Inside information and the MNPI wall: building the control the FCA assumes you do not have
UK MAR, in particular the definition of inside information (Article 7), the prohibitions on insider dealing and unlawful disclosure (Articles 8 and 10), and the market soundings regime (Article 11); and the FCA's expectations on the buy-side's use of expert networks and primary research.
A fund running expert calls, channel checks, corporate access and market soundings routinely encounters inside information. In primary research, proximity to a genuine edge means proximity to non-public information. The FCA works from a plain premise about firms that operate this way. The FCA’s working assumption is that most firms cannot show a functioning wall between the information research produces and the trades the desk places.
The FCA looks for documented evidence, not a policy statement. A policy describing a wall is not proof one operates; the firm has to evidence its operation. The fund has to show, from end to end, that when potentially price-sensitive information arrived it identified it, contained it, decided what to do with it, and kept it off the order pad until that decision had been made and recorded. On MNPI, the gap between a documented control and an evidenced one can carry criminal liability.
Most failures start before any trade, with a fund that does not recognise it is holding inside information at all. So the definition comes first. Under Article 7 of UK MAR, inside information is information of a precise nature, which has not been made public, relating to one or more issuers or financial instruments, and which, if it were made public, would be likely to have a significant effect on the price of those instruments or their derivatives. Precise rules out vague impressions but takes in specific facts and well-founded expectations. Not public is a low bar, and a private expert call often clears it. Price-sensitivity has to be judged in the moment, not reconstructed once a trade has gone wrong.
The prohibitions follow from that definition. Article 8 covers insider dealing: using inside information to acquire or dispose of the relevant instruments, and it reaches the cancelling or amending of an order placed before the information was held. Article 10 covers unlawful disclosure, passing inside information to someone outside the normal exercise of employment, as read through the FCA’s MAR sourcebook. None of this is abstract for a hedge fund. The analyst who hears something precise and non-public on an expert call, and the portfolio manager who trades on the back of it, are the chain these provisions target. And the guidance at MAR 1.2.3 is worth holding onto: UK MAR does not require the person to have intended to commit market abuse. A fund cannot defend a trade on inside information by saying nobody meant any harm. The offence is the use of the information, which is why the wall has to stop the information reaching the trade rather than rest on the good faith of whoever holds it.
The market soundings regime in Article 11 carries an obligation that belongs specifically to the buy-side, and it is widely under-implemented. A market sounding is the communication of information ahead of an announced transaction to gauge investor interest. Much of the regime governs the disclosing side, but Article 11(7) puts a duty on the recipient: whatever steps the discloser has taken, the person receiving the sounding must assess for itself whether it now holds inside information, and when that information has ceased to be inside information. The implementing standards require the recipient to run procedures that control the internal flow of that information, to train the relevant staff, and to keep records. Both sides of the regime are under review. The EU’s Listing Act is recasting parts of it as more of a safe harbour, and the UK is reviewing its own market abuse framework. But the recipient’s duty to assess and to control flow is the substance a fund has to operate whatever the wrapper becomes.
One distinction is worth settling, since funds routinely confuse it. The insider-list obligation in Article 18 of UK MAR falls on issuers and the persons acting on their behalf, who must list everyone with access to inside information about that issuer. A hedge fund is not usually the keeper of an issuer’s insider list, though it can find itself placed on someone else’s when it is wall-crossed, and it has to maintain its own restricted and watch lists to control the information it receives. The fund’s lists are not the same artefact as an issuer’s Article 18 list. They are the buy-side’s functional equivalent, and the FCA expects them to be as live and as controlled.
The assumption comes from the structure of the business, and the enforcement record bears it out. A fund that runs primary research and takes expert calls comes near inside information constantly, and its controls are usually thinner than its research effort.
Expert networks are the oldest vector and carry the longest enforcement history. The case against Primary Global Research, the expert-network firm that passed information from corporate insiders posing as consultants, set the template, and regulators on both sides of the Atlantic have probed expert-network controls since. The expectation now is spelled out: a hedge fund manager should monitor how it uses expert networks, control the receipt, identification and use of inside information, and run pre- and post-trade controls around it. The good practices the FCA recognises are the ones a fund is expected to evidence: capping the number of calls with any single expert, because the risk of picking up MNPI rises as the relationship deepens; obtaining the network’s logs of who spoke and what was discussed; and excluding experts who are current or recent insiders of relevant issuers.
The enforcement context is live. The FCA has kept bringing insider dealing cases, including against individuals who traded on precise, non-public information for modest profits, which tells the market it pursues the behaviour regardless of size. For a fund, the exposure is that a single trade can trigger enforcement. Its defence, if it has one, is the evidenced wall showing it identified the risk and controlled it.
MNPI reaches a hedge fund through a small set of recurring channels, and an evidenced wall answers each one on its own terms rather than leaning on a single general policy. The list below maps each vector to the control and to the evidence the FCA looks for.
- Expert networks. The risk is that a current or recent insider, or an over-familiar expert, conveys precise, non-public, price-sensitive information on a call. The control is an approved-network list, pre-approval of each consultation against the fund’s holdings and restricted names, exclusion of disqualified experts, limits on repeat calls with the same expert, a compliance presence or a contemporaneous note, and the network’s own log. The evidence is a per-call record of who was approved, what was discussed, and the post-call assessment of whether MNPI was received.
- Primary and channel-check research. The risk is that non-public data points, supplier checks, footfall counts and channel inventory aggregate into something precise and price-sensitive, crossing from a permissible mosaic into MNPI. The control is a research methodology that documents the public and non-material inputs behind a thesis, and an assessment whenever a check returns something specific and non-public. The evidence is a contemporaneous record that the investment thesis rested on a lawful mosaic of analysis rather than a single piece of inside information.
- Corporate access and management meetings. The risk is that an issuer or its management conveys something precise and non-public in a meeting, an idea dinner or a conference. The control is a wall-crossing protocol, a post-meeting assessment, and an update to the restricted or watch list where information was received. The evidence is the wall-crossing log and the assessment, dated to the meeting.
- Market soundings. The risk is that the fund is sounded on an upcoming transaction, receives inside information, and is restricted from trading the relevant instruments until that information is public or cleansed. The control is a named gatekeeper who receives soundings, the Article 11(7) assessment of whether the fund now holds inside information, immediate addition to the restricted list, and a cleansing process to determine when the restriction lifts. The evidence is the sounding record, the assessment, the restriction and the cleansing decision, from one end to the other.
- What is regulated is the firm’s assessment of the information, which must be made and recorded at the time. A fund that gathers information through expert calls, channel checks, meetings and soundings but keeps no contemporaneous MNPI assessment has skipped the control the law requires.
Behind the four vectors sits the wall: the set of controls that contains inside information once it arrives and proves, afterwards, that it was contained. An evidenced wall has six working parts.
A control room function that owns the restricted and watch lists. Someone has to hold the lists. The restricted list names instruments the fund may not trade because it holds, or may hold, inside information; the watch list names instruments under heightened monitoring. A list that exists is not enough. The lists have to be live, updated as information arrives from any of the four vectors, and wired into the trading process rather than parked in a spreadsheet nobody opens.
Pre-clearance of trades against the lists. The wall only does its job if it catches trades before they happen. An order in a restricted name should be blocked, and the block, along with any override and the reasoning for it, should be recorded. To function as a control, a restricted list must block orders pre-trade.
A cleansing process. Inside information does not stay inside forever. It becomes public, or it goes stale, or the transaction it related to completes. The fund needs a defined process for deciding when information has ceased to be inside information, so a name comes off the restricted list on a reasoned basis rather than quietly drifting off it. Article 11(7) makes that assessment an express duty for soundings, and the same discipline should apply to information from every vector.
Information barriers and need-to-know. Where one part of the firm holds inside information, the barrier that keeps it from the rest, physical, system-based and behavioural, has to be real and tested. The smaller the firm, the harder real separation becomes. A two-person investment team cannot wall information from itself, so for a small fund the control is more often restriction and abstention than separation, and the policy should say so.
The mosaic record. For a research-driven fund, a contemporaneous note that the thesis rested on lawful analysis is key evidence. When a position is questioned later, what decides whether it reads as a defensible mosaic or an indefensible tip is whether the fund can show, from the time, the public and non-material inputs that built the view.
The surveillance overlay. Trade and communications surveillance reinforces the wall by catching whether information that should have been contained reached a trade anyway. The firm needs both the wall and the surveillance so it can show the regulator that it does more than intend to keep MNPI off the order pad. It has put controls in place to check that it did.
This insight is provided for general informational purposes only and doesn’t constitute legal, investment, or regulatory advice.