A Narrow Reading with Wide Consequences: The California Supreme Court Limits Leave Cashout Credit Under PEPRA

Every public safety professional who has banked unused leave across a long career understands the quiet promise embedded in that ledger. Those hours represent time not taken with family, shifts covered for colleagues, and holidays spent in uniform. The expectation has always been that those hours would count when the pension calculation finally arrived. On July 27, 2026, the California Supreme Court unsettled that expectation for legacy members of the state’s county retirement systems. In Ventura County Employees’ Retirement Ass'n v. Criminal Justice Attorneys Ass'n of Ventura County, No. S283978 (Cal. July 27, 2026) (“Ventura”), the Court held that the California Public Employees’ Pension Reform Act of 2013 (“PEPRA”) bars a retiring employee from counting leave cashouts above the annual limit fixed by the terms of employment, even when the employee designates a final compensation period that spans two calendar years. The decision affirms the Court of Appeal (see 98 Cal.App.5th 1119 (2024)), and it deserves the close attention of every association that negotiates leave and retirement terms on behalf of California’s peace officers, public sector attorneys, and other impacted public sector employees.

The Issue the Court Decided

The dispute turned on a single sentence in Government Code section 31461(b)(2). That provision excludes from compensation earnable any payment for unused leave “in an amount that exceeds that which may be earned and payable in each 12-month period during the final average salary period, regardless of when reported or paid.” (Cal. Gov’t Code § 31461(b)(2).) The retirement system read the phrase to incorporate the annual cashout cap set by an employee’s terms of employment, so that a member who may cash out 200 hours in a calendar year cannot inflate that figure by straddling two calendar years within a single final compensation window. The employee associations read the same phrase to permit inclusion of every hour cashed out during the elected final compensation period, subject only to what the period itself allowed. The Court adopted the retirement system’s reading and confirmed that section 31461(b)(2) caps includable cashouts at the annual allowance no matter how the final compensation period is drawn.

The facts illustrate the stakes with unusual clarity. The retired member accrued 368.16 hours of leave each year, and his terms of employment permitted him to cash out 200 hours in any single calendar year. He designated October 10, 2019 through October 10, 2020 as his final compensation period, and within that window he cashed out 40 hours in December 2019 and another 200 hours in February 2020, for a total of 240 hours. He asked the system to include all 240 hours in his final compensation. The system included only 200, and the Supreme Court agreed that the 40 additional hours fell outside compensation earnable because they exceeded the annual allowance that governed a single calendar year per the terms of his employment.

The Court Conceded Ambiguity and Then Resolved It Against the Member

The most consequential feature of the opinion is not the result but the route the Court traveled to reach it. The Court did not hold that the statute plainly compelled the retirement system’s position. It acknowledged instead that the system’s construction “may not be the most immediately obvious or intuitive reading of the statutory text,” and it described that construction as merely “plausible.” (Ventura, slip op. at p. 22.) Both the trial court and the Court of Appeal had likewise found the language ambiguous. Confronting genuine ambiguity, the Court turned to legislative purpose and concluded that “[c]onsiderations of statutory purpose conclusively resolve the issue” in the retirement system’s favor because the competing reading would reopen the door to the pension spiking that the Legislature enacted the statute to close. (Id. at p. 23.)

In a concurring opinion, Chief Justice Guerrero underscored the divergence from the actual text. She wrote to explain that the majority’s interpretation is not apparent on the face of the statute, that the ordinary use of the word “during” is inconsistent with the majority’s construction, and that the reading survives only because a latent ambiguity emerges once the statute’s anti-spiking purpose is considered. (Ventura, Guerrero, C.J., concurring op. at pp. 2-4.) Her concurrence is candid confirmation that the words the Legislature chose pointed one way while the result points another.

The Court’s analytical move carries real significance for public employees because it required the Court to set aside a rule that has long favored pensioners. For decades, California courts have construed pension statutes liberally and resolved ambiguities in favor of the applicant. (See Barrett v. Stanislaus Cnty. Emps. Ret. Ass'n (1987) 189 Cal.App.3d 1593, 1603.) The employee associations pressed that principle here, and it pointed squarely toward the broader reading. The Court answered that liberal construction “must be consistent with the clear language and purpose of the statute,” and it held that the pensioner favoring canon must yield here because it conflicts with the Legislature’s overarching anti-spiking design. (See Ventura Cnty. Deputy Sheriffs' Ass'n v. Bd. of Ret. (1997) 16 Cal.4th 483, 490.) In other words, the tie no longer runs to the employee once a court identifies a countervailing legislative purpose, and pension spiking supplies a purpose that a reviewing court can invoke across a wide range of compensation disputes.

Why the Reasoning Matters Beyond Leave Cashouts

The holding governs a discrete question about leave cashouts in a final compensation period that straddles two calendar years, yet its method reaches much further. PEPRA is a sprawling statute enacted quickly, and many of its provisions have given rise to disputes and are likely to continue doing so in the future. Every one of those provisions is now subject to interpretation under the framework this Ventura decision reaffirms, and that framework instructs courts to consult the anti-spiking purpose whenever the words admit more than one reasonable meaning. A retirement association that wishes to exclude a category of pay will attempt to frame inclusion as a potential avenue for spiking, and after this decision, that framing will carry substantial weight. Members should therefore expect systems to press expansive exclusions in future disputes over on-call pay, specialty premiums, uniform and equipment allowances, and other items whose treatment PEPRA did not spell out with precision.

The decision also weakens a tool that member advocates have relied upon for a generation. When a pension statute was ambiguous, the liberal construction canon gave employees a thumb on the scale, and it often proved decisive in close cases. This opinion subordinates that canon to legislative purpose and thereby shifts the balance of interpretive advantage toward the systems in exactly the situations where the statute is least clear. Associations litigating future ambiguities will need to build their arguments on text and structure rather than on the presumption that ambiguity favors the applicant, because the Court has now signaled that the presumption gives way whenever the anti-spiking rationale is available. 

The Signal to the Legislature

The opinion carries a further lesson for those who shape pension policy in Sacramento. The Court repeatedly grounded its reading in the practical concerns of plan administration and funding, observing that the retirement system’s interpretation allows counties to anticipate their obligations rather than absorb liabilities that swing between one hundred and two hundred percent of the annual cashout limit depending on how a member elects a final compensation period. (Ventura, slip op. at p. 24.) Legislation should be assessed against this backdrop in which courts will fill statutory gaps with cost containment and anti-spiking assumptions rather than with the older presumption favoring the pensioner. If the Legislature intends a benefit to be included, it will need to say so with precision, because ambiguity will no longer be construed generously in the member’s favor. Associations that seek legislative clarification of favorable treatment for particular pay items should draft proposed amendments with that reality in mind and should not assume that silence or generality will be read to their advantage.

What This Means for Your Members

The immediate practical consequence is straightforward. Legacy members who plan to cash out substantial accrued leave near retirement can no longer increase their final compensation by designating a final compensation period that reaches into a second calendar year in order to stack two annual allowances. The annual cap set by the terms of employment now controls the entire 12-month final compensation period regardless of when it falls in relation to calendar years. Associations should review their memoranda of understanding and their members’ retirement planning materials so that no member approaches retirement expecting credit for cashouts that this decision now excludes.

The larger consequence is strategic. This case reflects a determined and well-resourced effort to secure the broader reading for public employees. Nevertheless, that effort did not prevail before the state’s highest court despite thoughtful advocacy and a serious textual argument that even the Chief Justice found more faithful to the statute’s words. The outcome confirms that the interpretive terrain has shifted and that future gains for members will come from careful bargaining and precise legislative drafting rather than from favorable canons of construction. We stand ready to help our client associations reassess their leave and retirement provisions, to model the effect of this ruling on individual members nearing retirement, and to press for the statutory clarity that will protect the benefits our public safety professionals have earned through decades of service.

When the Verdict Second Guesses the Officer: Appellate Reversal Affirms Officers’ Right to Defend Themselves

Every officer who has ever cleared a doorway understands the arithmetic of a fraction of a second. The decision to use force is made in the moment, on the information then available, against a threat that does not pause to let anyone deliberate. California law honors that reality, yet officers know that courtrooms and juries are frequently invited to second-guess split-second decisions against a leisurely reconstruction that no officer on the scene ever had. A recent Texas reversal of a verdict against an officer illustrates the point. The court’s reasoning is a timely reminder of the constitutional and statutory protections that govern peace officers in California. Although the decision is not binding here, its reasoning is persuasive in holding that a conviction built on hindsight analysis cannot survive when the governing standard is properly applied.

A Dangerous Encounter

In an appeal decided by a Texas appellate court, a police officer had responded with three colleagues to a report that a man was roaming the halls of a downtown building with a knife held to his own throat and threatening suicide. The officers were told the man had a history of mental health crises and was on an upper floor with a large kitchen knife. When the officers reached that floor by elevator, the man was facing a mirror with the knife at his throat. When the doors opened, he turned toward the officers, reoriented the knife away from himself and toward them, and advanced in close quarters while ignoring the commands shouted at him. The officers fired, and the man died. The officer was convicted of a firearm offense and sentenced to prison, notwithstanding testimony from fellow officers that they would not have acted differently, expert testimony that the use of force was justified, and the conclusion of the department’s own investigators that the conduct did not warrant criminal charges. The appellate court reversed and rendered a judgment of acquittal, holding that the prosecution never carried its burden of disproving justification once the objective evidence was examined at the only moment that mattered. Taylor v. State, 729 S.W.3d 124 (Tex. App. 2025).

Reviewing the body worn camera footage, it found that the officers were confined with no meaningful avenue of retreat, that the man turned toward them and advanced with the knife oriented in their direction, and that this presented an immediate and potentially lethal threat. The prosecution’s theory, by contrast, rested almost entirely on what the officers supposedly should have done differently, including waiting, using the stairs, or relying on less lethal options.

The court held that this evidence, even if accepted in full, did not answer the dispositive question, because it showed at most that different decisions could have been made. As the court explained, could have, would have, and should have do not disprove justification, and tactical misjudgments or hindsight critiques do not render an otherwise reasonable belief in the necessity of deadly force unreasonable. The court further rejected the notion that a subject who advances with a weapon is demonstrating compliance, holding that the law does not require officers to treat such conduct as surrender or to wait until an attack is completed before responding. Finally, it observed that a mental health crisis is dynamic, so that a person who is a danger only to himself in one moment may become an immediate threat to others in the next, and that reasonableness is measured on what the officer perceived at the moment force was used rather than on earlier assurances or missed opportunities to de-escalate.

The California Framework

California officers enjoy the same core protection, secured by both the federal Constitution and state statute. As the Supreme Court held in Graham v. Connor, the reasonableness of a use of force must be judged from the perspective of a reasonable officer on the scene rather than with the twenty-twenty vision of hindsight, and the analysis must make allowance for the fact that officers are forced to make split second judgments in circumstances that are tense, uncertain, and rapidly evolving. That inquiry turns on whether the person confronting the officer posed an immediate threat to the safety of the officer or others.

The Legislature wrote these principles into Penal Code Section 835a, which states that the decision to use force be evaluated from the perspective of a reasonable officer in the same situation, based on the totality of the circumstances known to or perceived by the officer at the time, rather than with the benefit of hindsight, and it expressly accounts for the reality that officers must sometimes make quick judgments about using force. The statute permits deadly force when the officer reasonably believes, based on the totality of the circumstances, that such force is necessary to defend against an imminent threat of death or serious bodily injury. A threat is imminent when a reasonable officer would believe the person has the present ability, opportunity, and apparent intent to immediately cause death or serious bodily injury, a harm that must be instantly confronted rather than a mere fear of future harm. The same section confirms that an officer need not retreat and is not rendered an aggressor, nor deprived of the right of self-defense, by using objectively reasonable force.

California courts have applied this framework with a clear eye toward the dangers of hindsight analysis. Applying AB 392, Koussaya v. City of Stockton, (2020) 54 Cal. App. 5th 909, confirmed that the reasonableness of force is still judged from the perspective of a reasonable officer on the scene rather than hindsight, and the standard provides deference to the split second decisions of an officer who, unlike a private citizen, is charged with acting affirmatively and using force. Critically, so long as an officer’s conduct falls within the range of conduct that is reasonable under the circumstances, there is no requirement that the officer choose the most reasonable available action or the conduct least likely to cause harm.

California decisions confronting the knife wielding advancer have reached the same conclusion the Texas court reached, holding that where deadly force is otherwise justified there is no constitutional duty to use less forceful or non-deadly alternatives first, and that the proper inquiry is whether the officer acted reasonably, not whether some less intrusive alternative was theoretically available. Martinez v. Cnty. of Los Angeles, 47 Cal. App. 4th 334, 54 Cal. Rptr. 2d 772 (1996). The focus remains on the vantage of the reasonable officer, not on the after the fact opinions of bystanders who did not perceive a threat.

Why a No Reasonable Alternative Standard Cannot Be Squared With the Constitution

This settled framework is the measure against which the earlier legislative proposals to rewrite California’s deadly force law must be judged. As those proposals were originally introduced, they would have gone considerably further than any jury in the case discussed above, restricting an officer’s use of deadly force to circumstances in which it was strictly necessary and in which no reasonable alternative existed. Such a standard would unreasonably invite the factfinder to convict an officer whenever some other course could later be described as available. Representing the Peace Officers Research Association of California (PORAC), our firm opposed that approach because it would have transformed the objective reasonable officer inquiry into an impossible mandate to exhaust every conceivable option before acting. In holding such a standard unconstitutional, our Ninth Circuit has recognized, “(r)equiring officers to find and choose the least intrusive alternative would require them to exercise superhuman judgment. Scott v. Henrich (9th Cir. 1994) 39 F.3d 912, 915.” Our firm was privileged to assist the law enforcement advocacythat ultimately helped secure the amendments to Assembly Bill 392 that removedthe most punitive features of the introduced language and preserved thereasonable officer standard in the law as enacted.  

The reversal of the Texas conviction reenforces the point that such a standard, had it been enacted in California as introduced, would be constitutionally infirm. A criminal conviction may rest only on evidence from which a rational factfinder could find every element beyond a reasonable doubt, and in a justification case that necessarily includes proof beyond a reasonable doubt that the officer did not reasonably believe force was necessary in the moment.

A no reasonable alternative standard collapses that inquiry. It permits conviction not because the officer’s real time perception was unreasonable, but because a prosecutor, working backward from a known outcome, can always describe another path not taken. The Texas court rejected this reasoning when it held that could have and should have do not disprove justification. It is also precisely what California law already forecloses, because there is no constitutional duty to resort to the least forceful alternative where deadly force is justified. An officer is not required to select the least harmful option to avoid liability. A statute or instruction that made the availability of a hypothetical alternative dispositive would nullify those protections and convert the reasonable officer standard into a strict liability regime, judged in the calm of a courtroom rather than from the position of the officer on the scene. Koussaya, supra. 

When jurors are told to ask whether any reasonable alternative existed, and to measure that question against a reconstruction assembled after the danger has passed, they are no longer deciding whether the officer’s belief was reasonable at the instant of the threat. They are grading the officer against an idealized response that the moment never permitted. That is the very danger the reasonable officer standard exists to prevent, because it refuses to let a sanitized reconstruction assembled after the fact displace the dangerous and rapidly evolving reality the officer actually confronted. An instruction that operationalizes hindsight in this way threatens the due process guarantee that a conviction rest on proof of an actual element of the offense, and it departs from the constitutional command of Graham that force be assessed from the perspective of a reasonable officer on the scene rather than hindsight.  

It is worth emphasizing that the deadly force statute the Legislature ultimately enacted, as shaped by the amendments described above, stopped far short of the standard the original proposals contemplated. Section 835a expresses the intent that officers use deadly force only when necessary in defense of human life and that they use other available resources and techniques, but it qualifies that expectation by requiring only what is reasonably safe and feasible to an objectively reasonable officer, and it preserves throughout the reasonable officer perspective and the rejection of hindsight. SB 230 also codifies the legal definition of "feasible” is defined as "reasonably capable of being done or carried out under the circumstances to successfully achieve the arrest or lawful objective without increasing risk to the officer or another person." (emphasis added.)  This definition clarifies that tactics increasing the risk to officers are not "feasible." 

In other words, the enacted law retains the constitutional anchor that the introduced version would have disgaurded. That distinction is the difference between a statute that channels officer discretion and one that would punish officers for failing to achieve a perfection the law has never demanded.

Practical Significance of Taylor for California Officers

For officers and the agencies that stand behind them, several lessons follow. Justification is assessed on the threat as it reasonably appeared at the instant of the decision, and neither a prosecutor nor a jury may substitute a menu of tactical alternatives for proof that the officer’s belief was unreasonable. A subject who advances with a weapon and disregards commands is not demonstrating compliance merely because no blow has yet landed, and officers are not required to absorb the first strike before responding. Prior information that a person is troubled or suicidal does not freeze the threat assessment, because the encounter is judged as it evolves. Where objective evidence such as body worn camera footage confirms the threat, that evidence can be decisive. And under existing California law, an officer who acts within the range of reasonable conduct has no separate duty to have chosen the least forceful option that a later critic might prefer.