If you're a member of the party establishment in one of the big two parties of a system, you stand to lose a lot by moving to an instant runoff and nothing to gain.
Unfortunately, most of the power to change the voting system is in the hands of these kinds of people, so it's going to be pretty hard to change it. Kind of a Catch-22.
I understand that things get a bit muddled because the individual representatives are also members of the electorate, and thus they may have warped incentives relative to the vast majority. But they are such a small fraction of the overall electorate that we should be able to ignore them in any kind of efficiency discussion.
I mentioned it because you said something about Pareto efficiency, and moving to instant runoff would not be Pareto efficient because some people (the parties are made up of people, remember) would be worse off.
Declaring something Pareto efficient has a specific meaning, and it applies to all people, not all the people that stanleydrew thinks are important. Though it's a comparatively small number of people, it's not some weird edge case that can be discounted. Rather, these are precisely the best people to convince to get the law changed. Note well that even this story in Maine, the measure was passed by ballot initiative, not by a legislature vote.
I am talking about the efficiency of the voting system itself, and its ability to express the preferences of the electorate.
Note that I'm not talking about changes related to the election outcomes that such a shift in voting systems would potentially cause.
Put another way, it would seem to be efficient to move to any voting system that would allow a voter to express more preferences than just one, since then we're getting strictly more information.
I am talking about the efficiency of the voting system itself, and its ability to express the preferences of the electorate.
Note that I'm not talking about changes related to the election outcomes that such a shift in voting systems would potentially cause.