In June, I made my oral argument debut, before the Cincinnati-based U.S. Court of Appeals for the Sixth Circuit in a case called F.P. Development v. Canton, Michigan. As described when Cato filed its amicus brief, the court considered whether a municipal ordinance fining landowners for removing trees from their property effects an unconstitutional taking.

The Fifth Amendment’s Takings Clause prohibits the government from taking private property for a public use unless it provides the owner “just compensation.” In the century since the Supreme Court first explicitly held that a regulation (not just an exercise of eminent domain) could be a taking if it went “too far,” as Justice Oliver Wendell Holmes cryptically put it, judges have struggled to discern regulatory takings from legitimate exercises of the state’s inherent power to ensure public safety, including through the regulation of harmful land uses.

As my colleague Sam Spiegelman and his co-author Gregory Sisk explained in the latest Cato Supreme Court Review, courts before the New Deal had no difficulties distinguishing laws preventing public harms (police powers) from those conferring public benefits (violations of due process or takings requiring compensation). Political pressures in the 1930s compelled the Supreme Court to adopt what Justice William O. Douglas later called a “well-nigh conclusive” deference to legislative determinations of its own inherent powers.

In Cedar Point Nursery v. Hassid, decided last term a few weeks after my argument, the Supreme Court inched closer to the pre-New Deal consensus. It ruled what I had largely argued in F.P. Development: that interferences with what look like fundamental elements of ownership—those harkening to the ancient common law—warrant a more probing review than Justice Douglas’s deferential approach. If a law or government action interferes with such an element, that’s a “per se” taking. The magnitude of the disruption has no bearing beyond the amount of compensation owed.

Enter the Sixth Circuit’s opinion, which came out a few week ago. The side we supported won, but the court didn’t feel the need to consider our argument. The Sixth Circuit held that Canton was in the wrong because its fining F.P. Development was an “unconstitutional condition” on the company’s use of its property. Quoting a high-court ruling from 2013, the court reasoned that “the government may not deny a benefit to a person because he exercises a constitutional right.” The township’s imposition was an “exaction,” a third category of taking separate from per se and partial regulatory takings. Since the Supreme Court developed the exactions doctrine in the late ’80s/​early ’90s, subsequent cases have done little to move the needle in a more property-friendly direction. Ditto the infamous Penn Central test, in which courts balance several factors in considering whether a regulation at best works a partial taking of property.

Governments almost invariably win those cases. Per se takings tests offer the most promise for property rights, but for a long time were limited to regulations alleged to result in permanent physical invasions or the total loss of a property’s value.

In the decades-long fight to expand the category, Cedar Point is a welcome entry. In F.P. Development, it should’ve required the court to explain at least why the removal conditions, no matter how onerous, didn’t interfere with a fundamental strand in its bundle of property rights: the right to the use of trees and timbers affixed to one’s land have a long history in the common-law pantheon. The court instead offered, in a footnote no less, that “nothing in [Cedar Point] demands … we review [this case] under a per se or regulatory approach.” If other circuits discount Cedar Point on similar grounds, what I’ve been considering a watershed, and the most important case from last term, could instead prove to have been just another battle in a long war for the soul of the Takings Clause.

You can listen to my oral argument in F.P. Development v. Canton here (I start at about 36:50).