The Bills Both Parties Backed

The standard account of Congress is that it is polarised to the point of paralysis. The roll call record complicates that story — not by refuting it, but by showing where it does and does not apply.

A useful and deliberately narrow definition: a vote is cross-party when a majority of Republicans and a majority of Democrats land on the same side. It does not require unanimity, near-unanimity, or that anything meaningful was at stake. It asks only whether both parties' centres of gravity pointed the same way.

Higher than expected, and less meaningful than it looks

The share of votes meeting this definition is generally larger than the polarisation narrative would predict. That finding is real, and it is also easy to over-read.

Most cross-party votes are on matters where there was never a genuine dispute: naming a federal building, authorising a routine reauthorisation, honouring a person or event, or advancing a technical fix. These are legislative housekeeping. They consume floor time, they produce recorded votes, and they say very little about whether the parties can cooperate on contested questions.

Suspension of the rules amplifies this. That procedure is reserved for measures leadership expects to pass overwhelmingly and requires a two-thirds majority, so it mechanically generates cross-party votes. A chamber that moves many bills this way will show a high bipartisanship rate largely as a function of its scheduling practice.

The selection problem

There is a deeper issue, and it is the most important caveat on this page.

Leadership controls which questions reach a vote. Bills that would divide a party frequently never get scheduled at all — the disagreement is resolved by inaction, before any roll call exists. The set of recorded votes is therefore filtered toward questions where the outcome was already secure.

This means the cross-party rate measures agreement among the questions leadership chose to put to a vote. It cannot measure agreement on the questions that never made it that far, and those are disproportionately the contested ones. The rate is a real fact about floor activity and a poor proxy for the health of cooperation in general.

Comparing the chambers

The two chambers produce different rates, and the difference is structural rather than a matter of temperament.

Senate procedure requires supermajority support to end debate on most legislation, which means measures reaching a final vote have usually already assembled cross-party backing — or have been narrowed until they could. House procedure allows a simple majority to move most business, so a unified majority party can pass legislation with no minority support at all.

Those rules push the two chambers toward different equilibria. Reading a gap between them as a difference in collegiality mistakes an institutional design for a cultural one.

What counts and what does not

Only roll calls where both parties registered a majority position are included. A vote where one party was largely absent, or where the question did not divide along party lines in a way that produced two comparable positions, is excluded rather than counted as agreement.

Independents are not part of the calculation. With three in Congress, they cannot form a majority position that means anything, and including them would distort rather than inform. Non-voting delegates are excluded throughout this section.

The examples above are drawn from recent qualifying votes and are illustrative, not a ranking — they are not the "most bipartisan" bills by any measure, simply the most recent ones that met the definition. Full rules are in the methodology.