Every so often a member of Congress announces they are moving their portfolio into a blind trust, and the coverage treats it as the end of the story. The conflicts are gone, the trades are invisible, nothing left to look at. If you actually read disclosure filings, the announcement is closer to the middle of the story. The politician still knows every position that went in, the filings keep telling you more than you would expect, and the blindness, such as it is, arrives slowly over years. The mechanics are worth walking through, because they change how you should read that person's trading history from the day the trust is created.
What a qualified blind trust actually is
The formal version is called a qualified blind trust, and it comes out of the ethics law passed in 1978 after Watergate. The idea is simple enough. The official hands assets to an independent trustee, usually a bank or an investment firm with no prior relationship to them, and gives up the right to know what the trustee buys and sells. The trustee manages the money, handles the taxes, and reports the overall value once in a while. Communication between the two is restricted to a narrow set of topics, mostly in writing, with copies filed to the ethics office. The official cannot suggest trades, cannot ask what is in the portfolio, and cannot get a hint through a friend or a spouse without breaking the arrangement.
The word qualified is doing a lot of work in that name. The trust document and the trustee both have to be approved in advance by the relevant ethics body, meaning the House or Senate ethics committee for members of Congress and the Office of Government Ethics for executive branch officials. This matters because politicians use the phrase blind trust loosely. A family trust run by a sibling, or an account a spouse manages, is not a qualified blind trust no matter what the press release says. If the ethics committee has not signed off on the trustee's independence and the trust language, the official gets none of the reporting exemptions, and you should treat their holdings as fully known to them.
Here is the part most people miss. A blind trust is only blind going forward. The official knows exactly what went into it on day one, because it was their portfolio. The trust becomes genuinely blind only as the trustee sells the original assets and replaces them with new ones the official never learns about. The rules acknowledge this directly. The original holdings keep appearing on the official's annual disclosure until the trustee has sold each one down below a small threshold, at which point the trustee notifies the official that the position is effectively gone. So a senator who moves a large single-stock position into a qualified blind trust may know, for years, that the trust is still substantially that stock. The conflict decays on the trustee's schedule instead of vanishing at signing.
Why almost nobody uses one
Given the political cover a blind trust provides, you might expect them to be common. They are rare. Out of 535 members of Congress, the number with an approved qualified blind trust at any given time has historically been a small handful. A few things explain that.
Cost is the obvious one. You need lawyers to draft a trust instrument the committee will accept, an institutional trustee willing to take the account and its fees, and months of back and forth before approval. For a member whose portfolio is mostly index funds, that is a lot of money and paperwork to solve a problem they arguably do not have, since broadly diversified funds are generally treated as low-conflict to begin with. Control is the second one. Handing your savings to a stranger you are legally forbidden to talk shop with is genuinely unpleasant, and members with family businesses, farms, or concentrated real estate often cannot transfer those assets in any way that makes sense. The third reason is the quiet one, which is that nothing requires any of this. Members are allowed to trade individual stocks as long as they file the required disclosures, so the trust is a voluntary constraint with real costs, adopted mostly by people who either have simple portfolios or a strong political need to be seen doing it. The incentives point away from adoption, and the adoption numbers look exactly like the incentives.
What the filings still tell you
Once assets go in, the trust shows up on the annual financial disclosure as a single line with a value range and an income range, and the underlying holdings stop being itemized once they qualify as blind. Transactions inside the trust are also exempt from the periodic transaction reports that normally surface a trade within weeks. That is the real information loss for anyone tracking these filings. The stream of dated, ticker-level trades goes dark for whatever lives inside the trust.
Several things remain visible, though. The transfer itself is disclosed, so you know what went in and when. The original assets stay identifiable on disclosures until they are sold down, so you can watch the blindness phase in rather than assuming it happened at signing. The trust's total value range keeps updating each year, which tells you roughly how the pool is doing. And everything outside the trust behaves exactly as before. Spouse accounts that were never transferred, retained assets, deferred compensation, stock options from a former employer, all of it keeps generating normal disclosures. In practice most officials who set one up move a slice of their wealth into it rather than the whole estate, so the reportable surface shrinks without disappearing.
How to treat the signal history
If you use congressional trading disclosures as one input among many, which I think is the sane way to use them, a blind trust changes your read in specific ways rather than ending it. My working checklist looks like this.
- Confirm the trust is actually qualified. Look for the qualified blind trust designation on the annual filing or the ethics committee approval. A trust described as blind in interviews but managed by a relative deserves zero benefit of the doubt.
- Mark the transfer date. Every trade before that date was made with full knowledge of the portfolio, so the member's historical record stays valid and scoreable. Do not retroactively excuse old trades because a trust exists now.
- Assume known holdings until proven sold. The original positions are still known to the member and still disclosed, so if a committee assignment overlaps with a large legacy position sitting in the trust, that overlap means as much as it did before the trust existed.
- Keep watching what is outside. Spouse trades, retained accounts, and any new periodic transaction reports tell you the member is still active, and trades made outside the trust carry the same signal quality as before.
- When the feed goes quiet, record it as missing data rather than a flat position. A member whose trades stop appearing after a transfer did not necessarily stop trading. You stopped being able to see it, and a model that treats silence as abstinence will drift.
The main failure mode I see is people treating a blind trust announcement as a binary switch, where the member is either trading on knowledge or fully blind. The instrument does not support either conclusion. It is a gradual handoff of knowledge with a paper trail on both ends, and the paper trail is the usable part. This is also why, when we ingest disclosure data into Blockcircle's political trading feeds, trust transfers get flagged as events rather than treated as the end of coverage. The pre-trust record still ranks the member, the legacy holdings still matter, and the outside accounts keep reporting.
If you take one thing from this, make it the timeline. A qualified blind trust starts out fully transparent to the person who created it and becomes blind only as fast as the trustee turns over the original book, which can take years and is partially observable through the filings the whole way. Read a trust transfer as a slow fade in the data rather than a hard stop, and score the politician accordingly.