In response to this post and this post, Matthew Shugart emails:

The whole idea of trying to prevent this sort of packaging troubles me. I fully understand the logic of the “single-subject” requirement, like not tying the sale of state property to the regulation of the primary-election process, for example. But, first of all, are term limits and redistricting really “separate subjects”? Both concern how we elect legislators. Term limits advocates claimed term limits would increase competitiveness, and proponents of non-partisan redistricting claim the same. The former has been shown via academic research largely to have failed, though in the presence of districts drawn by nonpartisans, it might very well work out as intended. This question would seem to be separable from the specific length of the term limits, which might have to be adjusted (as a matter of political practicality) in order to bring about the overall public good of more competitiveness. Now I can see that taking back one component (i.e. lengthening term limits) to get the other is a logroll, and the purpose of the single-subject requirement is essentially to prevent logrolls.
There just is a real problem with that logic. If you can’t link A and B, often you get neither. You just get the status quo. I won’t give you A if you won’t give me B.
I have the same trouble with item vetoes. I think–though I am unaware of empirical confirmation–that item vetoes do not really make the executive more powerful (as is often assumed). I think they simply make the status quo safer.
I am no fan of direct democracy, but please tell me we have not so restrained it judicially that the status quo trumps any chance politically for a potentially rather large majority to approve a package of reforms to the election process.

John Gear writes:

Michigan is suffering terribly from the twin scourges of expert partisan redistricting and stiff term limits. Because the first is so bad, the reaction was the second — so now we have the worst of all worlds … the winners of virtually all races are decided in primaries where, 9 times out of 10, the most extreme candidate wins.
(The exception proves the rule: pro-choice GOP Rep. Joe Schwartz, new member of Congress, “won” his primary with 28% of the vote, with all six “Stop Joe” opponents foolishly splitting the vote rather than deferring to one another. A perfect case for the use of instant runoff voting, but an example of how dysfunctional the system is … the winner was opposed by 72% of his own party but, thanks to bulletproof gerrymandering, was unstoppable once nominated.)
The state overall leans D but the US House delegation is 9-6 R-D, and both state houses are R dominated, tribute to the skill of the gerrymander artists.
However, for just one example of the bizarre consequences, the Speaker of the House is a 34 year old with two years legislative experience. The folks in Lansing are chafing under term limits, but they are finding that they are still insanely popular (with the Rs finding that the monster they created was easier to unleash when the Ds had long been in control than it is to get back under control).
My thinking has been that there is only one subject here: competitive elections. Term limits were a ham-handed and wrongheaded approach — but they were devised in response to a very real problem (totally uncompetitive elections that verge on fraudulent).
I would (and am) arguing that the only way to get people to agree on backing off from term limits is to ensure that we don’t get a return to the status quo of soviet-style elections with 99% incumbent reelection rates, and that is only possible by taking control of redistricting away from the legislature. They may not be one subject in a law school text but they are absolutely a single subject (competitiveness in elections) in the real world.

Thanks for writing!