How long does Congress have to disclose a stock trade?

The STOCK Act gives members 45 days from the transaction to file a Periodic Transaction Report for any trade over $1,000. In practice the filings spread across that whole window and beyond. In our sample of House filings from 2021 through May 2026, the median gap between the trade and the public filing was 28 days and the mean was almost 48 days, which means a meaningful share land after the 45-day deadline has already passed.

What percentage of congressional trades are filed quickly?

Fewer than most people assume. Only 11.9% of the trades we measured were filed within 10 days of the transaction. Roughly a third landed inside two weeks. The image of a fast-moving copy trader reacting to fresh congressional information describes about one trade in eight, not the typical case. Any strategy built on fast filers is a strategy that sits idle most of the time.

Does the disclosure delay destroy the edge in copying Congress?

This is the popular explanation and our data does not support it. We tested entering at the member's own trade date, a price no copier can ever actually get, and that version still failed to beat the S&P 500. It came in about 0.36% behind over 90 days. If removing the delay entirely does not produce an edge, the delay was not what was standing in the way. The picks themselves were not beating the index.

Do trades filed faster perform better?

They rank better, yes. Trades filed within 7 days beat the index by about 0.71% on average, while trades filed after 45 days trailed it by 2.92%. That ordering is consistent and it is the strongest pattern we found. But the fast group's advantage has a confidence interval running from negative 0.82% to positive 2.24%, so we cannot call it real, and it still only beat the index 49% of the time.

What is the penalty for filing a congressional stock trade late?

The standard late-filing fee is $200, which is not a serious deterrent relative to the size of many disclosed positions. That weak penalty is the most likely reason the filing distribution is spread so wide instead of clustering before the deadline. For anyone trying to trade on this data, the practical takeaway is that you cannot count on timely filings, because the system does not meaningfully require them.

Is there a better way to use congressional trading data?

We think it is more useful as accountability information than as a trading signal. Knowing which committees a member sits on and what they bought is genuinely worth public attention. Using it to pick stocks is where the evidence runs out. If you want market exposure, the boring index approach beat the copy strategy in our testing without requiring you to monitor filings at all.

Congress Stock Trade Disclosure Lag: The Real Numbers

The congress stock trade disclosure lag runs 28 days at the median. We measured what that delay costs a copy trader.

There is a rule of thumb in the copy-Congress world that sounds airtight. The politician’s headline return is measured from the day they traded. You can only buy from the day they disclose. So the gap between those two dates is the tax you pay, and the whole game is finding members who file fast.

We believed this too. Then we measured it across 8,329 House stock purchases filed between January 2021 and May 2026, and the tax turned out to be almost nothing. Which sounds like good news until you work out what it implies.

How long is the congress stock trade disclosure lag?

The median is 28 days. The mean is 47.8 days, dragged up by a long tail of very late filings. Only 11.9% of trades reach the public inside 10 days.

The STOCK Act allows 45 days, and because the penalty for missing that deadline is a $200 fee, the filings sprawl across the entire window and well past it rather than bunching up before the cutoff.

Two dates, and most analysis uses the wrong one

Before any of this can be measured properly there’s a detail that quietly wrecks a lot of published numbers.

Open an actual Periodic Transaction Report and you’ll find two dates next to every trade. One is the transaction date. The other is labeled the notification date, which is when the member was notified the trade occurred, often by whoever manages the account.

The notification date is not public. It sits inside a document that has not been filed yet. The moment the information actually reaches you is when the report lands with the Clerk of the House, and that is a third date, recorded in the Clerk’s index rather than on the form itself.

If you measure lag from the trade date to the notification date, you get a much prettier number and a much faster-looking Congress. If you buy pre-cleaned data from an aggregator, you often cannot tell which one you’re getting. We used the Clerk’s filing date throughout, because that’s the only one a copy trader could ever act on.

What the delay actually costs

Here’s the test that surprised us.

We ran every trade twice. Once entering at the first available price after the public filing, which is what you could really do. Once entering at the member’s own trade date, which is a fantasy, a time machine, the best case that could ever exist.

	enter at trade date	enter at filing date	cost of the delay
90-day hold	+4.42%	+4.07%	0.35 points
30-day hold	+1.34%	+1.54%	−0.20 points

Over 90 days the delay cost about a third of a percentage point. Over 30 days it had a negative cost, meaning the filing-date entry did slightly better than the trade-date entry.

When the sign of an effect flips depending on how long you hold, you are not looking at an effect. You are looking at noise.

The implication nobody likes

If the delay only costs a third of a point, then removing the delay should hand you the politician’s full advantage. So we checked what that advantage was.

Entering at the member’s own trade date, the trades underperformed the S&P 500 by about 0.36% over 90 days, with a confidence interval that comfortably includes zero.

There is no advantage sitting behind the delay. With a time machine and perfect execution at their exact entry price, you would have done no better than owning the index.

This reframes the entire conversation. The disclosure lag has been the standard explanation for why copying does not work as well as the headlines suggest, and it is a comfortable explanation because it implies the edge is real and merely out of reach. Our data says the edge was not there to begin with. The delay has been taking the blame for a shortfall it did not cause.

Fast filers still rank better, and it still doesn’t help

None of this means the lag is meaningless. Sorted into buckets, the trades line up exactly the way the theory predicts:

filed within	trades	vs S&P 500	beat the index
7 days	724	+0.71%	49.0%
8 to 14 days	877	−0.05%	48.7%
15 to 30 days	3,297	−0.22%	45.9%
31 to 45 days	2,099	−0.86%	44.3%
more than 45 days	1,332	−2.92%	39.6%

That is a clean gradient and we would not dismiss it. Late filings are genuinely worse, and badly so.

But look at the top row before building anything on it. The fast bucket’s edge is 0.71% with a confidence interval from negative 0.82% to positive 2.24%. It beat the index on 49% of trades, which is worse than flipping a coin. When the average is positive and the hit rate is below half, a few large winners are carrying the entire result, and there’s no way to know ahead of time which trades those will be.

Then there’s the volume problem. Restricting yourself to filings inside 10 days means acting on about 12% of what gets filed. You would spend most of the year waiting for a signal whose advantage we cannot statistically separate from zero.

What we take from this

The disclosure lag is real, it is longer than people think, and it is a legitimate transparency complaint. A 45-day deadline backed by a $200 fee is not a serious reporting requirement, and the spread of filing dates in our sample shows exactly what that produces.

As a trading input, though, it is a dead end, and not for the reason the popular story gives. We went in expecting to find a real edge decaying across the delay. We found no edge at either end of it.

We laid out the full test, including a portfolio result that fooled us before we caught it, in our backtest of copying congress stock trades. If you want to track the filings for their own sake, and there are good reasons to, our guide to following congress stock trades covers the free tools.

And if you want to compare any of this against the unglamorous alternative, our compound interest calculator will show you what the index does over the same period while you’re not watching filings at all.

Two limits on our numbers, stated plainly. This is House data only, since Senate reports live on a separate system. And about 7% of the buys had no usable price history because the company was delisted, acquired, or renamed, which removes disproportionately bad outcomes and makes our results slightly generous to copying.