Minimal Privett Reduction and the Unsuspendable Floor for Category 1A Incitement in Decoy Cases: Commentary on R v Robson‑Durrance [2025] EWCA Crim 1280

Minimal Privett Reduction and the Unsuspendable Floor for Category 1A Incitement in Decoy Cases: Commentary on R v Robson‑Durrance [2025] EWCA Crim 1280

Introduction

In R v Robson‑Durrance [2025] EWCA Crim 1280, the Court of Appeal (Criminal Division) considered a reference by His Majesty’s Solicitor General under section 36 of the Criminal Justice Act 1988 alleging undue leniency in a sentence imposed for online sexual offences against a police decoy purporting to be a 13‑year‑old boy. The respondent, aged 59 at sentence, pleaded guilty to:

  • Attempted sexual communication with a child (Sexual Offences Act 2003, s 15A).
  • Attempting to incite a child to engage in sexual activity (s 10), specifically penetrative activity (self‑anal penetration while masturbating).
  • Attempting to cause a child to watch a sexual act (s 12), by sending a video of himself masturbating.

The Crown Court imposed concurrent sentences, taking count 2 (incitement to penetrative activity) as the lead count, resulting in a total of two years’ imprisonment suspended for two years, together with a Sexual Harm Prevention Order for 10 years and notification requirements.

The central issues before the Court of Appeal were:

  • Whether the judge unduly reduced the notional sentence for the Category 1A s 10 offence (penetrative activity) to reach a suspendable two‑year term, particularly given the case involved a decoy rather than a real child.
  • Whether the judge failed to reflect totality for multiple concurrent offences.
  • Whether the mitigation and plea credit were over‑generous.
  • Even if unduly lenient, whether the Court should exercise its discretion to leave the sentence undisturbed.

Summary of the Judgment

  • The Court held that the sentence was unduly lenient. For a Category 1A s 10 offence (incitement to penetrative activity), the guideline starting point is five years’ custody with a range of four to 10 years. Even allowing for the “decoy” context (Privett reduction), mitigation, and full one‑third plea credit, the notional post‑trial sentence could not reasonably be reduced as far as three years—a level necessary to justify a two‑year sentence after full credit.
  • The Court accepted that some downward adjustment at Step 2 was required because there was no real child, but re‑emphasised that the reduction is usually very small where the offender did not voluntarily desist early and where serious sexual activity was intended.
  • The Court noted that with multiple offences sentenced concurrently, an upward adjustment from the single‑count starting point should ordinarily be considered to reflect totality.
  • Despite finding undue leniency, the Court exercised its discretion not to increase the sentence. Key considerations included: the suspended sentence had already structured the respondent’s life; his engagement with mental health services post‑sentence; his assessed complex needs and vulnerability; and the protective orders in place.
  • Leave to refer was granted, but the sentence imposed by the Crown Court was not altered.

Analysis

Precedents Cited and Their Influence

  • R v Privett [2020] EWCA Crim 557:
    • Privett formulates a two‑stage approach for cases involving decoy victims or attempts:
      1. Identify the category of harm by the sexual activity intended.
      2. Apply a downward adjustment at Step 2 to reflect the absence of actual harm; where the decoy is the only reason the offence did not occur, the reduction is usually very small; no additional reduction for the label “attempt”.
    • Robson‑Durrance applies this approach to confirm that for Category 1A s 10 (penetrative) offending, the Step 2 reduction cannot rationally be so large as to pull a notional five‑year starting point down into a range that yields a suspendable two‑year sentence after plea.
  • R v Plaku [2021] EWCA Crim 568:
    • Addresses the circumstances for granting full one‑third plea credit, generally where the plea is at the first reasonable opportunity.
    • The Court accepted that the full one‑third discount was justifiable on the unusual facts: defence had signalled an intention to plead guilty upon receipt of viewable message evidence (as recorded in the Better Case Management form).
  • Attorney General’s Reference (No 4 of 1989) [1990] 90 Cr App R 366:
    • Defines “unduly lenient” as outside the range reasonably open to the judge, and stresses the Court of Appeal’s discretion to refrain from increasing sentence even where undue leniency is found.
    • Robson‑Durrance draws directly on this discretion, grounded in fairness and subsequent events, to leave a suspended sentence in place despite undue leniency.
  • Attorney General’s Reference No 132 of 2001 (Bryn Dorian Johnson) [2002] EWCA Crim 1418:
    • Explains that AG references serve to maintain public confidence where a judge has substantially departed from normative sentencing standards.
    • Supports the Court’s willingness to mark the sentence as unduly lenient, which helps restore public confidence, even though the Court ultimately declined to increase the sentence.
  • Attorney General’s Reference (Egan) [2022] EWCA Crim 1751 and Attorney General’s Reference Nos 14 and 15 of 2006 (French and Webster):
    • Egan summarises principles governing s 36 references: undue leniency must fall outside the reasonable range; leave is exceptional; s 36 targets gross error; and the “double jeopardy” or “second jeopardy” concern can inform the discretion not to increase.
    • French and Webster is cited on the role of double jeopardy considerations. The Court in Robson‑Durrance underscores that it is not barred from increasing a suspended sentence to immediate custody but may nevertheless decline to do so where fairness demands.

Legal Reasoning

The Court’s reasoning proceeds from the Sentencing Council guidelines and Privett’s two‑stage framework, calibrated by general principles governing s 36 references:

  1. Guideline anchoring for the lead count (s 10 incitement to penetrative activity).
    • Count 2 is Category 1A (intended penetrative activity; significant age disparity; soliciting images) with a starting point of five years, range four to 10 years.
    • Where a decoy is involved, Step 2 permits a downward adjustment to reflect the absence of actual harm. Crucially, the guideline and Privett state that where the offence is prevented only because the “child” is a decoy and there is no genuine early voluntary desistance, only a very small reduction will usually be appropriate—and there is no extra reduction for the fact it is an “attempt”.
  2. Concurrency and totality.
    • Counts 1 (s 15A sexual communication, Category 1A: SP 18 months; range nine months–two years) and 3 (s 12 causing a child to watch, Category 2A: SP two years; range one–three years) were properly made concurrent, but totality required considering whether the lead count starting point should be adjusted upward to reflect the aggregate criminality. The judge did not do so.
  3. Mitigation.
    • The Court accepted substantial mitigation: age 59; no previous convictions; engagement with mental health services; vulnerability; and a professional’s oral confirmation of community mental health involvement.
    • Nevertheless, the state of the evidential foundation—no written pre‑sentence report—made a large step outside the guideline range difficult to justify.
  4. Plea credit.
    • Although the plea was entered at the PTPH, the Court accepted full one‑third credit in light of the pre‑indicated intention to plead on receipt of viewable messages, consistent with Plaku’s pragmatic approach.
  5. The pivotal error: excessive downward adjustment producing a suspendable sentence.
    • Even crediting mitigation and full plea discount, the Court held that the notional post‑trial sentence could not reasonably be as low as three years. The judge’s journey from a five‑year starting point down to three years (to reach two years after a one‑third discount) overstated the Step 2 reduction for the decoy context and the weight of mitigation.
    • Put differently: for a Category 1A s 10 offence involving incitement to penetrative activity with multiple strands of serious online sexualised conduct, a post‑plea two‑year sentence (enabling suspension) is outside the reasonable range.
  6. Discretion not to increase.
    • Despite the undue leniency, the Court declined to increase the sentence, weighing:
      • Reliance interests and the respondent’s conduct under a suspended sentence already in force.
      • Post‑sentence engagement with mental health services and an assessment of complex needs.
      • Vulnerability and the presence of protective orders (SHPO; notification) to manage risk.
    • This was expressly characterised as a response to the “very unusual circumstances” of the case, not a general indulgence for similar offending.

What This Decision Adds

  • Privett reductions in decoy cases are constrained. Where penetrative sexual activity is intended (Category 1A), the Step 2 reduction for a decoy will usually be small—measured in months, not years—particularly where there is no genuine early voluntary desistance. Courts should resist large Step 2 reductions that push the sentence into a suspendable range.
  • A practical “floor” is signalled. Although not formulised as a rigid rule, the Court’s analysis indicates that for Category 1A s 10 incitement cases of this kind, a notional post‑trial sentence below three years will rarely be reasonable; after a full one‑third discount, this implies post‑plea sentences will ordinarily remain above two years (i.e., not suspendable) absent truly exceptional features.
  • Totality matters even with concurrency. Multiple concurrent sexual offences should ordinarily lead to an upward adjustment from the single‑count starting point to capture aggregate harm and culpability.
  • Discretion to leave unduly lenient sentences undisturbed remains robust. The Court visibly applies Attorney General’s Reference (No 4 of 1989), Egan, and French & Webster to respect fairness and reliance interests in appropriate cases, even for serious sexual offences, where post‑sentence developments and protective orders mitigate the need for immediate custody.

Impact

For Sentencers

  • In decoy/attempt cases, confine the Step 2 reduction to the narrow compass envisaged by the guidelines and Privett. Where penetrative activity is intended, reductions large enough to yield a suspendable term will usually be inappropriate.
  • Apply totality: even with concurrent sentences, reflect aggregate criminality by stepping up from the lead count starting point where justified.
  • Mitigation anchored by credible evidence carries weight, but absent a written pre‑sentence report, large excursions below the guideline range will be harder to support.
  • Do not add an extra “attempt” discount. The guidelines explicitly disallow a second reduction simply because the offence is labelled “attempt”.

For Prosecution

  • On references alleging undue leniency in decoy cases, emphasise Privett’s constraint on Step 2; highlight where desistance stems from suspicion or police intervention rather than early voluntary cessation.
  • Press totality adjustments where multiple offences are sentenced concurrently to avoid under‑assessment of overall seriousness.
  • Expect the Court to scrutinise suspended outcomes in Category 1A cases; even where mitigation is strong, a two‑year suspended term is likely to be viewed as below range.

For Defence

  • Full one‑third plea credit remains achievable at PTPH where there is documented early intention to plead (e.g., in BCM forms) conditional on receipt of essential evidence.
  • Where mental health and vulnerability are central to mitigation, ensure the court has reliable evidence (ideally written) to avoid an inference that a substantial downward adjustment lacks foundation.
  • While the Court can leave an unduly lenient suspended sentence undisturbed in exceptional circumstances, practitioners should not assume this outcome; the case is framed as unusual and fact‑sensitive.

Complex Concepts Simplified

  • Category 1A (harm/culpability): Sentencing Council matrices classify seriousness by intended harm (e.g., penetrative activity is Category 1) and culpability (A indicates the most blameworthy features, such as significant age disparity or soliciting images). Category 1A for s 10 has a five‑year starting point, range four to 10 years.
  • Privett two‑stage approach:
    1. Assess harm by what the offender intended (not what actually occurred).
    2. Reduce the sentence modestly if there was no real child or no actual harm. Where the police decoy is the only reason the offence did not happen, or where the offender only stopped due to suspicion, only a very small reduction is usually appropriate. Importantly, there is no extra discount merely because the offence is an “attempt”.
  • Notional sentence after trial vs. after plea: Courts identify the sentence that would have been imposed after a trial (the “notional” figure), then apply the percentage reduction for a guilty plea (up to one‑third for the earliest pleas).
  • The “magical” two‑year threshold and suspension: In general, only sentences of up to two years can be suspended. A judge who targets suspension must not reverse‑engineer the notional sentence to reach two years; the final figure must be justified by the guidelines and evidence, not by the desire to suspend.
  • Totality: When multiple offences are sentenced at the same hearing, the court ensures the overall sentence reflects the total offending. Even with concurrent terms, the lead count may be adjusted upward to capture the combined criminality.
  • Unduly lenient test (s 36 references): A sentence is unduly lenient if it falls outside the range a reasonable judge could have imposed. Even then, the Court of Appeal has a discretion not to increase the sentence, considering fairness, post‑sentence developments, and broader justice.
  • “Double jeopardy” in this context: Not the constitutional bar on re‑trial, but the concern that an offender faces the anxiety of sentence enhancement after initial sentence. It can influence the exercise of discretion not to increase a sentence, especially where a suspended order has been relied on and post‑sentence conduct is constructive.
  • Sexual Harm Prevention Orders (SHPOs): Civil orders designed to manage risk and protect the public; their presence can be relevant to the discretionary decision whether an immediate custodial increase is necessary.

Worked Illustration (Why Two Years Suspended Was Outside Range)

Although the Court did not prescribe an arithmetic formula, its reasoning can be illustrated:

  • Lead count (s 10, Category 1A) starting point: 5 years.
  • Step 2 (decoy) reduction: usually very small where desistance is not early and voluntary. Illustratively, a modest reduction might be months, not years (e.g., to c. 4 years 6 months–4 years 9 months).
  • Totality for counts 1 and 3 (concurrent but serious additional criminality): an upward nudge to reflect aggregate conduct (e.g., back toward c. 5 years).
  • Mitigation (age, no record, mental health, guilty plea): may bring the notional post‑trial term nearer the bottom of range, but a post‑trial figure as low as 3 years was “difficult to see” as reasonable on these facts.
  • After full one‑third plea credit, a three‑year notional term would become two years (suspendable). Because a three‑year notional was not reasonably attainable, the two‑year suspended outcome was unduly lenient.

Conclusion

Robson‑Durrance makes two significant contributions. First, it tightens the application of Privett in decoy cases involving intended penetrative activity: Step 2 reductions must be genuinely modest and will rarely drive the sentence into suspendable territory for Category 1A s 10 offending, especially where the offender did not voluntarily desist at an early stage. In practical terms, the decision signals a floor: a notional post‑trial sentence below three years is unlikely to be reasonable in comparable circumstances, making a two‑year suspended sentence generally out of range.

Second, the case exemplifies the Court’s principled restraint under s 36. Even when a sentence is unduly lenient, the Court can, and in appropriate cases will, leave it undisturbed where fairness, reliance on a suspended order, post‑sentence engagement with mental health services, and protective orders together make immediate custody neither necessary nor just. This is an exception grounded in the specific facts, not a softening of standards for serious online child sexual offences.

The key takeaways for future cases are clear:

  • Minimal Step 2 reductions in decoy/attempt cases; no additional “attempt” discount.
  • Rigid attention to the Category 1A starting point for penetrative incitement; suspendable outcomes will be the exception, not the rule.
  • Concurrent counts still require totality‑driven uplift of the lead sentence.
  • Plea credit can reach one‑third at PTPH where a genuine early intention to plead is documented.
  • Discretion not to increase unduly lenient sentences remains available but is reserved for truly unusual, fairness‑driven circumstances.

In synthesis, Robson‑Durrance both reaffirms the gravity with which the courts approach online sexual offending against children, even in decoy settings, and demonstrates a careful, humane use of appellate discretion where justice between the parties so requires.

Case Details

Year: 2025
Court: England and Wales Court of Appeal (Criminal Division)

Comments