I agree that there are cautionary lessons to be learned from the Detroit fiasco; but to my mind they're less about the particulars of, say, the Kilpatrick debt issue boondoggle, and more about the problem of incentives for elected officials. If you're looking at an average career in public service, you're going to be making decisions that have a time horizon of decades, but your electoral prospects are determined by the short term pain your constituents face, and so kicking the can down the road is entirely a sensible decision. But these sensible decisions compound, and that ends up causing enormous knock-on effects.
> instead of term limits, make it you have to win each election with 5% more of the electorate?
But that would be an exponential increase, an increase without bound that would sweep past 100% as if it was standing still.
If a candidate won a substantial majority in one election, that rule might decrease his chances to be re-elected because the stakes would be raised in the following contest. So being popular would count as a handicap.
> He would have to be head-and-shoulder above the competition, yes, but he could run.
More to the point, he would have to be 5% more popular each time he ran. That's not likely. And eventually he would have to win more than 100% of the vote.
I don't know of a way around this.