Does insider convergence add anything?

Honesty Does insider convergence add anything?

When a member of Congress buys a stock and corporate insiders at that same company were also buying it around the same time, does that trade go on to do any better than a congressional buy without that agreement? We test it the same honest way we test everything else: the price move relative to the S&P 500 over the 63 trading days after the filing date — look-ahead-free, measured only from what a follower could actually have acted on. And crucially, convergence is judged ex-ante: an insider trade only counts if it was public before the congressional filing.

Hypothetical backtest, not investment advice. Returns are measured against historical prices over a fixed window. This evaluates whether convergence carries information — it is not a prediction or a recommendation, and past patterns don't predict future results. Insider convergence is context, never a buy/sell signal.

No measurable edge. Convergence buys (-0.05%) did not reliably beat comparable buys without convergence (-0.28%) — a difference of +0.23 pp, within the noise. That's the honest result: insider convergence is interesting context, not evidence of a return edge. We show it as context only.

Confirmed vs unconfirmed congressional buys

Both groups are buys on tickers where corporate insiders were active, so this is a fair like-for-like — confirmed vs not, not insider-covered vs not. Each figure is the average 63-day excess return over the S&P 500 from the filing date, with its 95% confidence interval.

Group	Avg excess return vs S&P 500	95% confidence interval	Significant?	Trades
With strong/moderate insider convergence	-0.05%	[-1.1, +1.0]	not sig.	881
No insider convergence	-0.28%	[-0.7, +0.2]	not sig.	4670

Method: every congressional BUY on an insider-covered ticker is measured 63 trading days out from its filing date (look-ahead-free), as its return minus the S&P 500's over the same window. A buy is "confirmed" when, at the time of its filing, corporate insiders had already made open-market purchases in the same direction strong enough to grade as moderate or strong (buy/sell asymmetry and 10b5-1 rules per our methodology). The 95% confidence interval uses the t-distribution on each group's sample; "significant" means the interval excludes zero. This is a hypothetical backtest on public data, not actual performance.