How one claim for thousands is actually tried.

One plaintiff can speak for thousands — which creates a trial problem.#

A PAGA case is not a class action. It is a representative actionRepresentative actionA suit one employee brings on the State's behalf to recover penalties for an entire workforce — which, unlike a class action, needs no class certification.Lab. Code § 2699(a) — a single employee suing on the State’s behalf to recover civil penalties for an entire workforce. Because it stands in for the State, it does not have to clear the certification hurdles a class action would.

The California Supreme Court fixed that character in Arias²2Arias v. Superior Court, 46 Cal.4th 969 (2009). A PAGA action is a representative action brought on the State's behalf and need not satisfy class-action certification requirements.: a PAGA action is brought on the State’s behalf and needs no class certification. That is the source of its reach — and of the difficulty that follows.

Here is the tension. A single named plaintiff’s penalty claim can run to every employee the violation touched, across a year of pay periods. There is no certification stage to test, narrow, or define that group up front. So a fair question arises with real force: manageabilityManageabilityWhether a sprawling representative claim can actually be tried; after Estrada, a court may trim a PAGA claim's evidence or scope to make it triable, but may not dismiss it as unmanageable.Lab. Code § 2699(p) — how is a claim that sprawls across a whole workforce supposed to be tried at all?

Estrada: courts may manage a claim, but may not dismiss it as unmanageable.#

For a time, some trial courts answered that question by striking PAGA claims they judged too unwieldy to try. The Supreme Court closed that door in Estrada³3Estrada v. Royalty Carpet Mills, Inc., 15 Cal.5th 582 (2024). Trial courts lack inherent authority to strike a PAGA claim as unmanageable, but retain ordinary tools to manage complex cases — limiting the evidence or the scope of claims at trial, and using representative or statistical proof. Disapproved Wesson v. Staples on the dismissal point..

The holding has two halves, and both matter. Trial courts lack inherent authority to strike a PAGA claim as unmanageable — they cannot dismiss the case for being too big. But they retain the ordinary tools for managing any complex case: limiting the evidence or the scope of claims at trial, and allowing representative or statistical proof rather than instance-by-instance testimony. In reaching that result the Court disapproved the Court of Appeal’s decision in Wesson v. Staples⁴4Wesson v. Staples was the Court of Appeal decision that had endorsed striking a PAGA claim as unmanageable; Estrada disapproved it on that dismissal point. The reform then codified the surviving managing power at § 2699(p). (No reporter citation for Wesson is given here — it is named as the disapproved decision, as the holding does, not cited as authority.) on the dismissal point.

The reform codified the managing power: § 2699(p).#

The Legislature did not leave Estrada’s managing power to the common law. The 2024 reform wrote it into the statute, in a single sentence:

Read plainly, this is the codified answer to the manageability question. A court may limit the evidence presented at trial, or otherwise limit a claim’s scope, so that the claim can be effectively tried. It is a power to shape the trial of a representative claim — not a power to refuse to try it. That distinction is the spine of everything that follows.

Manageability is § 2699(p) — not § 2699(q).#

One precise correction prevents a common mistake. The managing power lives in § 2699(p) — the sentence quoted above. § 2699(q) is a separate grant. It addresses the consolidation and coordination of related claims — the machinery for handling overlapping or parallel PAGA cases together — and it is not about manageability at all.

The confusion has a cause worth naming. The 2024 reform rewrote and renumbered § 2699, so a subdivision letter that pointed one place under prior law may point elsewhere now — and a brief working from memory or an older form can land the managing power on the wrong letter.⁵5Re-derive the subdivision against the current statute rather than trusting a pre-2024 brief, treatise, or habit. The scope-limiting power is § 2699(p); § 2699(q) is the separate consolidation-and-coordination grant. Estrada’s own discussion keeps the two distinct. So when the question is whether a sprawling claim can be trimmed to be tried, the subdivision is § 2699(p), full stop.

How a representative claim is actually tried.#

Put Estrada and § 2699(p) together and the mechanics come into view. A workforce-wide claim is tried not by marching every employee through the witness box, but by managing the proof: limiting the evidence presented at trial, limiting the scope of the claim, and proving the rest by representative or statistical evidence — sampling, model workweeks, and a trial plan built to make the common questions answerable at scale.

That trial plan is the practical center of the case. It is where the parties and the court decide which questions are common enough to try in one stroke, how a representative sample stands in for the whole, and where the claim must be narrowed to stay triable. The court’s broader § 2699(e)(2) discretion over the size of the penalty sits alongside this⁶6Manageability under § 2699(p) governs how a claim is tried; the court’s § 2699(e)(2) discretion governs the amount of the penalty — it may award below the maximum, or, notwithstanding the caps, above it, where the alternative would be unjust, arbitrary and oppressive, or confiscatory. Two different levers on two different questions. — a separate lever on the number, not on the trial.

The open seam: does § 2699(p) authorize pre-trial scope-cutting?#

The codified sentence carries a seam the courts have not yet closed, and it is worth flagging as argument rather than law. Read the two clauses closely — the figure below lays them side by side:

The first clause lets a court limit the evidence to be presented “at trial.” The second clause — “or otherwise limit the scope of any claim” — omits those words. Two readings follow, and the statute itself does not choose between them.⁷7Several defense commentators read the omission of “at trial” from the second clause as reopening a pre-trial scope-limiting power — a way to narrow the claim before trial, not merely to streamline the evidence once trial begins. The plaintiff bar does not engage that reading. My read: manageability-as-dismissal is dead; manageability-as-scope is alive and codified; whether (p) authorizes pre-trial scope-cutting is unresolved, with no controlling decision as of July 13, 2026.

On one reading, both clauses operate at trial, and “or otherwise” simply adds a second trial-management tool. On the other, the dropped words mean the scope clause is not confined to trial — it can be used to cut a claim’s scope before trial. No published California decision has construed the point. Until one does, the pre-trial reading is a defense argument to be made, not a settled power to be relied on.

What we still don’t know.#

A treatise earns trust by marking its own edges, and manageability has several live ones. No published appellate decision has construed § 2699(p) since the reform. The provision is barely older than the cases now testing it, and the first authoritative gloss may come from the trial courts or the LWDA’s rulemaking before it comes from a published opinion.

Three questions in particular are untested. Whether (p)’s second clause reaches pre-trial scope-cutting is unresolved — the seam above. How much trimming is “too much” — the point at which limiting a claim’s scope stops being case management and starts being a back-door dismissal Estrada forbids — has no line drawn yet. And the represented group a court must try is itself shaped by the 2024 personally suffered eachPersonally suffered eachThe 2024 standing rule: a plaintiff may sue only over violations they personally experienced, and no longer over violation types that happened only to other workers.Lab. Code § 2699(c)(1) standing rule, whose interaction with the pre-reform representative-standing concept behind Adolph remains untested⁸8Adolph v. Uber Technologies, Inc., 14 Cal.5th 1104 (2023). Under California law a plaintiff compelled to arbitrate the individual PAGA claim retains standing to pursue the non-individual (representative) claims in court, rejecting the contrary suggestion in Viking River. Rested on the standing concept from Kim v. Reins. How “personally suffered each” (§ 2699(c)(1)) reshapes the group that must be tried — and whether it narrows what Adolph keeps alive — is an open valuation question, not a settled one..

These are the kind of seams that get litigated for years. Better to flag them than to paper over them — and better, for now, to read § 2699(p) for what its text plainly grants (the power to manage and trim a triable claim) than for the contested power its silence might be read to imply. The neighboring exposure questions — the caps, the cure machinery, the tiers — are worked out on the penalty and exposure page; here the open question is narrower and sharper: when, and how far, a court may trim.

What to do now.#

Diagnosis is only half of it. Where you stand depends on who you are.

Common questions.#

The authorities.#