One element that has not been given enough attention is the unwisdom of these opposed recall elections. Jonah Goldberg goes after them, but I think he is too sweeping in his condemnation.
There are recalls and then there are recalls.
Michigan has the proper form of recall – a pure yes/no referendum on the office holder. This is what recalls really should be for: to address malfeasance or a failure to live up to one’s promises.
Because it is so narrow, it is difficult to electioneer. The ballot question reads simply: Shall the candidate be removed? The election is wholly about their suitability for office. Only if the candidate is removed do we proceed to nominating potential replacements.
This does two things:
1. It makes it harder to have a “candidate in waiting” ala California or Wisconsin. If you want to recall someone, the argument cannot be that someone nominated later will do better because it is simply too far off. The focus is entirely on the incumbent.
2. It slows down the process, making it harder to “flip” control of either the Legislature or elective office in one act.
This latter point is key. In Michigan, recalling the governor would effectively elevate the lieutenant governor. Recalling both would simply elevate the Secretary of State. This removal would then set in motion a new primary election (taking at least three months) and then a general election (three more months). This means that while a narrowly divided Legislature can “flip” through a single recall election, the executive power is far more insulated.
The practical effect is that the recall in Michigan is used to punish individuals rather than parties. An officeholder who ignores constituents or embarrasses the district may well face a recall and the result may change the composition of the chamber, but this is far from guaranteed since the losing party may simply nominate a better candidate and win the seat back.
This is what happened in the 51st House District. Republican Paul Scott was recalled but the GOP nominated Joseph Graves, rendering the recall moot.
Where recalls can be decisive is at the local level and of course where there a tie or very narrow majority in one of the chambers. Here the removal from office could change the control of the chamber. For example, in a 55-55 tie in the House (which happened in the 1990s) the loss of a seat to recall would immediately break the tie and give one party (temporarily, at least) control.