Party Registration is Not the Same as Party Identification
At this time of year, many reporters, pundits, and other commentators talk about the importance of the independent vote in New Jersey by stating something along the lines of: “At 46% of all registered New Jersey voters, unaffiliated voters outnumber both Democrats and Republicans.” And then go on to describe how the “independent vote” breaks down in the polls, as if these two groups are the same.
How I can put this? Um. They are not.
I’ve written about this before, but it obviously bears repeating. When polls refer to party identification, they are talking about how voters consider themselves politically (e.g. “Regardless of how you will vote, do you usually think of yourself as…”). In many states, that aligns pretty well with how voters are registered. But not so much in New Jersey.
[A Suffolk University poll out this morning appears to have weighted party ID to party registration, a common mistake by pollsters unfamiliar with the New Jersey electorate.]
Being “unaffiliated” in one’s registration is not the same as being “independent” in one’s thinking. We consistently find that at least 1-in-5 unaffiliated New Jersey voters actually see themselves as partisan.
This is a byproduct of New Jersey’s semi-open primary system. Why bother registering with a party if you can wait until primary day and do it on the spot? And why bother to vote in primaries if they are rarely competitive? So, New Jersey ends up with a lot of “party-line” voters who never bother to register with their preferred party. They just see no need.
This is the major reason why the unaffiliated proportion on the voter registration rolls suddenly plummeted from 58% to 45% in just one day. When was that? The presidential primary in February 2008.
That unique event brought many covert partisans “out of the closet” (as I described here). I have yet to meet a voter who claims they were independent before the primary, only to wake up on February 5th with a sudden epiphany that they were Democrat or Republican in their political views.
The Unaffiliated Vote is Not as Sizable as it Appears
Even if most unaffiliated voters are actually independent in their thinking, there is another reason why focusing on the large size of “unaffiliateds” on the voter rolls is misleading. Most of them don’t vote! Or if they do, it’s only in presidential races.
Last year, unaffiliated voters made up 38% of the electorate even though they comprised 47% of registered voters. In other words, while more than 8-in-10 registered Republican and Democratic voters showed up last November, only 6-in-10 unaffiliated voters turned out.
This disparity is even larger in non-presidential years (i.e. like this year). In the 2006 election for U.S. Senate, about 7-in-10 registered partisans showed up, but only 1-in-3 unaffiliateds did. And that was when unaffiliated voters made up 58% of the voter rolls. My guess is that many of those folks probably voted in the 2008 presidential primary and are now registered with a party. Therefore, I wouldn’t be surprised if the unaffiliated turnout is even lower this year.
All things considered, the unaffiliated vote is important in an election that is as close as the current governor’s race appears to be. However, that has less to do with absolute size of their vote share and more to do with the volatility of their vote choice.
The Democratic Advantage
No matter how you slice it, Democrats have a natural advantage in recent New Jersey elections. In 2006, they had a 6 point advantage over Republicans in terms of party registration, but a 13 point advantage in party identification on election day based on that year’s exit poll.
In 2008, Democrats had a 13 point registration advantage and a 16 point identification advantage in the final analysis (further indicating that the bump in party registration after the presidential primary was more a corrective than anything else).
Currently, Democrats have a 14 point registration advantage. And that lines up with the 14 point identification advantage they hold in the most recent Monmouth University/Gannett New Jersey Poll of likely voters.
Voter Party Identification is Stable, but Not Immutable
Recently, former House speaker Newt Gingrich accused the ABC News/Washington Post Poll for cooking the books on its party identification numbers. Knowing and having worked with those pollsters, I can say without qualification that Gingrich’s charges are wholly without merit. I’ll also add they are petty, juvenile, and uninformed.
But this incident brings to light one of the realities of polling. Party identification is not a demographic “fact” like, age, education, race, gender. It is an attitude. And as such, it is vulnerable to change, although history indicates that such change is usually gradual (with the exception of lightening rod-type events like Watergate).
That phenomenon has been evident here in the Garden State. After falling somewhat during the summer, Democratic party identification has inched up a couple of points in the last two months.