Voting in the face of uncertainty

In the 112th US Senate, party-line voting is rampant. 78 members voted with their party affiliation more than 90% of the time.

Further, changing stance on an issue has considerable political consequences. Individual senators who have done so have faced charges of flip-flopping or politicking, even in situations where new evidence justifies such a change.

Junior senators are politically constrained to vote within party lines. These lines cannot change without exposing more senior members to political repercussions, and so do not. As a result, historical precedent shapes policy in perpetuity.

A senator is ethically obligated to vote in a manner that will bring the least harm upon his or her constituients. Since votes on many bills are politically constrained by the core issues they address, the only way to minimize harm — to even appear to do this — is to operate through side-channels. This is evidently done through favor trading, back-room deals, and adding pork to legislation.

Techniques to oppose legislation, from filibustering to the brinksmanship of recent weeks, have arisen in part due to the failure of the legislative process to produce bills which minimize harm without challenging precedent. In many cases, this is in fact an impossible task, as precedent may now be harmful.

A root cause, I believe, is the implied certainty of a yea or nay vote. This binary choice is the mother of all quantization errors, reducing the depth and nuance of most political issues to a single precedent-setting choice. By forcing every issue to become an if you’re not with us, you’re against us affair, the lack of voting granularity drives each party further from each other and from a sensible middle ground.

A solution presents itself: give each senator one hundred votes instead of one.

Under such a system, senators could meaningfully encode their uncertainty about a decision into the voting record as a proportion of votes cast for or against a bill. As information accumulates about an area of policy, those proportions can be adjusted by each senator gradually, providing a face-saving means to shift precedent over time without appearing inconsistent. Further, senators can forego the side-channel process of harm minimization in favor of directly expressing their estimates of harm in their voting proportions.

Or senators could just keep going all-in on every vote. That’s actually a good thing: it means that this idea can be adopted without removing any tools from incumbents’ political arsenals.

Nevertheless, changing the voting style of the Senate would be a dramatic move, and it is unrealistic to expect a direct campaign to be successful. It would be easier to prove the idea as beneficial to both politicans and their constituients by demonstrating it in the laboratories of democracy. If you find this idea interesting, petition your city council, school board, and other local bodies of governance to give it a try.


Steven Robertson makes things for people who make things for people. more...