The latest report from Ruy Teixeira and John Halpin of the progressive Center for American Progress contains a thoughtful examination of President Obama's re-election chances. There's an awful lot packed into the 60 pages of text, but the basic thrust is as follows: We should expect the non-white share of the electorate to grow at least 2 percent from the 2008 election, padding Obama's base line. If he can hold serve among either the white working class or college-educated whites, he should be able to pull out a victory, even amid troubled economic times.
In truth, the report is substantially less bullish on Obama's re-election chances than some of the articles analyzing it have suggested (see Dan Balz here and Michael Tomasky here). It acknowledges that the president has a “tight needle to thread” and that “Americans will be open to replacing President Obama with an even-tempered, nonthreatening GOP leader focused on the economy.” In other words, the triumphalism of Teixeira’s “Emerging Democratic Majority” argument of the early 2000s is decidedly tempered throughout the report, and with good reason. After all, the GOP just won its second-largest share of seats in the House of Representatives since 1928, with an electorate that had the second-smallest share of non-Hispanic white voters in history. In retrospect, those repeated “last gasps” of the GOP coalition (1994, 2002, 2004, 2010) look a lot more like “steady breathing.”
But the optimistic tones in Teixeira and Halpin’s piece need to be tempered even further. The “demographics versus economics” debate that Teixeira and Halpin suggest will determine the outcome of the next election isn’t a 50-50 proposition. It is weighted heavily toward the economics side, and I think it’s unlikely that demographics will save the president. There are three critical observations here:
1. The minority population may not grow substantially from 2008 through 2012.
Probably the central feature of the Teixeira/Halpin argument is that the nonwhite share of the electorate should have grown 2 percent by 2012, reducing the white share of the electorate to 72 percent (for simplicity’s sake, I’ll shorten “non-Hispanic whites” to “whites”). This is certainly possible, as the white share of the electorate has contracted by 2 percent, on average, in every presidential election since 1980.
But it hasn’t been a straight line. In 1992, the white share of the electorate actually increased by 2 percent, in response to H. Ross Perot’s candidacy and the economic contraction. In 2004, the white share of the electorate declined by 4 percent, in part due to the growth of the Latino population.
So why might we expect the demographic changes in the electorate to be more like 1992 than 2004? First, the Latino share of the electorate has actually remained stagnant for much of the past decade. In 2004, Latino voters comprised 8.24 percent of the electorate. In 2006, they were 7.94 percent of the electorate. In 2008, they were 8.38 percent of the electorate. In 2010, they were once again around 8 percent. In other words, for a variety of reasons, the surge in Latino population has not translated into a surge in Latino voting power (and remember, there was a huge registration and get-out-the-vote drive in 2008 among Latinos, both in the primaries and the general election).
And while the headline from the release of the decennial census was the surge in the Latino share of the populace, the lesser-known truth is that Latino immigration has largely stopped over the past several years. It may have even reversed. There are multiple reasons for this, including the United States’ deep recession and slow recovery, as well as the continued modernization of the Mexican economy. In other words, to the extent that Latino immigration is what accounts for the increase in the Latino share of the electorate from 1992 through 2004, we should not expect it to do so from 2008 through 2012.
But while the Latino share of the electorate was stable from 2004 to 2008, the white share of the electorate nevertheless decreased. Why would this be? The answer is simple: The increase in the non-white share of the electorate from 2004 to 2008 was largely driven by a surge in African-American voters. The African-American share of the electorate is typically between 9 and 11 percent. In 2008, it was 13 percent, by far the largest vote share in history.
The problem for the president is that he has probably maxed out among these voters -- the African-American share of the electorate in 2008 was about 10 percent more than their share of the population as a whole. In fact, I wouldn’t be surprised if the number of African-American voters declined somewhat in 2012. This isn’t because African-Americans are disappointed in Obama -- his approval among African-American voters remains stratospheric -- but rather because it will probably be much more difficult to energize marginal African-American voters with the prospect of re-electing the first black president than it was to energize them with the idea of electing the first black president.
But the bottom line is that we should be very surprised if the African-American share of the electorate increases further from 2008 to 2012. We likewise have good reason to believe the Latino share of the electorate will remain stable. This suggests a pretty strong argument that the minority share of the electorate will be roughly the same in 2012 as it was in 2008, and a decent argument that it might contract somewhat.
(2) Obama will have a difficult time winning either white working-class voters or upscale whites.
Even if the non-white share of the electorate does increase by an additional two points in 2012, Obama still faces an additional hurdle. As Teixeira and Halpin suggest, Obama cannot afford to lose white voters at the same rate Democrats lost them in 2010. In other words, he must hold serve among either upscale whites or downscale whites if he hopes to win.
But keeping the margin close, much less winning, among either group will be difficult. Remember, Obama enjoys a low-to-mid-40s approval rating right now for one reason: He maintains an approval above 80 percent among African-Americans. Among whites, it is a mere 33 percent. At that level, there are almost by definition very, very few subgroups of whites who approve of the job the president is doing.
So when we see, for example, that Obama’s job approval among all adults making more than $7,500 a month is 40 percent, we can probably imagine that his overall approval among upscale whites is a few points lower than that. His job approval among adults making $2,000-$7,500 a month is not much different, and his job approval among adults with “some college” or a “high school diploma or less” is also in the low 40s. Once again, we can pretty safely assume that his job approval among whites in those categories is somewhat lower.
In other words, Obama doesn’t just have some “tidying up” to do among various white groups. He has to either improve his image there by about a point a month over the next 11 months, or hope for a Republican nominee so unacceptable to the overall populace that Obama can convince a substantial number of voters who disapprove of him to nevertheless cast ballots for him. Right now, the latter looks much more likely than the former.
(3) Winning minority voters and white voters is something of a zero-sum game.
In a little more than a month, my book, “The Lost Majority,” hits the stands. The central argument of the book is that the famous permanent Republican and Democratic majorities that many commentators foresaw emerging in the 2000s were mirages, precisely because long-term, permanent majorities are almost always impossible in this country.
There are myriad reasons for this, but two are of particular importance here. First, as new issues emerge, the party that is in power will inevitably have to choose winners and losers on these issues from among its coalition. This is even more pronounced in a time of economic stagnation, when the question isn’t “who gets the new slices of pie?” but rather “who is going to have to give up their share of the pie?” Second, the party that is out of power will adapt, and will chase after groups that the other party either takes for granted or ignores.
So, for instance, Obama can try to shore up his support among Latino voters by embracing immigration reform and combating Arizona’s profiling law. But in doing so, he risks alienating white working-class voters and, to a lesser extent, upscale white voters. In fact, this is precisely what happened in Arizona in 2010. Jan Brewer won the state by three points more than John McCain, despite running about 13 points behind McCain among Latino voters. She more than made up for this decline among Latinos by increasing her share among whites (who are still three-quarters of the Arizona electorate) by three points. The idea that there is a zero-sum game at work here is an inherent assumption underlying the argument that Obama has to choose between a “Colorado strategy” focusing on upscale whites, or an “Ohio strategy” focusing on downscale whites.
To be clear, if Republicans win total control of the government in 2012, they’ll have to make similar tough choices. Holding together a party composed of semi-secular soccer moms in Loudoun County, Va., evangelical attorneys in Edmond, Okla., and Catholic auto workers in Youngstown, Ohio, is almost as difficult as holding together a coalition of blacks, Latinos, working-class whites, suburbanites and urban liberals. These types of difficulties run throughout history, and they help explain why parties almost never win the popular vote more than three times in a row.
But for now, Obama is the president. The state of the economy, as well as policy choices made early in his term, are forcing him to pick winners and losers among his 2008 electoral coalition. Republicans will craft their 2012 message based in large part around the choices he makes. Barring a gift from the Republicans in the form of their nominee -- and this is something we absolutely should not rule out -- the president will likely have a very difficult time holding it together.