Since the off-year Election Day, I’ve been noodling around with some exit poll comparisons for NJ and VA in 2009, 2012 and 2013 (sadly, there were no exit polls in either state in 2010, which would be useful to know about in looking ahead to 2014). I quickly discovered the composition of the electorate in both states was quite similar in 2009 and 2013–with one glaring exception in VA, as I wrote about at Washington Monthly:
In New Jersey the 2013 electorate looked an awful lot like it did in 2009, and quite different from its composition in 2012. The racial breakdown was 73% white, 14% African-American and 9% Latino in 2009, and 72% white, 15% African-American and 9% Latino in 2013. By contrast, it was 67% white, 18% African-American and 10% Latino in 2012. You see a similar pattern with the vote by age: in 2009, voters over 50 represented 55% of the vote while those under 30 were 10%. Yesterday voters over 50 were 59% of the vote while those under 30 were 10%. In 2012, over-50s were 49% while under-30s were 16%.
So New Jersey followed the expected pattern of an off year election producing a significantly older and whiter electorate than in a presidential year. Christie would have won with either electorate, but he did have a stiff wind behind him this year.
The age breakdowns in Virginia follow the same pattern. Over-50s were 54% in 2009 and in 2013, but only 43% in 2012. Under-30s were 10% in 2009 and 13% in 2013, but rose to 19% in 2012.
But the racial breakdowns broke the mold a bit: in 2009, the Virginia electorate was 78% white and 16% African-American (with 5% Latino or Asian). In 2012 it was 70% white and 20% African-American (with 8% Latino or Asian). And yesterday it was 72% white, 20% African-American (with 5% Latino or Asian). It’s unclear whether the McAuliffe campaign did an unusually good job of turning out the African-American vote, or something else was going on, but it is clear it was a key factor in his victory, since the additional 4% of the electorate that were African-American as compared to 2009 represented close to 90,000 votes. He won by just over 54,000.
Since I wrote that quick analysis, there’s been a lot of talk about the composition of the VA electorate resembling that of 2012, but little or no focus on the African-American vote specifically. This, too, I mentioned at Washington Monthly:
Now comes the magisterial Ruy Teixeira at TNR with a deeper look at Virginia, and he, too, focuses on the unexpected composition of the electorate:
In 2009, Virginia voters were 78 percent white and 22 percent minority. In 2013, they were just 72 percent white and 28 percent minority–not far off the 70/30 split in the 2012 presidential election. There you have the key to McAuliffe’s victory: Despite performing much better among white voters than the hapless Creigh Deeds, McDonnell’s Democratic opponent, McAuliffe would nevertheless have lost this election if the white/minority voter distribution had mirrored that of 2009. It was the increase in the minority vote that put him over the top.
But here’s the thing: according to the exits, the Hispanic/Asian percentage of the vote came in this year at 2009 (5%), not 2012 (8%) levels. And the age composition of the electorate was very much like that of 2009, not 2012. Nor was there any “super sizing” of the overall electorate; total turnout was up a bit from 2009, but nowhere remotely close to presidential levels. So what we are looking at is not some sudden change in the overall size or configuration of the off-year vote, but a pretty isolated but very significant surge in African-American turnout.
Ruy has no particular explanation for this phenomenon; nor have I. I’ve heard a few random folk cite the pre-election voter purge executed by Virginia (about 37,000 people suspected of dual registrations were disqualified, not the kind of purge most likely to overwhelmingly target minorities) as a provocation to black voters. And there’s a general sense that the McAuliffe campaign devoted a significant portion of its abundant resources to GOTV efforts, which would naturally affect African-American turnout. But that was quite a surge in the black vote, and Democrats looking ahead to 2014 ought to go to school on it.
So the mystery remains, but I’m sure there’s an answer that some Democrats in Virginia are chortling about.