 U.S.C. § 1332(c)(1), a corporation is a citizen of the state “where it has its principal place of business.” This concept makes little sense for a corporation with no physical presence. How, then, will distributed corporations (and their opponents in litigation) access federal courts through diversity jurisdiction? I will review current law and its rationales and examine some plausible analogies. I will then propose a solution: courts should consider distributed corporations to have no principal place of business for the purposes of § 1332.
I. Current Law and Considerations for the Path Forward
A. Statutory Language and the “Nerve Center”
Under Article III, federal courts can hear “controversies . . . between citizens of different states.” And under 28 U.S.C. § 1332(c)(1), a corporation is a citizen of both the state in which it incorporates and “the State or foreign state where it has its principal place of business.”2Though § 1332(c)(1) applies to foreign and domestic corporations, this Essay will address only distributed corporations based in the United States. At one time, corporations were only citizens of their states of incorporation; but, in 1958, Congress added the “principal place of business” language. Since corporations often sue, or are sued, by their neighbors via state-law causes of action, place-of-business citizenship eliminated diversity jurisdiction in many corporate disputes.
In the following years, some courts used what Judge Richard Posner called a “neurological metaphor”: to determine the principal place of business, a judge would “look for the corporation’s brain, and ordinarily find it where the corporation has its headquarters.” For a while other courts used different tests but, in 2010, in Hertz Corp. v. Friend, the Supreme Court endorsed this “nerve center” reading of § 1332(c)(1). Two aspirations motivated the Court: ensuring “administrative simplicity” and reducing “jurisdictional manipulation.” Because of this, Hertz also emphasized physical location: the “nerve center,” the Court said, “is not the State itself. . . . [It] is a single place” “within a State.” The Court acknowledged that there is “no perfect test,” and presciently recognized that, “in this era of telecommuting, some corporations may divide their command and coordinating functions among officers who work at several different locations, perhaps communicating over the Internet.” But it provided no answer as to what § 1332 would dictate should this happen.
B. Four Considerations
To apply § 1332(c)(1) to distributed corporations, courts should consider four crucial factors.
First, any solution should cohere with the text of § 1332(c)(1), which says the “principal place of business” must be one “place” within a state. That language has been in effect for more than sixty years, and corporations have relied on it when choosing their places of business. The rise of distributed corporations should not unsettle the law for traditional ones.
Second, a solution should effectuate Hertz’s goals. Administrative complexity and jurisdictional manipulation often go hand in hand, and any rule ought to reduce both. Straightforward rules promote predictability, thereby encouraging judicial and corporate efficiency. And, of course, reducing the number of motions challenging jurisdiction will decrease the federal court system’s workload—which was the purpose of the “principal place of business” language in the first place.
Third, a solution should take into account prior constitutional and statutory case law on diversity jurisdiction. For instance, in Mollan v. Torrence, the Supreme Court held that citizenship “depends upon the state of things at the time of the action” and “cannot be ousted by subsequent events.” Some seemingly plausible solutions may run afoul of longstanding diversity-jurisdiction precedent.
Fourth, a solution should not undermine one of the corporate form’s central rationales: limiting individual liability. Limited shareholder liability “is a corollary of the concept of a corporation as an entity.”3William A. Klein & John C. Coffee, Jr., Business Organization and Finance: Legal and Economic Principles 130 (4th ed. 1990). It underlies the basic principle that a corporation is a “person distinct from the shareholders who own it.”4Corporation, Black’s Law Dictionary (11th ed. 2019).
II. Potential Solutions That Fail
To determine the legal place of business of a distributed corporation, courts will sift through case law on unusual corporate forms in a hunt for the best analogy. I will present some existing cases and suggest a few ideas of my own.
A. Unusual Corporate Forms
Holding companies have little need for a physical presence, since they exist only to own other entities. Therefore, issues of diversity jurisdiction for incorporated holding companies may resemble what distributed corporations will face. In Johnson v. SmithKline Beecham Corp., the Third Circuit addressed the citizenship of a holding company, GSK Holdings. The corporation had three directors. Only one of the directors always attended board meetings “in person” in Wilmington, Delaware, while the other two “often participate[d] telephonically from other offices” out of state. Since the meetings were officially in Delaware, the court determined that the corporation had its principal place of business in that state.
But an analogy to Johnson, despite its similarity to remote work, offers no solution. While there was a de jure place where the conference call occurred (Wilmington), the official location of a call or videoconference can be manipulated too easily, defeating one of Hertz’s goals. Jurisdictional manipulation is a serious concern in the context of remote work. Consider, for example, a distributed corporation expecting to be sued by a plaintiff from the state where its board’s remote meetings are officially located. It could guide that anticipated lawsuit into federal court by assigning the responsibility of officially hosting a conference call to a director located in a different state.
Additionally, if different calls are officially located in different places, administration is no longer simple. A distributed corporation could assign a different officer or director in a different state to host each regular management call. Courts would then struggle vainly to determine which state serves as the corporation’s headquarters.
In 3123 SMB LLC v. Horn, the Ninth Circuit addressed another holding company, Lincoln One, that had never held a board meeting. Though Lincoln One had “engaged in no activity other than incorporation,” the court held that, since the firm had determined that its board meetings would be held in Clayton, Missouri, then that is its principal place of business. But Horn may violate Mollan, since the first board meeting—which the court used to determine its jurisdiction—did not occur until a year after the suit was filed. And its rule would again be easy to manipulate, since Lincoln One’s meeting plans could easily be changed on a whim to accommodate the corporation’s litigation strategy. Moreover, it would only apply to distributed corporations that planned to physically locate themselves in the future.
One judge dissented in Horn. Judge Andrew Hurwitz argued that “the ‘nerve center’ cannot be in a state where the corporate EEG is flat.” He would have held that Lincoln One was actually located in California, “where its shareholders and directors resided, and where the only corporate asset . . . was located.” Similarly, the Fourth Circuit determined one corporation’s principal place of business to be “where the majority of corporate officers were located and where those officers were responsible for oversight and strategic decision-making.” The majority-rule approach taken by the Fourth Circuit and Judge Hurwitz makes intuitive sense. But it, too, runs into administrability and manipulation problems when applied to distributed corporations. If officers reside in different states, a court will have to count them to label the principal state. It may be hard to determine who counts as a decisionmaker, and easy for a corporation to manipulate the titles of its own officers. Additionally, by zeroing in on the location of each shareholder, the Horn dissent begins to chip away at the distinct personhoods of the corporation and its owners.
Another possible analogy is to the defunct corporation—one that is winding down and has ceased to do normal business, but can still sue and be sued. In a pre-Hertz case, William Passalacqua Builders, Inc. v. Resnick Developers South, Inc., the Second Circuit held that a defunct corporation’s principal place of business is where it “last transacted business”—in the language of Hertz, where its last nerve center was. But while William Passalacqua’s most-recent-location test is tempting for corporations that recently transitioned to fully remote work, it would not work for corporations that were founded as distributed. Nor would it work for some corporations that switched from physical to distributed many years ago. Section 1332(c)(1), after all, speaks in the present tense. To quote Mollan, if a now-distributed corporation no longer has significant ties to its old place of business, it can hardly be said to be located there “at the time of the action.”
B. Other Possibilities
Unlike a corporation, an unincorporated entity is a citizen of each state where one of its members is a citizen.5Carden v. Arkoma Assocs., 494 U.S. 185, 195–96 (1990). While this rule does not stem from § 1332(c)(1), it might be tempting, as a practical matter, to treat a distributed corporation like an unincorporated entity—giving it multiple places of business, and assigning it the citizenship of each decisionmaker. But there are theoretical and practical reasons not to go this route. On a theoretical level, this idea, like the Fourth Circuit’s majority rule, runs counter to the idea that a corporation (unlike an unincorporated entity) is a distinct person. On a practical level, a corporation with many shareholders could be a citizen of many states—or even every state—without access to federal court even in some disputes with national implications. Alternatively, to be consistent with Hertz, the officers could be the “members.” But then a court would have to decide which officers are senior enough to qualify, creating another administrative difficulty.
Another possible rule is that the principal place of business is where the corporation’s chief executive works. After Hertz, in the words of the First Circuit, “the federal court is to look for the place where the buck stops.” This too is tempting: It has the look of administrative simplicity. But it would not be simple in application. While a corporation generally has a public-facing chief executive, appointed by its directors, that executive’s place of work may be far from obvious. She may live in one state or work in another, or—since the corporation is distributed—work from different locations at different times of the year. Additionally, this rule would risk allowing jurisdictional manipulation: A corporation fearing oncoming state-court litigation in a certain state, for instance, could move its CEO out of that state temporarily until after the suit is brought. And the rule might even incentivize directors, when choosing a CEO, to select someone working out of a low-population state where litigation is less likely to arise.
III. A New Rule for Distributed Corporations
A. No Physical Presence Means No Place of Business
The best solution comes from another defunct corporation case. In Holston Investments, Inc. B.V.I. v. LanLogistics Corp., the Eleventh Circuit joined the Third Circuit (and broke with the Second Circuit) to hold that “a dissolved corporation has no principal place of business,” and for purposes of § 1332(c)(1), is a citizen only of its state of incorporation. At first blush, applying this idea to distributed corporations seems less than ideal. But a no-place-of-business rule accords with all considerations listed in Part I.B.
First, it is consistent with the text of § 1332(c)(1). Federal law places a corporation in “the State . . . where it has its principal place of business.” Nothing in the plain text requires a corporation to have a principal place of business. Logically, § 1332(c)(1) means that if a corporation has a principal place of business in a state, then it is a citizen of that state. It is silent on corporations that lack one. Reasonably, courts often probe the corporate structure to find a principal place of business whenever possible. But the text does not prohibit a court from concluding that a certain corporation is not principally located in a single place—as the Third and Eleventh Circuits have done.
Second, this rule is consistent with the goals of Hertz. Once a court determines that a corporation is distributed, the path forward is simple and there is no room for manipulation. And due to persistent concerns about fully remote work, it is unlikely that many corporations will want to—or be able to—become distributed solely to take advantage of increased access to federal court.
Third, this rule is consistent with existing constitutional and statutory diversity-jurisdiction precedent. Between 1844 (when the Supreme Court decided Louisville, Cincinnati & Charleston R.R. Co. v. Letson) and 1958 (when Congress added the “principal place of business” language), corporations were citizens only of their states of incorporation. There is no constitutional bar to this rule and, as previously discussed, no statutory bar either.
Fourth, this rule does not undermine the rationales for corporate personhood. Unlike some other suggestions, under this rule courts would analyze the operations of the corporation itself, not the individual living and working arrangements of its officers, directors, or shareholders.
B. Addressing Potential Criticisms
There are two major lines of criticism that could be leveled at this proposal. After addressing an objection grounded in congressional intent, I will propose a burden-shifting framework to address administrative concerns with the no-place-of-business rule.
1. Congress intended to restrict access to federal court, but such intent cannot overcome the statutory text.
First, this solution admittedly does not accord perfectly with Congress’ intent in 1958. As a contemporary court wrote, the “primary purpose” of the principal place of business language “was to reduce the heavy workload of the federal courts.” But any conflict with congressional purpose is not dispositive. As the Supreme Court recently said, “the express terms of a statute” supersede any countervailing “extratextual considerations.”
If we do accept the import of legislative purpose, I admit that this proposal would move some litigation from state to federal court. But this would occur less than might appear. Even if distributed corporations were assigned to a state, they would likely have more access to federal court than traditional ones. Because of their lack of physical presence, distributed corporations are more likely to enter litigation with far-flung opponents—and less likely to enter litigation with their neighbors—relative to physically located corporations. Consequently, the marginal effect of adopting a no-place-of-business rule for distributed corporations would be limited. Moreover, the “principal place of business” language would still be important in many cases. There would still be far less litigation in federal court with it than without it.
2. This rule raises administrative concerns, but they can be alleviated by a burden-shifting framework.
Second, a no-place-of-business rule for distributed corporations may raise some of the administrative concerns Hertz sought to avoid: it may be difficult to decide which corporations are distributed and which are not. A corporation may try to manipulate its structure (or its appearance) to frame itself as one or the other, depending on the forum in which it wishes to litigate.
This should not be especially difficult in the usual case. People have strong feelings about remote work (both for and against it). Companies will want to attract employees who prefer—and work productively under—their policies. Corporations will therefore be motivated to publicly announce whether they have or do not have a physical presence. For example, one distributed corporation, Articulate Global, Inc., advertises on its career webpage that employees can “[w]ork where, when, and how [they] want. There’s no corporate office, no corporate B.S.” Articulate even offers employees “help [in] set[ting] up [their] home office[s] with awesome equipment and technology.” Similarly, most corporations will want to publicly communicate their locations to clients. Materials like these will be available to attorneys and judges.
Of course, there will still be hard cases, especially when a plaintiff lacks evidence about the nature of a corporate defendant. Corporate defendants often prefer federal court to state court. Therefore, the most common opportunity for manipulation will be a corporation that has not publicly advertised itself as distributed claiming, during litigation, that it is distributed in order to take advantage of the no-place-of-business rule. In this scenario, the best solution is a burden-shifting framework. A plaintiff can establish a prima facie case for the existence of a principal place of business by citing to corporate webpages, news articles, and publicly filed documents, that mention, or at least plausibly suggest, a physical presence.6This would constitute the “short and plain statement of the grounds for the court’s jurisdiction.” Fed. R. Civ. P. 8(a)(1). Once the plaintiff establishes its prima facie case, the corporate defendant can respond by pointing to publicly available sources the plaintiff missed. Or, it can produce internal documents attesting to its distributed nature, but by doing so likely opens itself up to jurisdictional discovery.
The reverse, a plaintiff claiming that a defendant is distributed in order to get into federal court, would be less common, but would presumably still happen in some cases. The plaintiff would establish its prima facie case by alleging distribution with some evidence, and therefore force an evidentiary response from the defendant. She could do this by pointing to documents, filings, or statements that collectively do not mention a physical presence in places where such presence would normally be mentioned. (For example, in publicly available SEC filings or on the defendant’s website.) Again, the defendant would respond by showing some evidence that it is fully distributed. But jurisdictional discovery should be permitted only if the plaintiff produces affirmative evidence that the defendant is distributed—evidence of absence would not be enough.
This second scenario might seem too easy for a plaintiff to allege jurisdiction. But there are systemic incentives pushing plaintiffs to file accurate, detailed complaints. As Professor William H.J. Hubbard has written, a conscientious plaintiff seeking a favorable settlement will “use the complaint as a credible signal of her willingness to pursue litigation.”7William H.J. Hubbard, A Fresh Look at Plausibility Pleading, 83 U. Chi. L. Rev. 693, 703 (2016). A thorough complaint allows “the plaintiff with a strong claim to separate herself from the plaintiff with a weak claim. By doing this, she brings the defendant to the settlement table.”8Id. Plaintiffs and their lawyers will not want to signal weakness to the judge and defendant by failing to take jurisdictional allegations seriously. Of course, vexatious or pro se litigants often have different incentives or are unaware of how to target settlement, and may not put much effort into alleging the defendant’s lack of physical presence. But then it would be easy for a defendant to quickly respond: she would merely point to the location of a corporate headquarters, and the case would be remanded to state court.
Taken together, my proposal would limit both a corporate defendant’s ability to manipulate jurisdiction and a plaintiff’s ability to raise the defendant’s costs without evidence, while increasing administrative complexity only in the occasional case.
Remote work brings great changes to which the legal profession is not immune. In particular, jurisdiction revolves around sovereignty, which itself is traditionally defined as exclusive control over physical territory. But as physical presence becomes less important to the corporate world, our rules tying corporate citizenship to territory grow obsolete. One casualty is the “principal place of business” clause of 28 U.S.C. § 1332(c)(1), which must be rethought for distributed corporations.
A rule that distributed corporations have no principal place of business is the strongest solution. It squares with the text of § 1332(c)(1), the policy rationales for the current “nerve center” test, existing case law on diversity jurisdiction, and the historical bases for the corporate form. Such a rule will not significantly undermine Congress’ purpose, and with a clarified policy for jurisdictional discovery it can be applied without much added complexity.
Longstanding rules on which businesses rely should not be changed on a whim. A no-place-of-business rule is only a modest shift, but would provide a needed update and increased stability for the distributed corporations of the future.
* * *
Nicholas Hallock is a staff member of The University of Chicago Law Review and a J.D. candidate in the University of Chicago Law School class of 2022. He received his B.A. from Columbia University in 2017. Thanks to the editors of The University of Chicago Law Review for their help with this piece.