All resources
Resource

Managing MNPI in Private Equity

19 min read8 chapters

Deal activity in private markets creates inside information about listed names. Bolt-on diligence, listed comparable analysis, committee seats, board positions, expert calls, and alternative data feeds all generate exposure to UK MAR and the Criminal Justice Act. This guide covers the Art 7 three-limb test, the three Art 14 prohibitions, information-barrier governance, and the evidence spine that survives FCA scrutiny.

Guide

Picture a single piece of information moving through the firm from source to restriction to trade. That mental map is the whole MNPI framework in one image.

Information enters from several directions simultaneously. Deal-team diligence on a bolt-on acquisition. Internal analysis of listed comparable companies. Origination on a public-to-private transaction. A seat on a lender or creditor committee. A portfolio-company board seat taken by a nominee director. An inbound market sounding from a banker. A call to an expert network. A newly licensed alternative data feed. Each source is distinct. Each can generate inside information about a listed name with no warning.

The MAR Art 7 test gates which sources produce inside information. Only if information is precise, non-public, and price-significant about a listed issuer or instrument does the MNPI machinery engage. Note that Art 7(1)(a) extends to information relating indirectly to issuers. A private bolt-on acquisition can be inside information about a listed competitor if the bolt-on is material to its competitive position.

If information passes the Art 7 test, three prohibitions fire at once under Art 14: do not deal in the listed security for the firm's own account or on behalf of the fund; do not recommend or induce any third party to trade; and do not disclose the information outside the normal course of duties. For individuals, the criminal parallel runs alongside under Criminal Justice Act 1993 s.52, covering dealing, encouraging, and disclosing, with penalties up to ten years on indictment.

The only way to keep trading a name the firm holds inside information about is a real, evidenced information barrier operating under MAR Art 9 and SYSC 10.2. In a fifteen-person boutique where the same partners see every deal and every trade, this is genuinely fragile. The honest answer is often simpler: restrict the name. Do not trade for the life of the MNPI event. The wall you cannot evidence is worse than no wall.

A second perimeter applies on take-privates. The Takeover Code runs alongside MAR as an independent clock. The MAR Art 17 disclosure obligation ("as soon as possible") and the Rule 2.2 price-movement announcement triggers and the Rule 2.6 PUSU clock are separate requirements that must be reconciled. Each has different trigger conditions and different responsible parties.

The evidence spine is the pass mark throughout. Documented means the policy says the right thing. Evidenced means a dated artefact shows the control operated on the day it was needed. Every enforcement action since 2018 has turned on the absence of artefacts, not on actual market abuse.

Not every piece of confidential information is inside information. The three-limb test under MAR Art 7 determines which information triggers the prohibitions.

Limb 1 is precision under Art 7(2) to 7(3). Information is precise if it relates to a set of circumstances that exists or may reasonably be expected to occur, or to an event that has occurred or may reasonably be expected to occur. In protracted processes such as staged acquisitions, refinancings, and capital raises, intermediate steps can each be precise. A signed letter of intent is precise. The bolt-on target's EBITDA, known at the letter of intent stage, is precise. The mere possibility that the firm might bid is not precise until there is a binding commitment.

A practical example: a deal team is diligencing a bolt-on for a portfolio company. The process has five stages: signed letter of intent, due diligence complete, board approval, financing committed, and close. Each intermediate step, once committed, is precise. The fact that the firm is exploring options at stage zero is not. Once the letter of intent is signed, all information flowing from the due diligence is precise from that date.

Limb 2 is non-public status. Information is non-public if it is not in the public domain and is not readily ascertainable by persons using ordinary diligence. Bloomberg data on a listed company is public. Your deal team's internal analysis of how that company's results affect a target's competitive positioning is not. The analysis adds non-public judgment even if the underlying data is publicly available.

Limb 3 is price-significance under Art 7(4). Information is price-significant if a reasonable investor would be likely to use it in deciding whether to trade the listed security. The legal threshold from Hannam v FCA [2014] UKUT 0233 is a real, not fanciful, prospect of a price effect. It need not be more likely than not. A bolt-on representing 4% earnings accretion to a listed competitor passes. A bolt-on representing 0.1% of that competitor's revenue probably does not. When in doubt, restrict first and analyse afterward.

The listed-adjacency point deserves emphasis. Private equity firms habitually say market-abuse law does not apply to private markets. It applies the moment the private deal generates inside information about a listed name. The bolt-on's financials are inside information about the listed competitor. The portfolio company's covenant breach is inside information about its listed parent. The fund's take-private offer is inside information about the listed target from the moment the approach is committed.

An information barrier is the only mechanism that allows trading to continue on a name the firm holds inside information about. Under MAR Art 9, a legal person does not use inside information if it has established, implemented, and maintained arrangements that ensure the decision-maker and anyone able to influence that decision-maker is genuinely uninformed about the inside information. The SYSC 10.2 mirror requires similar segregation.

What regulators actually test on an information barrier is four things. First, it must be written: a documented policy naming the restricted teams, the restricted information, the date the restriction starts, and the scope of the prohibition. Second, it must be enforced through physical or system-based controls that prevent the information reaching the trading team. This means separate email domains in practice, no joint portfolio meetings, restricted data rooms not accessible to trading personnel, and system access logs that can be produced. Third, it must be monitored continuously. Quarterly evidence that the barrier held is the minimum: email surveillance showing no contact across the barrier, system logs showing no data leakage, sign-offs from both sides certifying isolation. Fourth, if a trade occurs on the name during a live restriction, a contemporaneous memo by the trading decision-maker must exist, certifying that the decision was based on independent analysis and that the decision-maker was not in possession of the restricted information.

For most boutique PE firms with ten to twenty people, the honest assessment of a real information barrier fails quickly. The same founding partners sit on the investment committee and also direct portfolio management. There is no genuine information segregation between someone who sees the deal and someone who directs the trade. In those circumstances, the practical answer is to restrict the name entirely and not trade for the period of the MNPI event. The cost of restriction is a temporarily frozen position. The cost of a deficient barrier is an enforcement action, because a barrier that cannot be evidenced is, in law, no barrier at all.

The relevant enforcement lessons: in Sound Point Capital Management (SEC, 2024, $1.8m), there was no barrier between credit-analysis and CLO-trading personnel. The restricted list bit only on the issuer name, not on the CLO tranche exposed to it. In Marathon Asset Management (SEC, 2024, $1.5m), there was no requirement to independently assess adviser-sourced MNPI. In Ares Management (SEC, 2020, $1m), the compliance inquiry was undocumented. None involved a finding of actual insider trading. Each involved a finding that the barrier did not exist as an operative control.

A market sounding is a communication of information before the announcement of a transaction to gauge the interest of potential investors. Under MAR Art 11, if a disclosing market participant (DMP) conducts a market sounding in the proper way, the disclosure of inside information to the sounded party is a lawful disclosure. The DMP must assess whether the information to be disclosed is inside information, get the recipient's prior consent to be wall-crossed, inform the recipient of their obligations as an insider, keep full records, and cleanse the recipient when the information becomes public.

The most common boutique error on market soundings is misidentifying who owns the Art 11 procedure. When the AIFM is soliciting co-investors or arranging financing, the AIFM is the DMP, not the broker. The AIFM owns every step of the procedure. A named gatekeeper must approve the initial sounding list and every addition, with per-name justification recorded. The FCA's 2026 market-sounding review found that larger sounding lists did not generate proportionately greater demand, and the FCA has indicated it now treats list size itself as a risk factor.

Two further points on the outbound sounding. First, a single approved script must go to all sounded parties under RTS 2016/960 Art 3(5). Using different information packages for different recipients is non-compliant. Second, cleansing is active, not passive. A recipient who declined the wall-cross remains an insider until the information is public or demonstrably stale. A response of "no" to the wall-cross does not free the recipient from restrictions.

On inbound calls, the question is when inbound information becomes the firm's inside information. Before any NDA, information shared by a banker as a market sounding is not yet a firm obligation, but the prudent practice is to treat it as restricted pending the NDA. After the NDA is signed, the firm has a contractual obligation of confidentiality, and if the three-limb Art 7 test is satisfied on the information received, the MNPI regime is fully engaged. The firm must restrict trading on the listed target immediately and log the date of receipt, the nature of the information, and the restriction.

For each inbound market sounding or opportunity call, a confirmatory email to the banker noting the date, the NDA status, and the restriction status is the minimum contemporaneous evidence. It establishes when the information was received, that the firm recognised it as confidential, and that restrictions were imposed.

On a take-private, two regulatory frameworks run simultaneously and must be reconciled. The MAR framework governs disclosure of inside information. The Takeover Code governs the conduct of the offer process, the announcement obligations, and the insider-list management of the offer perimeter.

Under MAR Art 17, an issuer must inform the public as soon as possible of inside information that directly concerns it. A potential offeror that has obtained inside information about the target has obligations under Art 14. The target has the Art 17 disclosure obligation, with delay available only to the target under Art 17(4) provided the conditions for delay are met. The potential offeror does not have Art 17 obligations as an outsider, but it does have Art 14 dealing prohibitions from the moment it holds inside information about the target.

Under the Takeover Code, Rule 2.1 requires secrecy before announcement: keep the approach confidential, pass information only on a need-to-know basis, warn recipients of their obligations, and keep numbers to a minimum. Rule 2.2 requires announcement when the approach has extended beyond a very restricted number of people or when rumour or untoward price movement occurs. The price triggers in the Rule 2.2 Note are approximately 10% above the lowest price since the approach was made, with a lower intra-day trigger of approximately 5%. The potential offeror is primarily responsible for a leak announcement.

Once named, the potential offeror must under Rule 2.6 announce a firm intention to make an offer or a "no intention to bid" statement by 5pm on the 28th day (the PUSU deadline). Only the offeree board can seek an extension. Under Rule 2.7, the firm-intention announcement is effectively binding and, for a cash offer, requires a third-party cash confirmation under Rules 2.7(d) and 24.8.

The practical reconciliation: once a board recommendation has been obtained, the inside information is firm and MAR Art 17 considerations apply to the target. The joint RNS announcement satisfies both MAR and the Code. Before a binding offer, the potential offeror must monitor the target's price daily against the Rule 2.2 Note triggers and maintain a holding-announcement playbook with Panel contact details. The MAR clock and the Code clock start on different days, run at different speeds, and have different responsible parties. Running one register that captures both insider lists, timestamps both, and reconciles both clocks is the only way to manage the two-perimeter problem in a single controlled environment.

Personal account dealing under COBS 11.7 and 11.7A requires every employee who is a relevant person to pre-clear personal trades in securities related to the firm's activities. The pre-clearance system is not a formality. It is the last line of defence between the restricted list and a trade in a restricted name.

The system must work in both directions. When a name goes onto the restricted list, it must immediately become visible to the pre-clearance function. When a personal trade request is submitted, it must be checked against the current restricted list and the answer must be recorded, not just communicated verbally. A logged denial is the evidenced control. A verbal denial with no record is not a control.

The restricted list itself requires active maintenance. A name added for a deal that died six months ago should be reviewed for removal on a schedule. The dangerous failure is not a name left on too long; it is a name that should have gone on and did not. Where the firm cannot operationally resource genuine information barriers, the model effectively becomes a single restricted list capturing every name the firm has any material information about, firm-wide, with pre-clearance as the gate for all personal dealing.

The watch list is distinct from the restricted list. A watch list contains names where trading is permitted but each trade receives heightened surveillance. The watch list is confidential to compliance and the control room. Names move between watch and restricted as information status changes. The register must record the history: add date, reason, restriction date, clear date, and the most recent scrub date. A list that shows only current state, with no historical record of why names were added and removed, cannot demonstrate that the control was operating in real time.

The evidence file for PA dealing pre-clearance needs: the current restricted and watch list with add dates and reasons; a 12-month log of pre-clearance requests showing requester, date, instrument, and outcome; at least one documented denial in the period, or a note explaining why none occurred; a post-trade surveillance report comparing PA trading to the restricted list; and annual training acknowledgements from all relevant persons.

The gap between documented and evidenced is the central problem in MNPI compliance. A firm that has written the right policies but cannot produce dated artefacts showing those policies operated in a specific case will not pass an FCA inspection or an SEC examination. The policies are the floor. The artefacts are the pass mark.

For each MNPI event, five artefacts are required. First, an information-identification memo: a dated record of when the information was received, from whom, and why it satisfies the three-limb Art 7 test as to a specific listed name. The conclusion can be that the information is not MNPI, provided that conclusion is reasoned and dated. Silence is not deliberation. Second, a restricted-list entry timestamped at or before the identification memo, with a system audit trail showing it was not added retrospectively. Third, where an information barrier is erected, a wall-crossing log with an access list, a wall-cross request form with compliance sign-off, and system-access logs confirming the IC side had no data-room access. Fourth, an Art 18 insider list covering all persons who hold the inside information, including external recipients such as co-investors, financiers, and counsel. Fifth, if any trade occurs on the name during the restriction period, a contemporaneous trading-defence memo from the decision-maker, citing the independent analysis relied on and certifying no knowledge of the restriction.

The quarterly spot-check is the governance mechanism that confirms the spine is operating. Compliance or the board reviews 20% of MNPI events from the quarter: was information identified and logged on receipt; was the restriction imposed within one business day; if a barrier was erected, can isolation controls be evidenced; if trading occurred, does the defence memo exist and is it signed; was the restriction lifted on the date the information became public. Any row answered "it is in the policy" rather than "here is the artefact" is the finding.

The artefact standard is not onerous. A dated email to the banker confirming restriction status. A restricted-list system export with timestamps. An IC attendance sheet showing no overlap with the trading decision. A signed memo from the portfolio manager. These are instruments any firm can produce if the practice is embedded. They are instruments no firm can produce retroactively if the FCA asks for them two years later.

The board's role in MNPI governance is to own the framework, not to approve individual decisions. Board ownership means three things: confirming the policy is comprehensive and current; demanding quarterly metrics that evidence the framework is operating; and signing off annually on the policy and the evidence.

The policy must define the three-limb test in the firm's specific context, name the responsible compliance officer, specify the response protocol for each category of information source (deal-team diligence, committee seats, market soundings, expert calls, alternative data), and establish the restriction and barrier standards. A policy that delegates all decisions to compliance without specifying the expected response per scenario is not sufficient.

Quarterly MI to the board should show the number of MNPI events logged, the number of restrictions imposed and lifted, the number of PA dealing pre-clearances with the denial rate, and any events where the response was longer than one business day with reasons. An MNPI function that produces no denials, no restrictions, and no logged events is not operating. It is a policy without a practice.

The twelve-month implementation roadmap for a firm building the framework from scratch runs as follows. In the first month, map every information entry point: deal-team diligence, board seats, committee seats, market soundings, lender relationships, expert-network subscriptions, data feeds. For each, document which are likely to generate MNPI about listed names and who in the firm is the first point of receipt.

In the second month, draft the MNPI policy defining the Art 7 test in the firm's context, the response protocol, and the evidence standards for each scenario. In the second and third months, build the infrastructure: the restricted and watch list system with timestamps, the pre-clearance log, the trading-defence memo template, and the email or system surveillance capability.

In the third month, run training for deal teams, fund administrators, and trading staff on the identification protocol and the response steps. In months four to six, test the protocol on two historical MNPI events, pulling the artefacts for each. In month six, present the policy, the infrastructure, and the test results to the board for formal sign-off. From month nine onward, run quarterly spot-checks, produce quarterly board MI, and run the annual policy review.

Start here

Book a conversation and we'll help you apply it to your firm's specific situation.