That's because changes to state constitutions typically require voter approval in a referendum to be enacted. Although support for limiting the use of eminent domain for private economic development spans the ideological spectrum (in spite of opposition from many municipal officials), libertarian-minded conservatives feel far more passionately about the issue than anyone else. As a result, the eminent domain referendums may be an impetus for conservatives to show up in droves at the polls in 2006.
I've been able to find the idea of eminent domain constitutional amendments mentioned in 15 states, although many of these aren't likely to have anything on the ballot this year. A proposed amendment will definitely be on Michigan's ballot and other states that seem likely to have referendums include California, Missouri, Oklahoma, Georgia and North Dakota. Property rights groups are also gathering signatures in Oregon for a ballot measure that would change state statues, rather than the constitution.
Many of these states will host high-profile elections in November. Democratic governors Jennifer Granholm of Michigan, Ted Kulongoski of Oregon and Brad Henry of Oklahoma are all potentially vulnerable and, in tight races, could be defeated by the extra conservatives the referendums draw to the polls.
Perhaps the biggest potential impact is in California, where some conservatives have become dissatisfied with Governor Schwarzenegger. Without the extra incentive of an eminent domain referendum, they might stay home, but if they're showing up anyway to vote on the referendum they would probably support Arnold over the Democrat.
This certainly isn't the first time anyone has suggested a connection between ballot measures and turnout. Just two years ago, many observers expected votes on gay marriage bans to bring large numbers of conservatives to the polls. Just from my cursory analysis, it doesn't seem as though those expectations were met. Presidential turnout among the voting-age population increased by 5.5 percentage points from 2000 to 2004 in states with gay marriage votes in 2004, but that hardly differed from the 5.4 percentage point increase nationally.
However, the case that referendums could affect turnout this time around seems far more plausible. In 2004, with a presidential election between one candidate who supported a constitutional ban on gay marriage and one who opposed it, anyone who cared passionately about the issue was probably voting anyway. The turnout dynamics are entirely different for midterm elections, when many registered voters will only show up to support a candidate or issue they find particularly compelling. For many people, an eminent domain referendum might do the trick.
