Overruled? Court of Appeal panel overrules earlier decision, raising questions about precedent

Can a panel of the California Court of Appeal overrule an opinion decided by an earlier panel?

Well, it has. In a case in which our Supreme Court granted review on Wednesday, Cohen v. Superior Court, 102 Cal.App.5th 706 (2024), the court overruled Riley v. Hilton Hotels Corp., 100 Cal.App.4th 599 (2002). Riley had held that Government Code section 36900 creates a private right of action to enforce municipal ordinances. Now our Supreme Court will decide that issue, which I am not addressing here.

Cohen is one of a handful of Court of Appeal cases that state that they are overruling precedent. The first cases to do so seem to have been 1977 and 1980 cases from the Riverside division. People v. Yeats (1977) 66 Cal.App.3d 874; Saucedo v. Mercury Sav. & Loan, 111 Cal.App.3d 309 (1980). In a 2019 case, the Riverside division cited the 1980 case as authority for overruling precedent. Estate of Sapp, 36 Cal.App.5th 86, 109 n.9 (2019). Then another division cited the 2019 Riverside case as authority for doing so. State Dept. of Hospitals v. Superior Court, 84 Cal.App.5th 1069, 1079 (2022). Cohen cited the 1980 and 2019 Riverside cases for overruling authority. 102 Cal.App.5th at p. 716.

What, though, is the basis of a panel's authority to overrule an earlier opinion? After all, a panel with equivalent legal authority decided the earlier case.

To be clear, this question is not about when intervening authority changes the law, as with a new statute or an opinion from the state or federal Supreme Court. The question is whether a panel can overrule an opinion because, as in Cohen, it concludes the opinion was wrongly decided.

Consider the federal appellate system. There, one three-judge panel in the region that includes California (the Ninth Circuit) cannot overrule, and must follow, an earlier opinion by a three-judge panel, no matter how misguided the current panel thinks the earlier opinion is. Miller v. Gammie, 335 F.3d 889, 899 (2003).

But the federal system is unlike California in two ways: it has regional precedent, and it has an "en banc" procedure. Because precedent applies regionally, a federal court of appeal panel in another region (such as the Second Circuit or Eighth Circuit) cannot overrule Ninth Circuit opinions. Such a panel can create contrary regional precedent, and the conflicting holdings can separately govern trial courts in their own regions -- indefinitely, in theory. The United States Supreme Court could resolve the split by establishing a single rule for the nation.

The only way the Ninth Circuit can overrule its opinions is by convening an en banc panel drawn from all its judges. This is typically eleven of the Circuit's twenty-nine judges, though the "full court" also can convene, see Ninth Circuit Rule 35-3. The en banc panel has greater authority than a three-judge panel.

The California Court of Appeal, in contrast, has neither regional precedent nor an en banc procedure.

Once any division of our Court of Appeal issues an opinion, it is "binding . . . upon all the superior courts of the state."  Auto Equity Sales, Inc. v. Superior Court, 57 Cal.2d 450, 455 (1962). An opinion from the Fifth District in Fresno is binding on a trial court in Long Beach. An opinion from Division One of the Fourth District in San Diego is binding on trial courts in Modoc County, which borders Oregon and Nevada. For precedent, we have a single Court of Appeal, not regional courts.

If Court of Appeal justices -- even an overwhelming majority -- disagree with an opinion issued by their colleagues, there is no mechanism to take the case "en banc" by convening some higher authority within the Court of Appeal. But California appellate justices can do something that Ninth Circuit judges cannot. A three-judge panel can issue an opinion splitting from another panel's binding authority. "[T]here is no horizontal stare decisis within the California Court of Appeal." V Lions Farming, LLC v. County of Kern, 100 Cal.App.5th 412, 436 (2024).

Accordingly, and with some frequency, Court of Appeal panels issue opinions disagreeing with an earlier panel on a legal point. Once that happens, two conflicting but authoritative holdings govern all of California. There is still no regional precedent as a formal matter. When opinions disagree, a state trial court "can and must make a choice between the conflicting decisions." Auto Equity Sales, 57 Cal.2d at 456.

California appellate judges even can create splits in the law within a single district or division. An intra-division split, while uncommon, is more common in the four California districts (the Third through Sixth) with seven to ten justices in each division or undivided district. The First and Second Districts based in Los Angeles and San Francisco have only four justices per division. In those districts, splits between two divisions are common but not intra-division splits.

As justices may split with other divisions (or their own division or district) for "good reason," Lucent Technologies, Inc. v. Board of Equalization, 241 Cal.App.4th 19, 35 (2015), our Supreme Court often grants review of an existing split "to secure uniformity of decision." California Rule of Court 8.500(b)(1). Until the California Supreme Court acts, conflicting decisions -- from anywhere in the state -- provide dichotomous precedent for all the states' trial judges.

Accordingly, if Cohen had disagreed with Riley rather than overruled it, both opinions would remain precedential. Our Supreme Court could resolve the split by granting review, as it did anyway. Cohen, however, overruled Riley, explaining in detail why it found the earlier decision wrong and in need of overruling.

Our Supreme Court has never authorized the Court of Appeal to overrule its cases. It has said that a Court of Appeal opinion "stands . . . as a decision of a court of last resort in this state, until and unless disapproved by this [Supreme C]ourt or until change of the law by legislative action." Cole v. Rush, 45 Cal.2d 345, 351 (1955). That formulation does not allow for overruling. On the other hand, our Supreme Court has not expressly held that the Court of Appeal cannot overrule its cases.

In a journal article four years ago, I suggested that no clear theory had been articulated as to the source of one Court of Appeal panel's authority to overrule another. See The Puzzle of Precedent in the California Court of Appeal, 33 Cal. Litig. 11 (2020). I noted, however, that panels that had overruled opinions had done so only as to opinions from their own division.

There is intuitive appeal to a division's ability to overrule its precedent. In the larger divisions, though, even a unanimous three-judge panel leaves open the possibility that most division justices support the earlier opinion. If three justices in the Sacramento division overrule an opinion, seven off-panel justices might disagree with that ruling.

Cohen was a unanimous opinion from the same four-Justice Los Angeles division that issued Riley, so a current division majority supported the Cohen holding rather than the Riley holding that now-retired justices articulated two decades earlier.

But in my earlier article, I asked: "Because Court of Appeal precedent is not regional, but rather emanates from a single Court of Appeal, binding trial courts throughout the state, what gives judges that happen to sit in the issuing division the unique authority to overrule a decision from that division?" Puzzle of Precedent, 33 Cal. Litig. at 15.

Among other things, with court of appeal precedent binding statewide, panels in other divisions may have agreed with an existing precedent but found no need to publish (and so make precedential) a repetitive opinion. See Nemer v. City of Mill Valley, 2023 WL 8859942 (Dec. 22, 2023) (section 36900 "creates a private right of action for violation of local ordinances"); Kraus v. Grilli, 2015 WL 468634 (Feb. 3, 2015) (affirming judgment based on section 36900 private right of action).

Cohen may have gone further, though, than overruling a case from its own division. It also "disapproved" two published decisions from other districts (in San Francisco and San Diego) that followed Riley without adding to its analysis. See Cohen 102 Cal.App.5th at 727 n.11, disapproving Amaral v. Cintas Corp. No. 2, 163 Cal.App.4th 1157, 1181 n.10 (2008) (city ordinance enforced by "a private right of action for aggrieved employees" under section 36900) and Huntingdon Life Sciences, Inc. v. Stop Huntingdon Animal Cruelty USA, Ltd., 129 Cal.App.4th 1228, 1263-64 (2005) (section 36900 "expressly provides that a violation of a city ordinance may be redressed by civil action").

If a division has the unique authority to overrule its own precedent, then justices in the divisions that issued Amaral and Huntingdon would be the ones to do so. Indeed, some authority states that one division may not overrule a decision of another. Garza v. Asbestos, Ltd., 161 Cal.App.4th 651, 659 n.5 (2008); 16 Cal.Jur.3d Courts § 314. Cohen may be the first opinion that has purported to do so.

 Cohen "disapproving" of Amaral and Huntingdon could just have meant it "disagrees" with the cases. Yet disapproval is the term our Supreme Court uses when it abrogates Court of Appeal opinions after resolving a split in the law. And the difference between disagreement and overruling is significant. The former would mean that Amaral and Huntingdon remained binding authority, and a trial court may choose to follow them instead of Cohen. The latter would mean that trial courts must follow Cohen.

With our Supreme Court having granted review in Cohen, the narrow question about whether it could overrule Riley, Amaral, and Huntingdon no longer matters. Our Supreme Court is now poised to resolve the issue in these cases, whether a particular private right of action exists. But a broader question still lurks: does one Court of Appeal panel have the authority to overrule an opinion decided by another one? And if so, what circumstances warrant overruling rather than disagreeing with an earlier case?