The authority that decides the hardest question in any penalty appeal: what happens to the taxpayer who suspected something was wrong and chose not to look. Blind-eye knowledge needs a targeted suspicion and a deliberate decision not to confirm it. Negligence, however gross, is not enough.
On this page
Full name: Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd and others (The Star Sea)
Citation: [2001] UKHL 1; [2003] 1 AC 469
Court: House of Lords (including Lord Hobhouse of Woodborough and Lord Scott of Foscote)
Judgment: 2001
Subject: Blind-eye knowledge; privity to unseaworthiness under s39(5) Marine Insurance Act 1906; the duty of greatest good faith after proceedings begin
Result: The assured succeeded. Blind-eye knowledge requires a targeted suspicion and a deliberate decision not to confirm it.
Why a Marine Insurance Case Governs Tax Penalties
The hardest question in the behaviour category of a tax penalty is what to do with a taxpayer who suspected something was wrong and chose not to look. HMRC calls it turning a blind eye and treats it as deliberate. Taxpayers call it carelessness at worst.
There is no tax case that defines blind-eye knowledge. There is a House of Lords authority that does, and it is The Star Sea. The tribunals have drawn on it directly, most visibly in CPR Commercials Ltd v HMRC [2023] UKUT 61 (TCC), and it now supplies the working test in penalty appeals.
The Facts
The Star Sea was a vessel insured under a time policy. She was seriously damaged by fire and became a constructive total loss. The owners claimed under the policy. The underwriters resisted the claim on two principal bases.
First, they said the vessel had been sent to sea in an unseaworthy state, in particular that the crew were not competent to operate the fire-fighting equipment and that the carbon dioxide system had not been properly deployed, and that the assured was privy to that unseaworthiness within s39(5) of the Marine Insurance Act 1906. Two sister ships had previously suffered fires, which the underwriters said should have alerted the owners to the problem.
Second, they alleged a breach of the continuing duty of greatest good faith in the way the claim had been presented and pursued after litigation commenced.
The critical question on the first issue was what “privity” requires. It is not enough that the assured ought to have known. But did it extend to an assured who suspected the vessel might be unseaworthy and deliberately refrained from investigating?
The Ratio Decidendi
The two limbs unpacked
- A targeted suspicion. The suspicion must be of the particular facts said to be known. A general awareness that things might not be in order is not enough. Lord Scott put it in terms of the suspicion being firmly grounded and targeted on specific facts.
- A deliberate decision not to confirm. There must be a conscious choice to avoid the enquiry because of what it might reveal. This is the element that converts suspicion into knowledge: the person has effectively decided that they would rather not be told.
Why negligence is expressly excluded
Their Lordships were explicit that the concept is not a substitute for a failure to take reasonable care. A person who simply did not think to ask, or who was disorganised, or who trusted someone they should not have trusted, has not turned a blind eye. That is negligence, and the law of negligence has its own consequences. Blind-eye knowledge is a form of knowledge, and it must be reserved for cases where the person in substance knew.
Obiter and the Second Issue
- The continuing duty of good faith. The House also considered whether the duty of greatest good faith in s17 of the 1906 Act continued to apply, in its full pre-contractual form, after the assured had commenced proceedings. It held that it did not: once litigation has begun the parties’ obligations are governed by the rules of procedure and the court’s powers, and the draconian remedy of avoidance is not apt. Important in insurance law, and not part of the blind-eye ratio.
- Observations on the relationship with recklessness. The discussion of how blind-eye knowledge relates to recklessness and to dishonesty is analytical commentary. It matters in the tax context because the concepts are frequently run together in HMRC correspondence, and they are not the same.
- The evidential difficulty. Remarks about how rarely blind-eye knowledge will be established on the evidence are practical observation rather than rule, but they set the tone that appellate courts have maintained, and they are the best short answer to an HMRC officer who treats a failure to check as self-evidently deliberate.
How the Test Operates in Tax Penalties
The statutory question in a Schedule 24 FA 2007 penalty, and in the extended time limit provisions, is whether an inaccuracy was deliberate. The working definition comes from Auxilium Project Management Ltd v HMRC: the taxpayer knowingly provides HMRC with a document containing an error, intending HMRC to rely on it as accurate. The Supreme Court in HMRC v Tooth confirmed that a deliberate inaccuracy requires an intention to mislead.
The Star Sea fills the gap between those propositions and the reality of most enquiries, where nobody admits to having known. Its function is to mark the boundary:
| What the taxpayer did | Category |
|---|---|
| Knew the figure was wrong and submitted it anyway | Deliberate |
| Suspected the specific problem and consciously chose not to check because of what checking might reveal | Deliberate (blind-eye knowledge) |
| Had a general unease that the records were poor and did not investigate | Careless |
| Relied on an adviser without checking | Careless |
| Did not think about it at all | Careless, or no penalty |
Practitioner Application
Resisting a blind-eye allegation
- Make HMRC identify the suspicion. Ask, in writing, what specific fact the taxpayer is said to have suspected, and what evidence supports it. A generalised assertion that the taxpayer must have known something was wrong does not engage the test.
- Make HMRC identify the decision. The second limb is a positive act of avoidance. What is the evidence that the taxpayer decided not to enquire, and that they did so because of what the enquiry would have revealed?
- Reframe honestly as carelessness. Where the client genuinely did not check, say so plainly, explain why, and place it in the careless category. A candid carelessness case is far more credible than a denial that anything was amiss, and the penalty range is materially lower.
- Distinguish the concepts. Blind-eye knowledge, recklessness, dishonesty and carelessness are four different things. HMRC correspondence frequently uses them interchangeably. Separating them in the response often exposes that no case on deliberateness has actually been pleaded.
- Prepare the client to give evidence. The test is subjective and about state of mind. It is decided on the taxpayer’s own account under cross-examination, and findings made at first instance are almost impossible to disturb: see Edwards v Bairstow.
- Remember what turns on it. A deliberate finding opens the twenty-year assessment window, permits publication of deliberate defaulters’ details, supports a personal liability notice against a company officer, and raises the risk of a COP9 or criminal referral.
Frequently Asked Questions
What is blind-eye knowledge?
A suspicion that specific relevant facts exist, combined with a deliberate decision to avoid taking any step that would confirm them. Both limbs are required. The House of Lords in Manifest Shipping v Uni-Polaris (The Star Sea) [2001] UKHL 1 held that a failure to make enquiries which a reasonable person would have made, however careless or grossly negligent, is not blind-eye knowledge.
Why is a marine insurance case relevant to tax?
Because no tax case defines blind-eye knowledge, and the tax tribunals have drawn directly on The Star Sea when deciding whether a taxpayer who suspected a problem and did not check acted deliberately. It was applied in that way in CPR Commercials Ltd v HMRC [2023] UKUT 61 (TCC), and it now supplies the working test in behaviour disputes under Schedule 24 FA 2007.
If I should have known, does that make it deliberate?
No. “Ought to have known” is the language of carelessness, not of knowledge. The Star Sea expressly excludes negligence, however serious, from blind-eye knowledge. HMRC officers frequently reason that because a taxpayer should have realised, they must have suspected, and because they did not check, they must have chosen not to. That chain of inference is precisely what the case forecloses.
What is the difference between blind-eye knowledge and recklessness?
Recklessness involves appreciating a risk and going on to take it. Blind-eye knowledge involves suspecting a specific fact and deliberately avoiding confirmation of it, so that the person is treated as having known. They overlap but are not the same, and neither is the same as dishonesty or carelessness. HMRC correspondence often uses all four interchangeably, and separating them frequently exposes that no case on deliberateness has actually been pleaded.
How do I disprove a blind-eye allegation?
Force HMRC to particularise both limbs: which specific fact was suspected, on what evidence, and what evidence shows a deliberate decision to avoid confirming it. Then give a candid account of what the client actually did and why, placing it in the careless category if that is the truth. The test is subjective, so the client’s own evidence is usually decisive and should be prepared properly.