Cowan v. New Jersey State Parole Board: A Two-Step Justification Requirement for Extended FETs Tied to Recidivism Risk

3.1 Precedents Cited

Standards of review for agency action

• Russo v. Bd. of Trs., PFRS, 206 N.J. 14 (2011) and In re Att'y Gen. Law Enf't Directive Nos. 2020-5 & 2020-6, 246 N.J. 462 (2021): These cases supply the Court’s general administrative-law posture: courts give deference but will reverse final agency action that is arbitrary, capricious, or unreasonable. Their role here is methodological—confirming that the Parole Board is not insulated from searching review when its reasoning is untethered to governing standards.

• Trantino v. State Parole Bd. (Trantino IV), 154 N.J. 19 (1998): This case provides the familiar three-part test for reviewing agency action and serves as the Court’s anchor for examining whether the Board followed legislative policy, relied on substantial evidence, and reasonably applied the governing factors. It is also substantively important because Trantino IV set aside a parole denial and a 10-year FET where the Board’s reasoning did not sufficiently focus on the likelihood of future crime.

• Krug v. State Parole Bd., 261 N.J. 477 (2025): Cited for de novo statutory interpretation principles and for historical context on the Parole Act’s amendments. It also reinforces that the legal framework depends on which version of the Parole Act applies, and it is referenced regarding the 1997 amendments’ elimination of the “new information” limitation.

Parole decisions, discretion, and judicial intervention

• Trantino v. State Parole Bd. (Trantino VI), 166 N.J. 113 (2001) and Beckworth v. State Parole Bd., 62, N.J. 348 (1973): These cases frame parole determinations as individualized discretionary appraisals, while still requiring that the Board’s conclusion be reasonably supported by sufficient credible evidence. They are used to calibrate deference: the Court does not reweigh evidence but insists the Board’s decision must be rational and connected to the statutory standard.

• Acoli v. State Parole Bd., 250 N.J. 431 (2022): The Court draws from Acoli to reject “threadbare findings,” especially generic assertions about “lack of insight” that are not meaningfully connected to recidivism risk and that ignore significant countervailing evidence (such as institutional record, age-crime research, and risk assessments). Cowan extends this logic into the FET-setting context: conclusory narratives cannot justify a dramatic extension beyond the presumptive term.

The evolution of parole law: recidivism vs. punishment

• In re Parole Application of Trantino (Trantino II), 89 N.J. 347 (1982): The Court relies on this decision for a detailed historical account: under the 1948 regime the Board considered both likelihood of recidivism and sufficiency of punishment; after the 1979 changes, parole eligibility became tied to the sentence and parole decision-making was narrowed to recidivism risk. This history is essential to Cowan because it explains why a very long FET cannot be justified as a proxy for “more punishment” or “more rehabilitation” untethered to the recidivism inquiry.

• Royster v. Fauver, 775 F.2d 527 (3d Cir. 1985): Cited to characterize the 1979 reforms as a significant shift in penal law. It reinforces the Court’s narrative that parole decisions after 1979 are conceptually aligned with the new sentencing structure and its allocation of punishment decisions to the judiciary at sentencing.

• Trantino v. State Parole Bd. (Trantino V), 331 N.J. Super. 577 (App. Div. 2000), overruled in other part by Krug, 261 N.J. at 497: Used for the proposition that the 1979 Parole Act precluded the Board from considering whether the prisoner had been sufficiently punished for deterrence and retribution. Cowan uses this to highlight that the Board must not functionally reintroduce “punishment sufficiency” through inflated FETs.

• State Parole Bd. v. Byrne, 93 N.J. 192 (1983): Supports the doctrinal point that the 1979 Act created a presumption in favor of release and shifted the burden to the State to justify denial based on recidivism likelihood. This presumption informs the Court’s view that presumptive FETs similarly carry weight and cannot be displaced without a rigorous explanation.

Extended FETs and the “pick a number” problem

• Berta v. State Parole Board, 473 N.J. Super. 284 (App. Div. 2022): This is the decision the Court effectively elevates into statewide governing law for extended FETs. The Supreme Court adopts its two-step explanation requirement: overcoming the presumption (why the presumptive FET is clearly inappropriate) and justifying the chosen term (why it is necessary and appropriate). The Court endorses the warning that the Board cannot “simply pick a number out of thin air.”

  The Court also echoes Berta’s insistence that extended FETs must not be punitive and must be tied directly to reducing the likelihood of future criminal behavior—then adds its own emphasis: the extended term should be no longer than needed to address recidivism risk.

• Garner v. Jones, 529 U.S. 244 (2000): Appears in a footnote addressing potential ex post facto issues about applying current regulations. While not litigated here, its inclusion signals the Court’s awareness that parole rule changes can implicate punishment-increase concerns if they create a significant risk of longer incarceration.

3.2 Legal Reasoning

(a) The governing “north star”: recidivism risk under the 1979 Parole Act

The Court situates the dispute within the legal architecture created by the 1979 reforms: sentencing courts determine the punishment (including parole bars), and the Parole Board’s post-eligibility role is constrained to one question—whether there is a substantial likelihood of a new crime if released. This is not merely contextual. It is the Court’s primary tool for evaluating whether the Board’s FET-setting is lawful.

Because Cowan’s offenses occurred before the 1997 amendments, the Court reiterates that the 1979 Act controls. That matters in two ways:

• Substantive standard: denial and deferral must be linked to likelihood of future criminal behavior.
• Procedural/informational constraint on reconsideration: at subsequent hearings the Board may rely on only new information since the prior denial.

(b) The regulatory structure creates a presumption for the scheduled FET

The Court reads the FET regulations as establishing a structured baseline:

• N.J.A.C. 10A:71-3.21(a)(1) provides the presumptive 27-month FET for the most serious categories (including manslaughter).
• N.J.A.C. 10A:71-3.21(c) allows a limited discretionary adjustment of up to nine months (yielding a 36-month maximum in Cowan’s category).
• N.J.A.C. 10A:71-3.21(d) permits a larger departure only if the scheduled/adjusted FET is “clearly inappropriate” due to lack of satisfactory progress in reducing the likelihood of future criminal behavior, and only after considering the N.J.A.C. 10A:71-3.11(b) factors.

The Court’s key move is to treat this structure as imposing a meaningful burden of explanation: the schedule is not a suggestion. A dramatic deviation must be justified in a way that is traceable to the regulation’s recidivism-risk rationale.

(c) The two-step justification rule for extended FETs

The Court formally adopts the Berta-style framework as the governing standard:

1. Why is the presumptive term “clearly inappropriate”?
   The Board must explicitly address why 27 months (or at most 36 months via subsection (c)) will not suffice given the inmate’s current lack of satisfactory progress in reducing recidivism risk.
2. Why is the chosen term “necessary and appropriate”?
   The Board must articulate why the specific number selected—here, 200 months—is needed, and not merely why “more time” is desirable.

The Court then adds an important proportionality-like constraint grounded in the regulation’s purpose: the extended FET should be no longer than needed to address the perceived recidivism risk.

(d) Why Cowan’s 200-month FET failed under that standard

The Court’s arbitrariness finding rests on multiple, reinforcing failures:

• No explanation for bypassing 27–36 months: the Board did not explain why the presumptive FET (or the +9 month maximum adjustment) was “clearly inappropriate” under N.J.A.C. 10A:71-3.21(d).
• No rational bridge from record to 200 months: the Board did not connect its observations (insight/programming/infractions) to a reasoned conclusion that nearly 17 additional years were required to reduce the likelihood of future criminal behavior.
• Failure to meaningfully grapple with the recidivism-focused evidence: the decision did not meaningfully address Cowan’s disciplinary pattern over time (most infractions clustered in early decades, with only three since 2011), which could bear on whether risk is decreasing.
• Overlooking the objective risk assessment instrument: the LSI-R’s moderate-risk findings (including quantified re-arrest and reconviction probabilities) were not discussed in the three-member panel’s written FET decision, despite the regulation’s directive to consider N.J.A.C. 10A:71-3.11(b) factors and despite risk being the core statutory question.
• “Threadbare” reliance on “insight” and “programming”: echoing Acoli v. State Parole Bd., the Court treats the Board’s generalized statements about “criminal thinking” and unspecified “further programming” as insufficiently substantive to support a massive deviation from the schedule.

The upshot is not that the Board could never impose a materially extended FET; it is that the Board must show its work—explicitly, quantitatively where possible, and always tied to recidivism risk rather than to offense seriousness already baked into the schedule.

(e) Guidance on remand and future cases

The Court provides operational guidance that will shape future parole practice:

• Use the Berta two-step analysis whenever exceeding the presumptive FET and the nine-month adjustment.
• Ensure the duration is no longer than necessary to address the perceived deficit in reducing likelihood of future criminal behavior.
• Consider the N.J.A.C. 10A:71-3.11(b) factors in the extended-FET decision, with special attention to the objective risk assessment instrument, which should receive substantial weight.
• Apply the 1979 “new information” limitation at subsequent hearings in pre-1997 cases: only new information since the first denial may be used to decide whether to grant parole.

3.3 Impact

(a) A new statewide rule constraining extended FET discretion

The most immediate doctrinal impact is the Supreme Court’s adoption of a concrete, reviewable standard for extended FETs. By embracing the Berta framework, the Court converts what had been an Appellate Division corrective into a statewide rule with clear, enforceable steps.

(b) A “reasoned decision” requirement with measurable consequences

Extended FETs now face heightened vulnerability on judicial review if the Board does not:

• explicitly analyze why the schedule is “clearly inappropriate,” and
• justify the specific number chosen as “necessary and appropriate.”

This will likely increase remands in cases where the Board uses generalized language (“insight,” “criminal thinking,” “needs more programming”) without tying those concepts to a time-bound, evidence-based explanation of risk reduction.

(c) Elevated importance of objective risk instruments

By stating that “substantial weight should be given” to the risk assessment instrument, the Court materially raises the stakes of ignoring (or merely checking a box for) tools like the LSI-R. In practice, the Board will need to:

• discuss the score and its implications,
• explain any disagreement with the tool’s results, and
• connect any extended FET length to the risk factors identified.

(d) Reinforcement of the anti-punitive principle in FET setting

The opinion underscores that offense seriousness is already accounted for in the FET schedule and sentencing. Extended FETs must not become a backdoor mechanism for retributive delay. This framing may shape future litigation challenging extended FETs that look like punishment rather than risk management.

(e) Special consequences for pre-1997 cases: “new information” constraint

The Court’s reminder about the 1979 Parole Act’s “new information” limitation is likely to influence a subset of legacy cases. For those inmates, the Board’s ability to revisit old facts or re-litigate unchanged narratives at later hearings is constrained, potentially narrowing the permissible basis for continued denials and extended deferrals.