Politics being an American sport as much as football or baseball, there has already been a lot of discussion about the 2012 presidential election (in addition to more reasonable discussion about the upcoming midterm elections next fall). It seems to me that to get a sense of the landscape for 2012 even this early, we would do well to look at the demographic conditions prevailing in the 2008 election.
I got my information by looking at election and exit polling results from presidential elections in 1972 through 2008. Dave Leip's Atlas of Presidential Elections was very helpful in that part, while I had to use a variety of sources for the exit polling, although the New York Times was generally the most consistent source.
I am still chewing through the data, but found some interesting indicators worth consideration. For instance, the total voters for 2008 made up 41.5% of the population, slightly down from 2004's 41.9% but above the 1972-2008 average of 39.3%. Obama in 2008 took 2.8 million more votes than Kerry did in 2004, while McCain in 2008 took 4.5 million fewer votes than George W. Bush in 2004. Barack Obama won the White House by about 8.5 million votes, the largest popular-vote margin of victory since Reagan's 1984 re-election, when he beat Mondale by 16.9 million votes. 2.7 million more men voted in the 2008 election than in 2004, while women voters increased from 2004 to 2008 by only 450 thousand voters. White voters decreased from 2004 to 2008 by 1.3 million, while black voters increased from 2004 to 2008 by 2.9 million voters - since McCain won white voters by a margin of 55% to 43% while Obama won black voters by a margin of 95% to 04%, this demographic shift is an important element in the Democrat's victory. Hispanic voters increased by 1.5 million from 2004 to 2008, and since Republican support by Hispanics fell from 44% to 31% while support for the Democrat increased from 53% to 67%, this also represents a significant shift which favored Obama. Voters under 29 (a strong demographic for Obama) increased by 1.8 million, while voters over 60 (a key sector for McCain) decreased by half a million voters, which also played into Obama's advantage.
Consequently, the exit polls demonstrated that race, gender, and age all affected the outcome of the 2008 presidential election. The next step would be to consider the thresholds of those demographic qualities, or the historical levels of support needed to win.
With regard to male voters, Bill Clinton won in 1992 with only 41% of male support. Barack Obama came close to that, with just 43% support. If male voters continue to grow faster than female voters, this could be a problem for Obama in 2012.
The 13% representation by black voters was the highest ever in a federal election. Obama won the black vote by a margin of 14.9 million votes (meaning that Mccain took 6.4 million more non-black votes than Obama). To win re-election, therefore, President Obama needs to enjoy similar turnout by black voters; statistically every black voter who voted in 2008 but does not vote in 2012 is twenty-three times more likely to hurt Obama than his Republican opponent by staying home.
Seniors represent a readitionally significant and active voter bloc. If they return in numbers in 2012, this will be a significant problem for Obama.