In Knick v. Township of Scott, Pennsylvania, 139 S. Ct. 2162 (2019), the Supreme Court overruled Williamson County Regional Planning Commission v. Hamilton Bank, 473 U.S. 172 (1985), to hold that plaintiffs may file inverse condemnation claims in federal court without first exhausting their claims in state court. Advocates of takings expansion rejoiced: “[A]fter more than 30 years, Supreme Court reopens the federal courthouse door to property owners.” Robert H. Thomas, Knick Analysis, Part I: After More Than 30 Years, Supreme Court Reopens the Federal Courthouse Door to Property Owners, InverseCondemnation.com (June 24, 2019), https://bit.ly/2U1k34X.
Skeptics of takings expansion were less sanguine: Knick had “opened the federal courts to what may be a flood of cases, . . . [making] federal courts into the new zoning appeals board for every municipality in the country.” Joseph William Singer, Supreme Court Opens Federal Courts to a Floodgate of Takings Cases, Harvard.edu (June 21, 2019), https://bit.ly/2RvmoTL.
This author thinks both camps are wrong. Most inverse condemnation claims will still be filed or decided in state courts, because many such cases will turn on state law issues appropriate for abstention, and state courts will often be more receptive to inverse condemnation claims. Although the author is usually in the takings-skeptic camp, she agrees with the takings expansionists that Knick properly aligns takings jurisdiction with that in other constitutional cases. This article first lays out the Williamson and Knick decisions and then discusses why the effect of Knick may not be as great as both critics and celebrants assume.
From Williamson County to Knick
Williamson County was built on an afterthought and developed into a jurisdictional trap. Williamson County, Tennessee, had approved a preliminary plat for a housing development in 1973 but several years later refused to renew the approval because of changed density limits and several other alleged zoning violations. The developer filed alleging inverse condemnation, and a jury ruled in its favor. The district court ordered the county to apply the initial zoning regulations but held that damages for the intervening temporary taking were not available under the Fifth Amendment. The Sixth Circuit reversed, and the Supreme Court granted certiorari.
Although the Court granted certiorari on the temporary takings question, it reversed by holding that the claim itself was “premature.” First, in a holding not challenged in Knick, the Court found the denial of approval was not final because the developer had not sought a variance from the zoning requirements. Second, in the holding overruled by Knick, the Court held that takings claims against states are not “ripe” in federal court before state courts have a chance to rule on the takings claim.
This rule was different from that of most actions to vindicate constitutional rights. Congress created 28 U.S.C. § 1343 and 42 U.S.C. § 1983 in the wake of the Civil War specifically to ensure that challenges to violations of federal civil rights could be heard in federal court. The rationale for Williamson County was an almost metaphysical dissection of when, in fact, a takings claim arises. Unlike most constitutional rights, government takings of property are not themselves unconstitutional. The Constitution kicks in only when the government takes property “without just compensation.” Therefore, the Court reasoned, the state had not violated the takings clause until its agents both took property and refused to pay for it through its own post-deprivation procedures—in other words, through a completed state court claim for compensation.
Calling this a “ripeness” rule suggested that the federal takings claim would become ripe after state court litigation. Rules of issue preclusion, however, meant that this was not to be. In San Remo Hotel, L.P. v. City and County of San Francisco, 545 U.S. 323 (2005), the Court held that once a takings claim had been litigated in state court, it could not be relitigated in federal court. The decision was unanimous, but four justices concurred in the judgment, arguing that Williamson County should be repudiated. And in the lower courts, something odder began to emerge. Defendants would remove takings claims to federal court (as was their right under 28 U.S.C. § 1441), and then argue—sometimes successfully—that the court had no jurisdiction over the claim because it had not gone through state court. Opponents of Williamson County were justified in calling this Kafkaesque.
And then came Knick. Rose Mary Knick owns and lives on 90 acres of land in rural Pennsylvania. Her land includes a small cemetery where the ancestors of her neighbors are allegedly buried. In 2012, Pennsylvania enacted a statute requiring such cemeteries to be open to the public during daylight hours, and in 2013 state officials told Knick that her land was subject to the requirement. She responded with a takings claim in federal court. The trial and appellate courts dismissed under Williamson County, and the Supreme Court granted certiorari and reversed. The Court held that a Fifth Amendment takings claim arises as soon as the government takes property without compensation. Disgruntled owners need not first sue for compensation in state court and be rejected before their claims were ripe.
It is disturbing, as Justice Kagan wrote for herself and Justices Ginsburg, Breyer, and Sotomayor in dissent, that the Court violated stare decisis. Respect for stare decisis is part of what makes our courts instruments of the rule of law rather than political preference. But this may be the rare case where the unforeseen consequences and illogic and weak precedential support for the prior rule justify the violation.
First, Williamson County never made much sense when compared to jurisdiction over other constitutional claims. When an individual claims that a state official is discriminating against her in violation of the Equal Protection Clause, for example, she need not go to state court and persuade it to affirm what the official has done before suing in federal court. Second, this discrepancy from regular constitutional jurisdiction was even more glaring after San Remo made clear that takings plaintiffs would not get their day in federal court at all. Third, the decision does not violate any precedent before Williamson County. Although several cases before Williamson County rejected takings claims based in part on state or federal procedures for compensation, these were not jurisdictional cases but rather challenges to the constitutional power to take the property at all. See, e.g., Ruckelshaus v. Monsanto Co., 467 U.S. 986 (1984) (upholding the Federal Insecticide, Fungicide, and Rodenticide Act in part because Tucker Act procedures were available to seek compensation for trade secrets made available by the act). In other words, there was not much to support Williamson County before the decision, but there was a lot of reason to oppose it afterwards.
The Rumors of Knick’s Effect May Be Greatly Exaggerated
Rightly or wrongly decided, for three reasons Knick may not have as great an effect as its supporters hope and its critics fear. First, the Supreme Court has already resolved when damages are available for temporary takings, and Knick does not expand it. Second, many takings plaintiffs should still choose to file in state court, as they have a better chance of success there. Third, and most important, federalism will often require federal courts to refrain from deciding takings claims until unsettled state law issues have been decided in state court.
Knick Does Not Expand Temporary Takings Claims
Some might fear that Knick expands not only jurisdiction, but also takings claims and damages, by allowing owners to challenge planning delays and the like. The Supreme Court has already foreclosed this possibility. Just two years after Williamson County, the Supreme Court reached the temporary takings issue in First English Evangelical Lutheran Church of Glendale v. County of Los Angeles, 482 U.S. 304 (1987). The Court held that when there had been a taking, the government could not avoid paying damages by simply withdrawing the offending restriction. Instead, it was liable for damages due to the restriction that arose during the time it was in place. Fifteen years later, however, the Court’s holding in Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency, 535 U.S. 302 (2002), made clear that delays in approving development while reviewing planning decisions did not constitute per se takings for which compensation was required. Instead, the loss of value during the delay would be measured against the value of the property over its lifetime and therefore would likely not amount to a regulatory taking. Accordingly, beyond the jurisdictional implications, holding that a taking occurs as soon as property is taken without compensation has more formal than practical importance.
Given a Choice, Many Takings Plaintiffs Should Strategically File in State Court
Although Williamson County no longer bars takings plaintiffs from filing in federal court first, it may be strategically advantageous to file in state court first. Federal takings law makes it very hard to win an inverse condemnation claim, and state law—and state court interpretations of state constitutional parallels—often offer surer paths to victory. Even before Kelo v. City of New London, 545 U.S. 469 (2005), for example, states had begun to interpret the “public use” requirement more strictly than the Constitution requires, and Kelo increased this exponentially. Several states, moreover, have enacted laws requiring compensation for some regulatory restrictions that reduce the value of land, even when the impact is not so great as to constitute a taking. Other state courts have been more ready to find that government-authorized nuisances constitute takings than have the federal courts. Compare Thornburg v. Port of Portland, 376 P.2d 100 (Or. 1962) (holding noise from airplane flights worked a taking even though the planes did not fly over the plaintiff’s land) with Batten v. United States, 306 F.2d 580, 584 (10th Cir. 1962) (holding noise did not constitute a taking unless planes flew over plaintiff’s land).
The limited empirical evidence also suggests that federal courts are less likely to rule for plaintiffs claiming takings by state and federal governments. James Krier and Stewart Sterk reviewed more than 2,000 inverse condemnation cases decided between 1979 and 2012 in An Empirical Study of Implicit Takings, 58 Wm. & Mary L. Rev. 35 (2016). During most of this period, Williamson County barred federal courts from hearing most takings claims against state governments, so the sample is limited. Still, federal courts ruled on the merits of 142 claims of inverse condemnation by state and local governments in this period. In only 13 cases—or 9 percent of the total—did they find a taking. Id. at 78, Table 8. State courts, however, found a taking in 12 percent of such cases—still a low rate, but a third higher than that in federal court. In contrast, Krier and Sterk found that federal courts ruling on takings claims against the federal government ruled for plaintiffs 26% of the time, more than double the state court rate and almost triple the federal-state rate. This likely reflects both the absence of respect for local decision-making in such cases, and the fact that most such cases are decided within the Federal Circuit, an Article I court whose jurisprudence is notoriously friendly to takings plaintiffs.
State courts, of course, differ from state to state, and plaintiffs’ attorneys may appropriately deem some federal courts friendlier to their claims than those of a particular state. Even in such cases, as discussed below, abstention doctrines will often require them to litigate part of their claims in state court first.
Many Takings Cases Must Still Be Brought First in State Courts
Although, after Knick, federal courts are no longer automatically foreclosed from hearing inverse condemnation cases, federal courts should nonetheless abstain from hearing many such cases until there has been state court review. This is so under two related doctrines, Pullman abstention and Burford abstention, each named after the Supreme Court cases that announced them. Under the Pullman doctrine, federal courts should abstain from reviewing constitutional claims involving sensitive issues of state and local policy where those claims turn on unclear questions of state law. See Railroad Comm’n v. Pullman Co., 312 U.S. 496 (1941). Burford applies where federal review would interfere with state and local control on issues of substantial state and local concern, particularly those involving administrative agencies. Burford v. Sun Oil Co., 319 U.S. 315 (1943); Quackenbush v. Allstate Ins. Co., 517 U.S. 706, 728 (1996). Inverse condemnation cases will often fall into one or both categories.
First, state and local land regulation is surely a “sensitive area of social policy upon which the federal courts ought not to enter” unnecessarily. Pullman, 312 U.S. at 498. The Supreme Court has recognized that the power of eminent domain
is intimately involved with sovereign prerogative, [and the] considerations that prevailed in conventional equity suits for avoiding the hazards of serious disruption by federal courts of state government or needless friction between state and federal authorities are similarly appropriate in a state eminent domain proceeding brought in, or removed to, a federal court.
Louisiana Power & Light Co. v. City of Thibodaux, 360 U.S. 25, 28 (1959) (affirming abstention). See also Kaiser Steel Corp. v. W.S. Ranch Co., 391 U.S. 593 (1968) (ordering abstention in case challenging taking).
Circuit Courts regularly applied this reasoning to inverse condemnation claims before Williamson County made abstention unnecessary in most cases. See, e.g., San Remo, 145 F.3d at 1099 (applying Pullman abstention in regulatory takings claim); Sederquist v. City of Tiburon, 590 F.2d 278, 281–82 (9th Cir.1978) (applying Pullman abstention in regulatory takings claim); MLC Automotive, LLC v. Town of Southern Pines, 532 F.3d 269 (4th Cir. 2008) (applying Burford abstention in due process challenge to land use regulation). Land use regulation is also plainly part of a complex administrative scheme with well-established state fora for review. Knick does not curtail the federalism problems if federal courts became—as critics fear—the “zoning appeals board for every municipality in the country.”
Even in land use cases, of course, abstention is not warranted unless the state law questions are not fairly subject to alternative interpretations. Hawaii Hous. Auth. v. Midkiff, 467 U.S. 229, 236 (1984). But many takings cases turn on unsettled state law questions. Knick itself is one example. The Pennsylvania statutes mandate that landowners give the public access to cemeteries on their land would usually be a per se taking of the owner’s right to exclude—unless public access to cemeteries was already part of Pennsylvania common law. Some jurisdictions have held as much, but the question has not been definitively decided by the state courts. Although Pullman abstention was not raised below, there is a good argument that it should apply. Similarly, Stop the Beach Renourishment, Inc. v. Florida Dep’t of Envtl. Prot., 560 U.S. 702 (2010), depended on Florida’s common law of avulsion and accretion, and the economic wipeout in Lucas v. South Carolina Coastal Council, 505 U.S. 1003 (1992), would not be a taking if the challenged statute prevented a use banned under state common law.
Other cases depend on the construction of state statutes. Murr v. Wisconsin, 137 S. Ct. 1933 (2017), for example, arose from an ordinance requiring at least one acre to build next to a protected river. A grandfather provision, however, permitted a single residence to be built on pre-existing lots of less than an acre. The plaintiffs owned two adjacent lots, one with a residence on it, and they wanted to sell the second as a developable lot. The question that went to the Supreme Court was whether, in evaluating the economic loss as a result of the ordinance, courts should consider the impact on the two lots together, or on the undeveloped lot alone. (Murr held they should be considered together.) But this constitutional question depended on whether the county correctly interpreted the grandfather provision to apply only to lots in separate ownership, so that two adjacent lots qualified for only one structure. Only after the Wisconsin Court of Appeals decided this question against the plaintiffs, State v. St. Croix County, 668 N.W.2d 743 (Wis. Ct. App. 2011), was it clear whether the plaintiffs had any loss, and therefore any takings claim, at all.
In contrast to the Williamson County/San Remo rule, abstention need not bar plaintiffs from having their takings claims heard in federal court. Under England v. Louisiana State Board of Medical Examiners, 375 U.S. 411 (1964), federal courts may hold federal constitutional claims in abeyance until plaintiffs resolve their state law claims in state court. Quackenbush v. Allstate Ins. Co., 517 U.S. 706 (1996) held the same for federal courts exercising Burford abstention in cases for compensation. If plaintiffs choose to litigate their constitutional claims in state court at the same time, they will be precluded from relitigating them in federal court, but the choice belongs to them.
Don’t knock Knick. For the reasons discussed above, the decision makes sense as a matter of civil rights law, and likely will not prevent most takings cases from being first heard in the place where they belong: state courts.