Elections these days are determined in part by the swing of unaffiliated voters, which both sides closely contest. They also hinge on how strongly each party’s base turns out to vote.

The base vote can be a huge factor in elections. Consider, for instance, that in 2000 George W. Bush won 72 percent of people opposed to abortion in most or all cases. In 2004, he won 75 percent of that bloc. Moreover, turnout among this group increased from 40 percent to 42 percent of the population. Taken together, this increase among an already solidly Republican group added about 3.2 million votes to Bush’s total in 2004, more than his entire margin of victory over John Kerry.

When we look at Obama’s voting coalition, African Americans without doubt constitute the core of his constituency. And it is truly extraordinary how much better Obama did in 2008 with them than previous Democrats.

In 2004, John Kerry won 88 percent of the black vote, which made up 11 percent of the total electorate. In 2008, Obama won 95 percent of the black vote, which made up about 13 percent of the electorate. That means Obama’s increase in support relative to Kerry won him about 3.5 million extra votes, or nearly 40 percent of his total margin over John McCain.

This was a truly amazing feat, considering what we know about people’s voting habits. In particular, political scientists for generations have found that voting is highly correlated with socioeconomic status. About 61 percent of the white voting age population turned out to vote in 2004, compared to only 50 percent of the black voting age population.

But enthusiasm for Obama in 2008 among African Americans basically erased this underlying tendency: whites turned out at 62 percent; African Americans turned out at 60 percent.

Campaign strategists like to pat themselves on the back for how greatly their organizational genius influences elections, but the surge in the black vote is a good case in point of how demographic trends actually influence campaigns, rather than vice-versa. Consider the increase in turnout in 2008 compared to the African American population:

There are only two states on this list that have a relatively small share of African Americans – Colorado and Indiana. Indiana was a swing state for the first time in generations, so the turnout increase makes sense. Colorado is increasingly competitive and has a large and growing Hispanic population.

The remaining states have some of the largest percentages of African Americans in the country. In fact, eight of the top ten states in terms of black population are on this list – with only Delaware and (Katrina-affected) Louisiana missing. Nationwide, there was a very strong relationship between black share of the population and turnout increases – a 10-point increase in a state’s black population suggested a 4.4 percent increase in its turnout:

So, to put it simply: in 2008 the first major party black nominee clearly overcame the historical tendency that socioeconomic status holds back black turnout.

My question today: will that happen again, especially since the socioeconomic position of African Americans has gotten worse relative to whites in the intervening years?

There is circumstantial evidence to suggest that Obama has worries about this bloc. If you’re wondering why Obama is reportedly having conference calls with black pastors like Jeremiah Wright, going on radio shows hosted by people nicknamed “Pimp with the Limp,” and sounding off about the feud between Mariah Carey and Nicki Minaj, perhaps this is the answer.

This might also explain Romney’s sustained polling edge in both Florida and North Carolina. Both states swung to Obama in 2008 because of substantial increases in black support. Now, Obama trails in both – suggesting that increased opposition among other blocs, namely non-Hispanic whites (and, in Florida, Cubans), is currently more substantial than his support among African Americans.

This should serve as a cautionary tale when looking at nationwide horse race polls. Those that mimic a nationwide Democratic advantage similar to 2008 perhaps should be taken with a grain of salt. That edge occurred partially because of turnout spikes in non-competitive states with large black populations – like Alabama, the District of Columbia, and South Carolina. What’s more, there was a substantial demographic shift in states like California, Connecticut, Illinois, Massachusetts, and New York: none of these states saw overall big increases in turnout, but did see large jumps in the black vote.

Again, very little of this was induced by the Obama-Biden campaign. To be clear: enormous credit is due to them for a careful study of the demographics in Indiana, North Carolina, Virginia (and, for that matter, Nebraska’s second congressional district, which is 10 percent black and went narrowly for Obama). Obama’s team intuited that it could use this expected surge in the black vote to pick up electoral votes that are not usually contested, but the campaign itself did not create this enthusiasm.

Final point: Obama has pressure to do as well with the black vote this year because he is on track to do substantially worse with white voters than four years ago.

In 2008, he won 43 percent of the white vote; this year, I am betting on 40 percent or less. That would make for at least a 4.5-point nationwide shift toward Romney relative to 2008; and remember that Obama’s total margin of victory was 7.3 points. But there’s potentially more to the story: white turnout was flat in 2008 relative to 2004 – 62 percent of the voting age population compared to 61 percent, i.e. within the exit poll’s margin of error. On the other hand, white turnout increased between 2000 and 2004 by 5 points.

So, if whites turn out at a greater rate than they did in 2008 and vote for Obama at a lower rate, that is going to exert greater pressure on the president to match his extraordinary haul among African Americans.

Nobody knows for sure exactly what will happen on Election Day. But this is an underlying trend to keep a close eye on. At the end of the day, it comes down to the fact that the black vote will break at least 90:10 in favor of the president. So, for every African American who does not turn out this cycle, the president is about 90 percent likely to lose a vote. This means that seemingly small shifts with this relatively small demographic group can have huge implications on who wins on November 6.

Jay Cost is a staff writer for THE WEEKLY STANDARD and the author of Spoiled Rotten: How the Politics of Patronage Corrupted the Once Noble Democratic Party and Now Threatens the American Republic, available now wherever books are sold.

Next Page