
Season 11 · Episode 10 · Trusts Law · 23 min
A trustee sells a field at full value, reinvests every penny properly, leaves the fund no worse off, and is still in breach of trust.
In this episode
Try it yourself
A trust instrument provides that the trustee is not to be liable for any loss except loss caused by his own fraud. The trustee makes a series of investments in ventures he has not troubled to investigate, disregarding written warnings from his co-trustee that they are unsuitable. Heavy losses follow. It is accepted on all sides that he took no benefit for himself and believed throughout, however unreasonably, that what he was doing was in the beneficiaries' best interests. They say his recklessness was fraud.
Does the trustee's conduct amount to fraud, so that the clause cannot protect him?
Listening teaches. Practice passes.
This topic has 30 exam-style questions in the bank — 4,400+ across SQE1, with mock exams, flashcards and weak-topic tracking. Lifetime access is £69.99.
A man is the sole trustee of a family trust whose main asset is a farm. The trust instrument says he may not sell any of the land without the life tenant's prior written consent. Believing that applied only to the farmhouse, he sells one of the fields. Full market value. Proceeds properly reinvested. The fund is not a penny worse off. Has he committed a breach of trust? He has.
Innocence is no defence, and neither is the absence of loss. Those go to the remedy, and to whether a court will excuse him. Not to whether he was in breach at all. This is Trustees' Liability and Protection, and it has a clean shape. What counts as a breach, what it costs, and what can save you. Keep our farmer in mind.
Here is the route. What a breach is, and the three remedies that follow. Then causation, which decides how much. Then the protections, in order. An exemption clause, relief from the court, and the beneficiaries' own consent. Then limitation, where the trap is bigger than people expect. Then the trustee's indemnity and personal liability. And last, varying the trust.
A breach of trust is a failure to comply with the terms of the trust instrument, or with the duties the law imposes. Positive duties, like the duty to invest. Negative duties, like the duty not to profit. And it does not matter whether the trustee meant it.
The standard categories are worth having. Misappropriating trust property. Negligence in managing investments. Failing to invest at all, or investing outside the permitted range. Conflict of interest. Failing to distribute income. And improper delegation. Leave money sitting idle in a current account for two years while you decide nothing, and that is a breach.
Now the remedies, and be precise, because the exam rewards precision here. There are three, and they do different jobs. Restoration puts the fund back where it would have been. An account of profits strips the trustee of gains. Equitable compensation makes good losses. The court may order one, or two, or all three.
An account of profits does not depend on the trust having lost anything. Take a trustee who quietly puts £50,000 of trust money into a start-up in her own name. The shares are now worth £120,000. The rest of the fund has done well, so the trust is no worse off. She must still account for the £70,000 gain.
The no-profit rule does that, and the fact that the beneficiaries would probably have agreed is no answer where she never asked them. She took the opportunity as a trustee, so the profit belongs to the trust.
Equitable compensation is the other side. It is compensatory, so it needs a causal connection between the breach and the loss, and that is where most marks are won. The question is not what went wrong during this trusteeship. It is what would not have happened but for the breach.
That is Target Holdings v Redferns. A solicitor trustee released mortgage money in breach of trust, but the lender's loss flowed from a fraudulent overvaluation and would have happened anyway. So the trustee was not liable to restore the sum. The Supreme Court reaffirmed that but for approach some years later.
Work the arithmetic, because it comes up. A trustee takes £100,000 in breach of the investment restrictions and puts it into a speculative venture. The whole lot is lost. Expert evidence shows that in the authorised investments he should have chosen, it would have fallen with the market and be worth £60,000 today. What does he pay?
£60,000. Not £100,000. Compensation is assessed with the benefit of hindsight, at the date of judgment, by asking what the fund would be worth had the trust been performed properly. Properly invested it would hold £60,000. It holds nothing. That gap is the measure, and the market fall that would have happened anyway is not his to bear.
Hindsight cuts the other way too. Sell trust shares in breach when they were worth £50,000, and if they would be worth £90,000 today, the trustee restores £90,000. He cannot freeze his liability at the value on the day he got it wrong.
Two more points on quantum. A trustee generally cannot set a gain on one breach against a loss on another, unconnected one. Make £30,000 on one unauthorised investment and lose £50,000 on a separate second, and he accounts for the gain and makes good the loss. Set-off needs a single transaction or course of dealing.
And a trustee who restores misapplied money pays interest on it too. Usually simple interest. But where he used the money in his own trade or business, the court may award compound interest.
Where two trustees commit the same breach they are jointly and severally liable, so the beneficiaries may recover the whole loss from either. Sue the wealthy one for the entire £120,000 and he cannot insist on paying half. He may seek contribution from his co-trustee, but his co-trustee's poverty is not the beneficiaries' problem.
A new trustee is not liable for breaches committed before she was appointed. What she must do is acquaint herself with the state of the trust on taking office. If she finds a predecessor's breach, she must take proper steps to remedy it or recover from him.
Now the protections, and there are four worth knowing. The first is in the trust instrument itself. An exemption clause.
Armitage v Nurse settled how far these go. Such a clause is valid and takes effect according to its terms, and the only liability it cannot exclude is the trustee's own actual fraud. Liability for negligence, even gross negligence, can be excluded, and professional trustees rely on that every day.
Which makes the definition of fraud the whole battleground. Millett LJ put it as an intention to pursue a course of action, either knowing it is contrary to the beneficiaries' interests, or being recklessly indifferent whether it is. That is a state of mind, not a measure of carelessness.
So carelessness, however gross, falls outside it and can be excluded. Reckless indifference to whether you are harming the beneficiaries does not, and cannot. Hold that distinction, because you are going to need it in a few minutes.
The second protection is the court. Under s.61 of the Trustee Act 1925 the court may relieve a trustee wholly or partly from personal liability. That is where he has acted honestly and reasonably, and ought fairly to be excused for the breach.
Three limbs, and all three must be satisfied. Honestly. Reasonably. And ought fairly to be excused. So never write that a trustee will be relieved. The jurisdiction is discretionary, and relief may be partial only.
Back to our farmer. He was honest, and he misread a document. Whether that was reasonable will turn on how clear the clause was and whether he took advice. Contrast a lay trustee who instructs solicitors and a regulated adviser, gives a written policy statement, follows the advice, and finds the advice was wrong. Honest, reasonable, and fairly to be excused.
The third protection is the beneficiaries themselves. One who is of full age and capacity, and who knowing the material facts consents to a breach or afterwards ratifies it, cannot later sue on it. Change your mind two years later and the release still stands.
There is a sharper version. Under s.62 of the same Act, take a trustee who commits a breach at the instigation or request of a beneficiary, or with that beneficiary's consent in writing. The court may impound that beneficiary's interest to indemnify the trustee.
The fourth protection is statutory. The Trustee Act 2000 imposes a duty of care measured by the ordinary prudent person of business. Then, by s.23, it provides that a trustee is not liable for the acts or defaults of an agent, nominee or custodian.
But read the exception. He is not liable unless he failed to comply with the duty of care when entering into the arrangement, or when reviewing it under s.22. Choose a regulated firm carefully, then hear nothing for six years while it abandons the policy statement and loses the fund, and the review duty is what catches you.
Now limitation, and this is where candidates lose marks they should not. Under s.21(3) of the Limitation Act 1980, an action by a beneficiary for breach of trust must be brought within six years of the date the right of action accrued. For breach of trust, that is the date of the breach.
Note what that does not say. Not six years from discovery. There is no general discovery rule here. A beneficiary who never asked to see the accounts, and finds out eight years later about an honest unconcealed breach, is simply too late.
Time is postponed to discovery only where the trustee deliberately concealed the breach, or committed it deliberately in circumstances unlikely to be discovered. That is s.32, a different rule with a different trigger. Failing to volunteer information is not concealing.
Then two situations where no limitation period applies at all. First, under s.21(1)(a), fraud. Where the breach is fraudulent and the trustee was a party or privy to it, the beneficiary may sue however many years have passed. Twenty years, and a retirement from the trusteeship, make no difference.
Watch the wording there. It is not six years from discovery of the fraud. It is no period at all, subject only to equitable defences like laches. Mixing that up with the s.32 concealment rule is the classic error.
Second, under s.21(1)(b), an action to recover from the trustee trust property, or its proceeds, still in his hands or converted to his own use. This one needs no fraud. A trustee who honestly but wrongly moved a trust painting into his own name eight years ago, and still has it, hands it back.
Now the trustee's own protection in the other direction. A trustee has a right to be indemnified out of the trust assets for liabilities and expenses properly incurred in administering the trust. That right is secured by a lien over the trust property.
The lien is proprietary, which matters. Pay an insurance premium out of your own pocket because the trust account is empty, and you may hold the building until you are repaid. That holds even against beneficiaries who are together absolutely entitled and demanding it now.
But properly incurred is doing all the work there. The indemnity is lost where the expense flows from the trustee's own breach, where he acted dishonestly, where he exceeded his powers, or where he was in an unauthorised conflict. Invest improperly and lose the money, and you cannot reimburse yourself out of what is left.
It is not, though, forfeited across the board. A trustee who makes one unauthorised investment still recovers the accountancy fees he properly incurred on the annual accounts. The breach costs him the indemnity for the breach, not for everything else.
Which leads to personal liability, and one structural point explains a lot. A trust is not a legal person. It cannot contract or be sued in its own name. So a trustee who signs a contract for the trust contracts personally, and is liable to the third party for the whole of it.
Order £150,000 of building work within your powers, the trust assets fall to £100,000, and the contractor comes to you for the other £50,000. Your indemnity is worth only what the assets can bear. The shortfall is yours.
Last, variation. Start with the beneficiaries, because they can often do it themselves. Where they are all of full age and capacity, and between them absolutely entitled to the whole beneficial interest, they may together end the trust. They take the property on the terms they choose. That is Saunders v Vautier.
So take a widow with a life interest and two adult children with vested remainders. Between them they can require a country house to be sold, though the will directs it must not be. Their interests exhaust the beneficial interest, and there is nobody else for the court to protect.
What they cannot do, while the trust subsists, is dictate to the trustees how the trustees' own powers are to be exercised. Ending the trust is theirs. Running it is not.
When somebody cannot consent you need the court, under s.1 of the Variation of Trusts Act 1958. The court may approve an arrangement varying or revoking the trusts on behalf of four categories. Minors and others incapable of consenting. Persons who may become entitled in the future. Unborn persons. And discretionary beneficiaries under a protective trust.
For every category except the last, the court may approve only if the arrangement is for the benefit of the person it approves for. And benefit is not confined to money. A move offshore saving a large amount of tax may be refused where it would take young beneficiaries out of the settled family environment the settlor intended.
Why does the Act exist? Because the inherent jurisdiction is narrow. In Chapman v Chapman the House of Lords held there is no general inherent power to approve a variation for minors or the unborn merely because it benefits them. It reaches only limited categories, like salvage, emergency, maintenance, and compromising a genuine dispute.
A word on how SQE1 tests this. You will not be asked to recall section numbers, and you will meet few case names. You get a scenario and five answers, and you pick the best. The names here are memory pegs, and this topic has four good ones.
If you keep only three. Target Holdings, because compensation needs a causal link. Armitage v Nurse, because an exemption clause stops at actual dishonesty. And Saunders v Vautier, because beneficiaries who between them own everything can simply take it.
Four traps. One. An innocent breach is still a breach. Honesty and the absence of loss go to the remedy and to relief, never to whether there was a breach.
Two. Six years runs from the breach, not from discovery. There is no general discovery rule here. Discovery matters only where the trustee deliberately concealed what he did, and that is a separate provision with a separate trigger.
Three. Do not say a trustee will be relieved under s.61. All three limbs must be met, honest, reasonable and fairly to be excused, and even then the jurisdiction is discretionary and relief may be partial.
Four. Keep personal liability and trust liability apart. A trustee contracts personally, and the indemnity from the fund is a separate question that depends on whether the liability was properly incurred.
Quick check, and you were told this one was coming. A trust instrument provides that the trustee is not liable for any loss except loss caused by his own fraud. He makes a series of investments in ventures he has not troubled to investigate, disregarding written warnings from his co-trustee. Heavy losses follow. He took no benefit for himself, and believed throughout, however unreasonably, that he was serving the beneficiaries' interests.
They say his recklessness was fraud, so the clause cannot protect him. Three answers. One: yes, because conduct this careless is fraud however honestly he believed in it. Two: yes, because ignoring a co-trustee's written warnings is necessarily dishonest. Three: no, because fraud here requires dishonesty, and he honestly believed he was serving their interests. Pause here if you want a moment.
The answer is three. Actual fraud is an intention to pursue a course of action either knowing it is contrary to the beneficiaries' interests, or being recklessly indifferent whether it is. Both limbs are about his state of mind towards their interests. This trustee's investing was indefensible, but he took nothing for himself and believed he was serving them, so neither limb is met and the clause holds.
Why the others fail. Option one treats carelessness as a quantity that becomes fraud once it gets bad enough. It does not: gross negligence can be excluded. Option two makes ignoring a warning conclusive of dishonesty, when it is only evidence towards it. And note what has not saved him. He may still lose relief under s.61, which needs reasonableness.
Five things to take away. One: our farmer is in breach, because innocence and the absence of loss go to the remedy, not to liability. Two: there are three remedies, restoration, account of profits and equitable compensation, and compensation needs a causal link, assessed with hindsight at the date of judgment.
Three: an exemption clause covers everything short of actual dishonesty, and s.61 relief needs honesty, reasonableness and fairness together. Four: six years from the breach, with no limitation at all for the trustee's fraud, or for trust property still in his hands. Five: beneficiaries who between them own the whole beneficial interest can end the trust themselves, and where somebody cannot consent, the court does it under the 1958 Act. Next time, Equitable Remedies and Tracing.
Practise this topic with exam-style questions at sqe1prep.co.uk. This episode is for education and exam revision only, not legal advice, and we are not affiliated with or endorsed by the SRA or Kaplan.
Free study plan
Tell us your exam date and we’ll email a schedule that fits Trusts Law alongside the other FLK2 subjects.
Narrated by an AI voice from a script written and checked by the editors at sqe1prep.co.uk. Educational content only — not legal advice. SQE1 Prep is not affiliated with or endorsed by the SRA or Kaplan. The SQE and SOLICITORS QUALIFYING EXAMINATION trade marks are the property of and are used under licence from the Solicitors Regulation Authority.
Enjoying this? Unlock all 144 topics, mock exams & flashcards.