Some members of Congress beat the market and most do not - and which answer a study reaches depends almost entirely on how the comparison is measured. The only fair test benchmarks each trade against the S&P 500 over the same entry and exit dates, rather than comparing average per-trade returns to a full-period index return. Pre-STOCK Act research found Senate portfolios beat the market by roughly 12% annually; post-2012 studies find smaller or no average excess returns, with enormous member-to-member variance.

Do Congress Members Beat the Market? What a Fair Test Shows

Some members of Congress beat the market and most do not - and which answer a study or headline reaches depends almost entirely on how the comparison is measured. The only fair test benchmarks each trade against the S&P 500 over the same entry and exit dates, rather than comparing average per-trade returns to a full-period index return. That is how HoldingsIntel scores every tracked member, trade by trade.

Why Most Claims About Congress Returns Mislead

“Congress crushed the market” and “Congress underperforms” headlines regularly coexist because the underlying comparisons break in predictable ways:

  • Mismatched windows - the classic error: averaging per-trade returns and comparing the average to the index's return over a full year. A 60-day trade and a 12-month benchmark measure different things; the comparison is apples-to-oranges by construction.
  • Headline selection - coverage concentrates on a handful of famous portfolios. Whatever those members did becomes “what Congress did” in the public imagination, while hundreds of members who trade little or hold index funds go uncounted.
  • Amount ranges - STOCK Act disclosures report ranges ($15,001-$50,000), not exact sizes, so dollar-weighted returns can't be computed precisely. Anyone quoting one made assumptions they rarely state.
  • Disclosure lag - trades surface up to 45 days late (later when members file late), so a copier's achievable return and the member's actual return are different numbers. Some trackers quietly report the former as the latter.
  • Open positions - a member who hasn't sold has no realized return, so every method must choose how to mark open positions - and different choices produce different league tables from identical data.

The Fair Test: Per-Trade Matched Benchmarking

The fix for the window problem is to make the benchmark live through exactly the same dates as the trade. For every disclosed purchase: compute the stock's return over the member's holding window, compute SPY's return over those same entry and exit dates, and take the difference. Every trade is judged against what a plain index fund would have done with the same dollars over the same days - no timeframe mixing, no flattering denominators.

Two more rules keep the measurement clean:

  • Score buys and sells separately - buy performance is the return question; sell quality is a timing question, scored by whether the stock declined after the exit. Blending them muddies both.
  • Report win rate next to average return - one lottery-ticket winner can mask a losing process. The share of trades that beat their matched benchmark tells you whether the average is skill or a fluke.

This is the methodology behind every performance figure on HoldingsIntel's congressional pages. It is deliberately conservative: it will make a famous portfolio look ordinary when the famous trades merely rode a rising market.

What the Research Actually Says

The academic record splits around the STOCK Act. Ziobrowski et al. (2004) studied Senate portfolios from 1993-1998 and found they beat the market by roughly 12% annually - the study that launched the entire genre. Research covering the post-2012 era, after mandatory 45-day disclosure took effect, generally finds smaller or no excess returns on average.

The consistent finding across both eras: the average conceals enormous member-to-member variance. A few members show persistent, benchmark-beating timing; most look like the market; some reliably lag it. “Does Congress beat the market” is the wrong question - which members do, measured fairly, is the one with an actionable answer.

Check the Live Numbers, Not a Screenshot

Any article quoting a member's return froze that number the day it was written; the next disclosure makes it stale. This page deliberately contains no per-member figures for exactly that reason. The congressional trading rankings show each tracked member's buy returns against the matched SPY benchmark, win rate, and sell timing - recomputed as disclosures arrive.

Performance is also only half the read. Each member's trades are cross-referenced with 13F filings from 551+ institutional funds (data through Q1 2026) in the Smart Money Convergence view - a member buying what hedge funds are also accumulating is a different read from a member trading alone. For how the disclosure system itself works, see the congress trade tracker guide.

Related Terms

For the legal mechanics behind these disclosures, see the STOCK Act guide. Key terms:

Frequently Asked Questions

Do members of Congress beat the market?

Some do, most don't - and the average answer depends on the study period. Research from before the STOCK Act (Ziobrowski et al., 2004) found Senate portfolios beat the market by roughly 12% annually in the 1990s; studies covering the post-2012 disclosure era find smaller or no excess returns on average. The average hides enormous member-to-member variance, which is why per-member measurement matters more than the headline.

Why is comparing a trade's return to the market's annual return misleading?

Because the windows don't match. A trade held for 60 days earning 8% looks weak next to a 24% market year - but the market may have returned only 3% over those same 60 days, making the trade strongly market-beating. Any comparison of per-trade average returns against a full-period benchmark return mixes timeframes and produces a number that means nothing.

What is a per-trade matched SPY benchmark?

For each disclosed purchase, you compute the stock's return over the member's actual holding window and SPY's return over the exact same entry and exit dates. The difference is that trade's excess return. Aggregating those per-trade comparisons gives a member-level performance figure where every trade is judged against what an index fund would have done with the same dollars over the same days.

Why are sells scored separately from buys?

A sell's quality is what it avoided, not what it earned. Mixing sells into a buy-return average double-counts noise, so sell timing is scored on its own terms: the share of a member's sales where the stock declined afterward. Good sell timing means exiting before drawdowns; the buy-side return question stays clean.

Where can I see a specific member's real performance?

HoldingsIntel's congressional pages compute each tracked member's buy returns against a per-trade matched SPY benchmark, plus win rate and sell timing, and recompute as new disclosures arrive. Each member's trades are also cross-referenced with institutional 13F filings from 551+ funds, so you can see whether the smart money agrees with the politician.

Analytics tracking helps us improve the product. Essential functions stay on either way. How analytics data is used