Vulnerability in Fraud Sentencing Counts for Both Culpability and Harm; Prior Similar Convictions Aggravate; Unrelated Drug and Fraud Offences Ordinarily Require Consecutive Terms

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

Neutral citation: [2026] EWCA Crim 490

Date: 1 April 2026

Procedure: Attorney General’s Reference under s.36 Criminal Justice Act 1988 (unduly lenient sentence)

Judges: Lord Justice Stuart-Smith, Sir Robin Spencer, and the Common Serjeant of London (HHJ Marks KC)

Core holding (in practical terms)

  • In fraud guidelines, victim vulnerability may properly influence both culpability (targeting) and harm (impact) without “double counting”.
  • Multiple high-culpability features create upward pressure from the guideline starting point; the starting point is not reserved for cases where “all” listed factors are present.
  • Relevant previous convictions—especially for strikingly similar offending—are a separate aggravating factor and should not be discounted as “double counting” merely because the offender is not “naïve”.
  • For totality, where drug offending is unrelated to fraud, consecutive terms will often be required (or an uplift if concurrency is used).
  • A Goodyear indication does not prevent intervention on an unduly lenient sentence reference.

1) Introduction

This Attorney General’s Reference concerned sentences imposed on two offenders for a “courier” bank card fraud conspiracy targeting elderly and otherwise vulnerable victims, and (for the first offender) substantial possession with intent to supply Class A drugs discovered during the fraud arrest warrant execution. The sentencing judge (a Recorder) imposed:

  • for the first offender: 30 months’ immediate imprisonment for fraud, concurrent with 30 months for each Class A drug offence; and
  • for the second offender: 2 years’ imprisonment suspended for 2 years, with a 25-day rehabilitation requirement.

The Solicitor-General contended the sentences were unduly lenient due to misapplication of the Sentencing Council guidelines, inadequate recognition of the offenders’ extensive histories of similar fraud, and an error in totality (treating unrelated drugs and fraud as fully concurrent).

2) Summary of the Judgment

The Court of Appeal granted leave and held that the Recorder made several guideline errors, producing sentences that were not just lenient but unduly lenient. It quashed and substituted:

  • Second offender: 3 years’ immediate imprisonment for conspiracy to defraud.
  • First offender: 2 years 8 months for fraud, plus 16 months for each drug offence (drug terms concurrent with each other but consecutive to the fraud term), producing a total of 4 years.

3) Analysis

3.1 Precedents and authorities cited

R v Wharf [2015] EWCA Crim 2320 and R v BAB [2024] EWCA Crim 712

The Recorder treated reliance on “high impact” as impermissible “double counting” because vulnerability had already been considered under culpability. The Court of Appeal rejected that approach as contrary to the structure of the fraud guideline and confirmed that vulnerability may be relevant in two different ways:

  • Culpability: where offenders deliberately target victims on the basis of vulnerability; and
  • Harm: where the actual impact on victims is especially severe because of that vulnerability.

This is consistent with the endorsement in R v Wharf [2015] EWCA Crim 2320 (at [9]) and R v BAB [2024] EWCA Crim 712 (at [34]) that guideline assessment can legitimately account for vulnerability under both headings without committing an error of duplication.

R v Samurio [2016] EWCA Crim 1948

The Court reiterated that in conspiracy sentencing the court should have regard to intended loss as well as actual loss. Citing R v Samurio [2016] EWCA Crim 1948 (at [25]–[27]), the Court emphasised that conspiracies are often detected before full harm is realised; sentencing must reflect the harm the conspirators set out to cause (and the risk they created), not merely the sum successfully stolen.

Statutory anchor: section 63 Sentencing Act 2020

The Court linked the “intended loss” principle directly to s.63 Sentencing Act 2020, which requires consideration of harm that:

  • was caused,
  • was intended to be caused, or
  • might foreseeably have been caused.

That statutory framing reinforces why “attempted” or incomplete fraud conspiracies are not sentenced by reference to actual loss alone.

3.2 Legal reasoning

(a) Starting points are not contingent on “all factors” being present

A central error was the Recorder’s apparent view that the category 3A fraud starting point (3 years) applied only if “all” high-culpability factors were present. The Court stated the opposite: the guideline starting point applies to the category, and the presence of multiple high-culpability features should create “significant upward pressure” and lead to an upward adjustment.

On these facts, the Court accepted multiple high culpability features including sophistication/planning, sustained activity, very large numbers of calls and potential victims, and deliberate targeting of vulnerable victims. Those features demanded movement upwards within the range.

(b) Vulnerability: not double counting, but two distinct functions

The Court held it was “entirely appropriate” to move the fraud harm categorisation upwards (from category 4 by monetary loss into category 3) due to high impact on particularly vulnerable victims. The Recorder’s “double counting” objection was an error because vulnerability can operate:

  • as a culpability driver (selection/targeting), and
  • as a harm escalator (real-world impact and trauma).

(c) Previous similar convictions: a separate aggravator, not something to be “neutralised”

Both offenders had extensive and strikingly similar fraud convictions, including a prior conspiracy committed together. The Recorder treated further reliance on prior convictions as “double counting” because the offenders’ “previous form” contributed to classifying the case as culpability A. The Court rejected that reasoning: the culpability A factors were made out on the facts of the offence itself (sophistication, targeting, scale), irrespective of antecedents. The antecedents were therefore a distinct aggravating feature requiring further upward adjustment.

(d) “Scatter gun” calling does not reduce culpability

The Recorder suggested that making many calls in a “scatter gun” approach lessened culpability. The Court held this was wrong: mass calling was part of a sophisticated system to find and exploit susceptible victims; at minimum it reflected a hope and expectation of locating vulnerable targets. It therefore aggravated, rather than mitigated, culpability.

(e) Goodyear indications and Attorney General references

The second offender had sought (and effectively received) a Goodyear indication that a suspended sentence would follow. The Court made clear it was not bound by the indication when determining undue leniency on a s.36 reference; the appellate task is to decide whether the sentence passed was unduly lenient and, if so, to substitute one that is not.

(f) Totality and consecutive sentencing for unrelated offences

For the first offender, the Recorder imposed full concurrency between fraud and drugs despite “no connection” between the two. The Court held totality required a different structure: it reduced the drug component to 16 months (reflecting totality moderation) but made it consecutive to the fraud term, producing a 4-year aggregate that properly reflected distinct criminality. This illustrates a practical appellate approach: rather than simply stacking full terms, the Court calibrates consecutive sentences to achieve a proportionate overall figure.

(g) Note on an apparent slip in the judgment

At [59] the judgment refers to “the sentence of 30 months imposed by the Recorder on the second offender”; this appears inconsistent with the earlier description that the second offender received a 2-year suspended sentence. The Court’s substantive disposal, however, is clear: it substituted 3 years’ immediate custody for the second offender and 4 years total for the first offender.

3.3 Impact

  • Fraud guideline discipline: Sentencers should avoid treating “vulnerability” as a one-use factor. Targeting vulnerability (culpability) and heightened real-world effect (harm) can both be reflected.
  • Recidivist “courier fraud” cases: Where offenders have prior similar convictions, this decision supports robust upward movement towards the top of the category range, especially where the modus operandi is repeated.
  • Conspiracy valuation: The decision reinforces that intended loss/risk created can justify higher harm categorisation even where actual loss is modest, particularly with large-scale attempted victim contact.
  • Totality in mixed-offence sentencing: It provides a clear template: where offences are unrelated, some consecutivity is ordinarily necessary; if concurrency is chosen, an uplift must ensure overall criminality is reflected.
  • Limits of Goodyear comfort: Defendants and practitioners are reminded that a Goodyear indication does not immunise a sentence from later increase on an unduly lenient sentence reference.

4) Complex concepts simplified

Unduly lenient sentence (s.36 Criminal Justice Act 1988)

This is a mechanism allowing the Attorney General (or Solicitor-General) to refer certain Crown Court sentences to the Court of Appeal if they are arguably too low. The Court does not ask whether it would have imposed a different sentence; it asks whether the sentence fell outside the range reasonably open to the judge—i.e., it was “unduly” lenient.

Culpability vs harm in guidelines

Guidelines typically separate:

  • Culpability: how blameworthy the conduct was (planning, sophistication, targeting, leadership, abuse of trust); and
  • Harm: the damage done or risked (loss amount, intended loss, number of victims, seriousness of impact).

A single real-world feature—like victim vulnerability—can legitimately be relevant to both, because it can describe both the offender’s method (targeting) and the victim’s experience (impact).

Starting point, category range, and upward/downward adjustment

A guideline “starting point” is not a cap and not reserved for extreme cases; it is the reference sentence for the typical case within that category. Judges then move up or down within the “category range” depending on aggravating/mitigating features, including prior convictions.

Totality and consecutive sentences

When sentencing for multiple offences, the court must ensure the overall sentence is proportionate to overall offending. If offences are unrelated, concurrency can understate total criminality; consecutive terms (often with a moderation discount) may be needed to reach a fair overall total.

5) Conclusion

[2026] EWCA Crim 490 is a clear corrective in high-volume “courier” fraud sentencing. It confirms that vulnerability can properly elevate both culpability and harm; that repeated, strikingly similar convictions must be treated as a powerful aggravating feature rather than “double counted away”; and that unrelated drug and fraud offending typically requires consecutive sentencing (with appropriate totality moderation). The decision also underscores that Goodyear indications cannot prevent appellate intervention where the resulting sentence is unduly lenient.