Conflict of interest policy.

What a conflict of interest policy is for

A conflict of interest policy sets out how your organisation recognises, declares and manages situations where someone's personal interest could influence, or appear to influence, a decision they take on the organisation's behalf.

The purpose is not to prevent conflicts. In a small sector, on a volunteer board, or in any organisation that buys from local suppliers, conflicts are unavoidable and often harmless. The purpose is to make sure a conflict is visible before the decision, and provable afterwards.

The test a good policy applies is not "was this decision improper?" but "could a reasonable outsider, knowing what we knew, think it might have been?" That is why a perceived conflict is treated as seriously as a real one.

It helps to distinguish three states early in the document, because staff use the words interchangeably and the policy should not. An actual conflict exists now. A potential conflict of interest would arise if something else happened, such as a tender being awarded. A perceived conflict exists whenever an outsider could reasonably believe one does, whether or not it is real. All three are declarable, and a policy that only asks about the first will collect almost nothing.

The types of conflict a policy must cover

A policy that only names financial interests will miss most of what actually arises.

• Financial interest. A direct or indirect benefit: a contract with a company you own shares in, a payment, a discount, a gift or hospitality above a stated threshold.
• Personal relationship. A family member or close friend who is an applicant, a supplier, a service user or a colleague you line manage.
• Conflict of loyalty. A duty to another organisation. Someone serving on two boards owes both of them the same undivided loyalty, and cannot always give it.
• Professional or reputational interest. A decision that affects your own standing, research, or future employment.
• Political interests. Where a role or affiliation could be seen to shape a decision taken in a public or charitable context.
• Perceived conflicts of interest. No actual benefit, but an appearance that would be hard to defend if published.

Each of these can be actual, potential or perceived. A policy should say plainly that all three are declarable, because staff routinely assume that only the first counts.

What the policy must include

Whatever its length, a workable conflict of interest policy answers seven questions.

1. Who it applies to. Trustees and non-executive directors, employees, volunteers, contractors and anyone taking a decision on the organisation's behalf. Say so explicitly rather than leaving it to inference.
2. What must be declared. With examples, and with a stated threshold for gifts and hospitality.
3. When. On appointment, annually, and immediately whenever a new conflict arises. The third is the one that gets skipped.
4. How. To whom the declaration goes, in what form, and to a line manager or the governance lead rather than into thin air.
5. Where it is kept. A register of interests, who maintains it, and who may inspect it.
6. How the conflict is managed. The options, in ascending order: note it; withdraw from the discussion; withdraw from the decision; step back from a role entirely. Say who chooses.
7. What happens if it is not declared. The consequences, and the route for raising a concern about someone else.

The management step is where most policies go vague. "The chair will decide an appropriate course of action" is not a procedure, and it puts the whole weight on one person's judgement at the worst moment. Say what mitigation applies to which kind of conflict, and give an escalation route for the case where the conflict sits with the chair.

Two supporting documents make the policy work. A declaration of interest form short enough that people complete it honestly, and an interest register that is genuinely maintained rather than compiled the week before an audit. Cross-reference the code of conduct too, since deliberate non-disclosure is normally disciplinary action rather than a governance matter, and staff should be able to see that in one place.

Managing conflicts in a meeting

The usual sequence is: the person declares the interest at the start of the item; the meeting decides whether the conflict is significant; the person withdraws from the discussion, the vote, or both; and the minutes record the declaration, the decision about it, and the withdrawal.

Two details are commonly missed. A conflicted person should normally leave the room rather than sit silently, because presence itself influences a discussion. And where the conflict is such that the person cannot be counted for quorum, the meeting needs to know that before it proceeds, not afterwards.

For charities, benefiting personally from the charity is restricted by law rather than only by policy, and trustees must usually obtain authority before doing so. That is one of the six main duties of a charity trustee and not a matter of internal preference.

What happens if a conflict is not disclosed

The consequences run further than an awkward conversation. A decision taken by a conflicted person who did not declare may be voidable, and a contract or a procurement award can be unwound. In a charity, an undeclared benefit may have to be repaid and can amount to a breach of trustee duty. In a regulated organisation it becomes a governance failing an inspector will record.

The reputational consequence is usually worse than the legal one, and it does not depend on anything improper having happened. "Nobody was told" is the whole story a journalist needs.

Deliberate non-disclosure is normally a disciplinary matter, and a policy should say so rather than implying it.

Is a conflict of interest policy a legal requirement?

There is no single statute requiring every organisation to hold one. But the duty it implements is legal in several settings: company directors owe statutory duties to avoid conflicts and to declare interests in proposed transactions; charity trustees must act only in the charity's interests and manage conflicts; and public bodies and many funders impose it by condition rather than by statute.

In practice, a funder, an auditor, an inspector or a bank will ask to see the policy and the register of interests. Whether or not a law names it, you will be asked to produce it.

The register is only worth what you can produce

Almost every organisation has the policy. Far fewer can produce the evidence, and the gap is not the register itself. It is the declarations that never reached it.

A conflict is very often declared in the moment and in passing: a message to the chair before a meeting, a remark in a group chat while a shortlist is being discussed, a text saying "I should say I know one of the applicants". The person did exactly the right thing. But that declaration lives on a personal phone, and the register only ever received the ones somebody remembered to transcribe.

Two years later, when a procurement is questioned, the organisation's position rests on whether it can show the interest was known and managed. If the honest answer is "it was declared, but we cannot show you where", the policy has not protected anyone, and the person who behaved correctly is the one most exposed.

That is a records problem rather than an ethics one, and it is worth asking at your next governance review whether the declarations your people actually make are landing anywhere you could produce them from.

Where to read the official guidance

For charities, the Charity Commission's Conflicts of interest: a guide for charity trustees (CC29) is the authority and includes a decision tree worth copying. For companies, the duties sit in sections 175 and 177 of the Companies Act 2006.

This page is a summary, not legal advice. Where a decision turns on the detail, take proper advice.

Related guides

• Charity trustee responsibilities: the six main duties
• Minute taking: what to record and what to leave out

All guides