AbstractMembers of the U.S. House must disclose their stock trades within 45 days, and a whole genre of websites invites readers to copy them. We tested 8,941 stock purchases and sales disclosed by 99 House members, traded from November 14, 2024, measuring what each stock did over the following one, three and six months against the S&P 500 and against the equal-weighted S&P 500, in which every company counts the same. Measured from the day a trade became public, the median purchase did −2.4% against the equal-weighted index over six months (90% range −3.9% to +0.1%), and the median stock members sold did −1.7%: what they bought did no better than what they sold.
Measured from the members' own trade dates, purchases did −1.5%. No group of purchases (by party, size, account or disclosure delay) did clearly better than the index.Keywords: Congress, STOCK Act, Periodic Transaction Reports, event study, copy trading1. IntroductionThe STOCK Act of 2012 requires members of Congress to report their stock trades in Periodic Transaction Reports, filed with the House Clerk within 45 days. Because members vote on laws and hear briefings that move markets, many people assume their trades carry an edge, and websites rank members by their trading and alert subscribers to new filings.
This paper asks the question a reader of those sites faces: had you bought each stock a House member reported buying, as soon as the report was public, would you have beaten the market? It is a companion to Arkanum's tests of company insiders' purchases and sales, and uses the same methods where they apply.2. DataThe trades come from the House Clerk's Periodic Transaction Reports, as collected for Arkanum's Congress pages (file updated November 14, 2026). We kept stock purchases and sales with a ticker and a clear trade date, one per line of a report: 9,188 trades.
Of these, 177 had no usable price history and 70 had a ticker that now belongs to a later listing, leaving 8,941 trades by 99 members: 4,142 purchases and 4,799 sales. The Senate is not covered: its disclosure site blocks automated access.Disclosure is slow. The median trade was disclosed 27 days after it was made, one in ten took 56 days or more, and 12% were disclosed after the 45-day deadline.3. MethodTwo starting points.
For each trade we measure the stock from the close on the trade date (what the member got) and from the first close after the report was filed (what a reader could get). Returns run one, three and six months (21, 63 and 126 trading days), using daily closes adjusted for splits and dividends.Two benchmarks. Each return is compared with the S&P 500 over the same days, and with the equal-weighted S&P 500 (RSP). Over this period a handful of very large companies drove the ordinary index, so most individual stocks trailed it; the equal-weighted index removes that, and is the fairer test of stock picking.
We report the equal-weighted results first.Purchases against sales. If members knew something, the stocks they bought should do better than the stocks they sold. Comparing the two is a test that doesn't depend on the benchmark at all.Statistics. We report medians, since a few large moves dominate averages, and the share of trades that beat the benchmark.
Trades disclosed in the same month share the same market, so they are not independent: each six-month median carries a 90% range from a bootstrap that resamples whole disclosure months (400 resamples), as in the insider papers. Report amounts are given only as ranges ($1,001–15,000 and up), so each trade counts once, whatever its size; a dollar-weighted view is in the appendix table.Not investment advice.
Texto de arkanumdata.com tal como lo publicó, leído de su página y mostrado aquí con el formato de El Autómata. El original, en arkanumdata.com · Volver a la noticia de El Autómata