Hurricane Sandy and the Election in New Jersey

Cross-posted at PolitickerNJ

An unknown number of provisional ballots remain to be counted in New Jersey, but a few threads are emerging on the presidential election.  Turnout in the Garden State was down by a lot.  Currently, the number of people who casts votes in the presidential election about 500,000 less than in 2008 – about a 14% drop.

That gap will certainly shrink as provisional ballots are tallied, but it will still mark the biggest drop in turnout of all the states.  Nationwide estimates provided by Edison Research of Somerville – the firm that conducts the TV networks’ exit poll – suggest that turnout will only have dropped by about 2% nationally compared to 2008.  New Jersey’s turnout is far behind that figure.

Let’s assume that total turnout in New Jersey ends up being nearly 3.5 million.  This represents about 63% of registered voters, which would be the lowest percentage on record since 1972, when 18-year olds were given the right to vote.  But the voter rolls may not be the best base for comparison.  Registration numbers took a big jump in 2008 because of concerted registration efforts and in 1996 because of the Motor Voter law. Prior to that, fewer eligible voters were actually registered.

If we consider turnout as a percentage of the total voting age population (VAP) or of the voting eligible population (VEP), this year’s numbers hold up against past elections.  Using about 3.5 million voters as a final estimate, New Jersey turnout may wind up being 51% of VAP or 59% of VEP.  Those results either match or exceed statewide turnout in both 1996 and 2000.

Given what the state has gone through over the past two weeks, these turnout numbers don’t look all that bad.

Now let’s look at how New Jersey voted in the presidential contest.  Nationwide, Barack Obama’s winning margin was smaller than it was in 2008.  This trend was true in nearly every state.  In fact, only four states showed Obama improve on his margin from four years ago.

These four states include Alaska, where he narrowed his losing gap by 8 points, and the Gulf States of Louisiana and Mississippi, where he lost by about a point and a half less than in 2008.

And this group also includes one blue state where Obama actually increased his winning margin.  That would be New Jersey, where the president’s margin went from about 15.5 points in 2008 to 17 points this year.

It’s worth noting that polls conducted before Hurricane Sandy hit the state showed Obama with only a 12 point lead on average.  It’s also worth noting that those same polls showed U.S. Senate incumbent Bob Menendez with an average 19 point lead – which is what he actually got on Election Day.

There is no doubt that Hurricane/Superstorm Sandy had an impact on how New Jersey voted in the presidential race -- 54% of New Jersey voters told exit pollsters that Obama's response to the disaster was an important factor in their vote.  Some observers, though, put Obama's winning margin down to a lower turnout in the harder hit Republican shore towns.  This certainly happened, but Democratic urban areas were also affected.

Using the preliminary vote counts, turnout in Ocean County was down about 19% compared to 2008.  But it was also down 19% in Essex County and 17% in Hudson County.

The difference is who turned out in those counties.  Obama cut his losing margin in Ocean County from about 18.5 points in 2008 to 17.5 points in 2012.  And he improved his winning margins in Essex by 3 points and Hudson by 9 points.

In Gloucester County, an area of the state spared most of Sandy’s wrath, turnout was down by just 4%.  Obama’s winning margin there went from 12 points in 2008 to just under 11 points this year.  Based on this result, even if more voters could have made it out to vote, Obama’s statewide margin may have dropped by only a couple of points.  This is still better than how he was doing in Garden State polls prior to Sandy.

A note on national polling:

It appears that nearly all national polls performed well within their individual margins of error, but most – including Monmouth’s – had a slight Republican skew in the nominal horse race.  So all those folks who claimed that we needed to “unskew” the polls were partially right.  They just had it in the wrong direction – which they would never admit, of course.  As Fox News anchor Megyn Kelly asked Karl Rove, “Is this just math that you do as a Republican to make yourself feel better or is this real?”  As we now know, it’s definitely not the latter.

My first-read suggests that the pollsters who came closest to the mark – which may end up being as much as a 3 point win for Obama when all the votes are counted – employed samples with more voters who are contactable by cell phone only.  This gibes with the exit poll findings that showed an increase in the proportion of the electorate who were under the age 30 or not Caucasian (i.e. Black, Latino, and Asian).  Young voters made up 19% of the electorate – compared to 18% in 2008 – and non-white voters comprised 28% of the electorate – up from a then-record 26% in 2008.

These groups are emerging as solid Democratic voting blocs.  As recently as eight years ago, young voters and Asians, and to a lesser extent Latinos, were much more up for grabs to the GOP.  Now they are solidly Democratic – and they are reachable only by cell phone or other electronic device.