I hate to break it to you, but you have a bunch of people in Congress right now gaining politically by being disruptive. So your failure scenario is occurring now without your mechanism.
If the failure mode of this proposal (excessive splitting of bills) is less disruptive than the current situation (attaching bill-killing, shutdown-threatening riders), then it seems to me we would still be better off with this implemented than without.
In addition, as long as the votes for this are public, it seems like it would be harder to defend capricious action for political gain, since anyone could see that this group of senators all voted to split legislation that clearly should not have been split and then take them to task for it. It could even be mandatory that a proposal to split a bill must be accompanied by an explanation (also publicly available) of why it should be split that way. It's one thing in my opinion to put something controversial into an unrelated bill--there will always be people who are in favor of that, so it is politically defensible and can score points--but it is another thing entirely to try to (literally) rationalize the frivolous division of a cohesive bill.
It actually strikes me as quite an elegant solution to the problem of riders and sprawling legislation.
You believe that by increasing transparency and granularity, voters will punish or reward Congresscritters more effectively and efficiently.
I disagree strongly. I submit that most voters do not follow the legislative process very closely and vote accordingly. Rather, I submit that most voters make decisions emotionally using far less than the totality of the relevant information currently available to them. At this very moment, we have a batch of Congresscritters who gain the support of their constituents by obstructing their opposition by any means possible. Questions of frivolousness or caprice are not considered. This is the situation here, today, and now.
I think your notion fails because it adds extra information that voters will disregard. Because this information will be disregarded, it will not significantly impact the behavior of voters. The net result is likely to be an increased legislative overhead, more procedural tools to be wielded as partisan weapons, and voter behavior not shifting significantly. As a result, the failure mode of this proposal is everything wrong with the current (attaching bill-killing, shutdown-threatening riders) PLUS excessive bill-splitting for the sake of obstruction.
Might I suggest that your solutions should not hinge on sudden and dramatic shifts of voter behavior at a scale of many millions?
