Politici bij naam noemen in klantmateriaal zonder regelgevingsrisico

The research is not the problem. Legislative transaction disclosures are public records published under statutory regimes, and building a systematic strategy on them raises no more marketing issue than building one on filings of any other kind. The problem is the sentence somebody writes on the factsheet, because that sentence is a communication offering advisory services, and it is regulated on entirely different grounds from the data underneath it.

This gets missed because the two live in different parts of the building. The quant is thinking about signal construction. The compliance question arrives six weeks later, attached to a pitch deck, and by then a named-member framing has been agreed with three people who liked it. What follows is the shape of the constraint and the language that generally clears it. It is not legal advice, and the specific rule text that binds you depends on your registration and jurisdiction, so the mapping is a conversation with your own counsel rather than something to take from a blog post.

The regulated act is the sentence, not the trade

Adviser marketing rules across the major regimes converge on a handful of concerns, and a named-member claim manages to touch three of them at once.

The first is endorsement and testimonial treatment. A statement built around a named public figure whose trades you follow can read as implying association, and endorsement provisions typically impose disclosure, oversight and in some cases compensation-related requirements when a third party is presented as connected to the adviser. The awkward part is that the named person has not endorsed you, cannot be asked, and has no relationship with the firm at all, so the disclosures those provisions contemplate cannot honestly be made.

The second is performance advertising. If the claim implies your results derive from a named person's results, you have put a third party's track record in your marketing, with all the presentation requirements that attach to performance, and with no ability to compute it on the required basis because you do not have that person's account.

The third is the general prohibition on materially misleading statements, which is the one that actually bites. Any framing implying you replicate a named individual's portfolio is misleading on the facts, and the reason why is visible in the product itself.

What the leaderboard ranks and why it will not support the claim

Open the leaderboard tab and read the column headers before the names. The sort control was set to Total Trades at capture, the roster showed 7,083 politicians, and the filter chips offer House, Senate, and in office against former. The columns are Trades, Alpha 30d, Win Rate and Compliance.

Now read the top three rows as a compliance officer would. The first is labelled Federal Judge under Judiciary, with 49,087 trades, a Compliance score of 100, and dashes in both Alpha 30d and Win Rate. The third row is another judiciary entry with 7,845 trades and the same two blanks. The only sitting legislator in the visible top three carries a 30 day alpha of minus 1.1 percent and a win rate of 44 percent.

Three things follow, and each of them kills a different version of the marketing line. The roster is not exclusively legislators, so "top congressional trader" misdescribes what is being ranked. The ranking is by activity, not performance, so "top ranked" carries no return meaning whatsoever. And where a performance figure does exist it is a 30 day window on a single filer, which is not a track record by any standard your reviewer will accept.

None of this is a defect in the module. It is a filing roster with an activity sort and some derived columns, which is the correct design for a research tool. It is simply not a performance table, and marketing copy that treats it as one is making a claim the underlying data cannot support.

Three claim shapes and what each one triggers

In practice the drafts that come back marked up fall into three families, and it helps to name them so the conversation is about structure rather than word choice.

• Association claims. Any phrasing that puts the firm and a named individual in the same sentence as collaborators or as follower and followed. This is the endorsement exposure, and it is the hardest to remediate because the disclosures the rules expect cannot be truthfully provided.
• Derived performance claims. Any phrasing that attributes returns, skill or edge to a named third party and then connects it to the product. This imports another party's performance into your advertising, and the presentation requirements that follow are ones you cannot satisfy without their statements.
• Implied replication claims. Any phrasing suggesting the portfolio mirrors a named person's holdings. This one is misleading on the facts as well as risky in form, because disclosures report dollar bands rather than amounts, so nobody knows the filer's actual weights and no product can replicate them.

Notice that the third family fails even if you resolved the first two. That is the useful insight for the strategy team, because it means the compliant description is not a watered-down version of the true one. It is the more accurate description.

Language that survives a marketing review

The pattern that works is to describe the data source and the process rather than any person, and to keep every quantitative claim about the strategy's own realised results.

Three footnote obligations travel with the compliant column. State the disclosure lag explicitly, since the dashboard reported an average of 32.5 days between transaction and filing and a reader is entitled to know the input is not contemporaneous. State that disclosures report value ranges rather than amounts, so weights are the strategy's own construction. And state that inclusion of any filer's disclosed activity is not an assertion about that person's conduct, which protects them as much as it protects you.

Naming is a data problem before it is a compliance one

Even setting the rules aside, the operational case against person-level marketing is strong, and it is the argument that tends to actually land with a portfolio manager who wants the name in the deck.

Entity resolution is unreliable at the individual level. The top trader tile and the leaderboard row directly beneath it render the same person's name in two different forms, one full and one short, which is ordinary for filing data assembled from multiple official sources and is precisely the kind of thing you do not want underneath a claim printed on a factsheet.

Sample size is the second problem. A single filer produces too few observations for anything you would call a track record, which is why the alpha and win rate columns are blank on two of the three top rows. Cohort level aggregation across a roster of thousands is the statistically defensible unit and it happens to be the compliant one too.

Then there is the compliance column itself, which measures punctuality of disclosure and nothing else. It is a data quality field. Late filing has mundane causes including administrative error, adviser reporting delays and account transfers, and a punctuality score is not a finding about a person. Treat it as an input to how much you trust a timestamp, never as a characterisation, and keep it out of client-facing material entirely.

The operational rule I would write into the research protocol is short. Individual names may appear in the research log, the data warehouse and the internal review record, where they are necessary and where they are read by people who understand what they are. They do not appear in any client-facing artifact. The strategy is described by its inputs and its rules, and the performance shown is its own.