Every quarter the same story appears. A member of Congress bought shares. Weeks later the company reported earnings. The stock jumped. The headline writes itself.
The problem is that the headline is built on two dates that do not mean what most readers assume. Understanding those two dates is the difference between a real signal and a coincidence dressed up as one. This article walks through the mechanics of disclosure timing around earnings season, what it can support as evidence, and where the honest limits are.
Two clocks, not one
A congressional trade produces two timestamps. They are separated by a gap that the law permits to be large.
The first is the transaction date. That is when the trade actually happened.
The second is the filing date. That is when the Periodic Transaction Report, usually called a PTR, was submitted and became public.
The STOCK Act of 2012 sets the rule. You can read the bill text on congress.gov. A member must file within 30 days of becoming aware of a transaction, and in no case later than 45 days after the transaction itself. The threshold is transactions above $1,000, and it covers the member, a spouse, and dependent children.
That 45-day ceiling is the whole story for earnings analysis. A typical company reports earnings roughly every 90 days. A disclosure window of up to 45 days swallows half a quarter. A trade made three weeks before an earnings release can legally surface three weeks after it.
So when you read that a member "bought before earnings," you are almost always reading a reconstruction. The public did not know at the time. The public learned later and drew a line backwards through two points.
What "before earnings" actually means as a claim
There is nothing inherently improper about trading before an earnings date. Members of Congress are not corporate insiders of the companies they hold. They are not subject to the issuer blackout policies that bind a company's own executives. There is no rule that says a senator must stop trading Apple stock in the two weeks before Apple reports.
An earnings date is also public information. Companies announce them in advance. Anyone can look them up for free. A trade timed around a known public date is not evidence of anything by itself.
The claim that would matter is narrower. It is that the member had material nonpublic information about that specific company, and traded on it. The STOCK Act made explicit that members owe a duty of trust and are not exempt from federal securities law, including the prohibition on insider trading.
Notice what disclosure timing can and cannot do for that claim.
It can establish sequence. Trade happened on this date. Event happened on that date. That is real and checkable.
It cannot establish knowledge. A PTR contains no reason, no rationale, and no source of information. It has a date, a ticker, a transaction type, and a dollar range.
Sequence without knowledge is a starting point for a question. It is not an answer.
The base rate problem
This is the part most coverage skips, and it is the part that changes conclusions.
Earnings season is not a rare window. Roughly four times a year, most of the S&P 500 reports within a compressed six-week span. Depending on how you define proximity, a large share of the calendar sits near somebody's earnings date.
Congressional trading volume is also not evenly spread. Filing activity clusters. Members with active portfolios trade in bursts, often around portfolio rebalancing, tax dates, or the end of a reporting period.
Put those together. If trades cluster and earnings cluster, then trades near earnings will occur frequently for reasons that have nothing to do with information. Before you treat proximity as meaningful, you need to know what proximity would look like under pure chance.
The practical test is straightforward. Build the distribution of days between transaction date and the nearest earnings date across all congressional trades. Then compare a specific member or a specific trade against that baseline. If the whole population sits at a median of, say, a few weeks from an earnings date, then a single trade sitting a few weeks out is unremarkable.
Any analysis that reports the interesting cases without reporting the base rate is selecting on the outcome. That is the most common analytical error in this entire field.
Where an information advantage would plausibly come from
If you are looking for a mechanism, earnings is a weak candidate. Members of Congress do not sit in on corporate earnings preparation. Regulation FD, which the SEC administers, restricts selective disclosure by issuers to outside parties. The SEC's own overview of Regulation FD sets out the framework. Companies that disclose material information selectively must make it public promptly.
The more plausible channels run through government activity, not company activity.
Committee work touches regulated industries before rules become public. Appropriations decisions move defense and infrastructure contractors. Health committee work sits close to drug and device policy. Briefings on macroeconomic or geopolitical conditions can precede market moves across whole sectors.
None of that is earnings information. It is policy information. And policy information often shows up in the market through a different door, such as a contract award, a rule publication, or a company's own Form 8-K. The SEC's Form 8-K guidance describes the material events a company must report between quarterly filings.
So if you want to hunt for information advantage in congressional trades, aligning trades against committee assignments and policy calendars is a stronger design than aligning them against earnings dates. Earnings is the event everyone can see. Policy timing is the event fewer people track.
Reading the fields you actually get
A PTR gives you less precision than the headlines suggest.
Amounts come in ranges, not exact figures. The lowest band starts at $1,001 and runs to $15,000. Larger bands follow. A trade at the top of a band and a trade at the bottom look identical in the data.
Owner codes matter. Many disclosed trades belong to a spouse or a dependent child, or sit inside an account the member does not direct. Attributing intent to the member for every filing under their name overstates what the document says.
Asset types vary. Some filings cover broad funds or bonds rather than single stocks. A diversified fund purchase carries no company-specific information at all.
Amendments exist. Filings get corrected. A trade you analyzed in March may carry a different date or amount after a later amendment.
Doing this yourself, for free
You do not need a paid product to check any of this.
The House publishes PTRs through the Clerk's portal at disclosures-clerk.house.gov. The Senate publishes through its electronic financial disclosure system at efdsearch.senate.gov. Both are free and require no account, though the Senate system asks you to accept an access agreement.
For the other side of the comparison, company filings are free on the SEC's EDGAR full-text search. Earnings dates and 8-K filings are public the moment they land. Investor relations pages publish scheduled reporting dates in advance.
Several open-source projects and volunteer-maintained datasets also mirror congressional disclosure data in machine-readable form. If your goal is a one-off check on a single member, the official portals are enough. If your goal is a full historical distribution, a mirrored dataset saves considerable parsing work.
An honest summary of what timing proves
Disclosure timing is good evidence of one thing. It tells you how stale a signal is by the time you see it. That is genuinely useful, and it is underrated.
Disclosure timing is weak evidence of information advantage. The gap between trade and filing means proximity to earnings is almost always discovered after the fact. Earnings dates are public. Trades cluster. Earnings cluster. Coincidence is cheap.
The strongest version of an information advantage argument needs more than two dates. It needs a plausible channel, a base rate to compare against, and consistency across many trades rather than one memorable example. Academic work on congressional returns has produced mixed results, and reasonable researchers disagree on whether an edge survives risk adjustment.
None of this is investment advice. Treat any pattern you find in disclosure data as a hypothesis you still have to test.
If you want the timing math done for you, our Congress Stock Trades report tracks new PTR filings, records both the transaction date and the disclosure date, and scores each trade with the filing lag included, so you can see how old a signal is before you act on it. Browse the current filings on the Congress Stock Trades report.
Want the signal instead of the raw filings? Get the free Congress Trades preview. Prefer the tool to the write-up? Browse all data feeds or connect the free MCP server.