This was a decisive but not an overwhelming victory for Barack Obama and the Democrats. As I put it in the lead of my U.S. News column for next week, it was a victory that was overdetermined and underdelivered. Obama's apparent percentage margin (we're still waiting for a lot of returns) is 52 percent to 46 percent, slightly higher than Bush 43's score in 2004 and slightly higher than Bush 41's in 1988. He owes his impressive 364-to-174 margin in the Electoral College to narrow wins in Florida, North Carolina, Virginia, Ohio, and Indiana; he nearly added 11 more in Missouri.
I haven't had time to comb through the election returns yet to make definitive conclusions. I hesitate to crunch a lot of numbers that will change when final returns are in, but here are a few observations.
The young vote. As I said at the American Enterprise Election Outlook session Thursday, the one sure way for John McCain to have won the election was to pass a constitutional amendment raising the voting age to 35. Not only Obama but Democratic House candidates, as Patrick Ruffini notes, won by huge margins among voters under 30, according to the exit polls. Obama got 66 percent among those under 30, won 30-somethings 54 percent to 44 percent, ran essentially even with 40-somethings and 50-to-64s, and carried only those 65 and over (53 percent to 45 percent). First-time voters were 69 percent Obama, 30 percent McCain.
It's highly unusual to see such a big difference between age cohorts, though we certainly saw it in the primary contests between Obama and Hillary Clinton. And I think it's enormously good news for the Democrats and bad news for the Republicans. Of course, President Obama may turn out to disappoint many of these voters. If it turns out, contrary to his victory statement after the last primaries, that this is not the moment that the seas stop rising and the planet begins to heal, they may be as turned off by him and his party in 2010 or later as they are turned on today. But in the meantime, Republicans need to think how they can appeal to these voters. As I've written before, I think important clues are available in Morley Winograd and Michael Hais's book, Millennial Makeover. Young voters trust government more than those who remember the 1970s, but they want choices, not government one-size-fits-all solutions.
The over 30s. Interestingly, the baby boom generation, voters 45 to 64 (born between 1944 and 1963), were 50 percent Obama, 49 percent McCain—another indication that this is not a liberal age cohort but a divided one (between Bill and Hillary Clinton, Al Gore and John Kerry on one side and George W. Bush and Dan Quayle on the other). Another way to look at it is that older boomers (born 1944 to 1957) are slightly more for Obama (50 percent-49 percent), than Obama's own age cohort, those in their 40s (born 1958 to 1968), who were tied 49 percent-49 percent. This is, roughly, the age cohort identified by Jonathan Pontell as Generation Jones. It's interesting that this group was (slightly) less enchanted with Obama than any other age group except the elderly (born 1943 or earlier)—or at least not nearly as enchanted as the young.
A footnote on Latinos. McCain, despite cosponsoring comprehensive immigration bills, did far less well with Latinos than George W. Bush. Barack Obama won elderly and young Latinos by overwhelming margins—the young presumably for the same reasons that other young voters were attracted to him, the elderly because they're pretty downscale. Of those in the middle age groups, 30 to 64, Obama led by a narrower margin of 60 percent-38 percent—a big margin, but closer to whites of that age than to blacks.
The top and bottom coalition. Obama led among those with incomes under $50,000 (big) and those above $200,000 (narrowly). Among the 56 percent with incomes in the middle, it was pretty much even. Similarly, Obama won 63 percent among those with no high school education and 58 percent among those with postgraduate degrees but led only very narrowly among those in between. That's reflected in the finding that McCain did better with noncollege whites (58 percent-40 percent) than college whites (51 percent-47 percent). At the moment, this top-and-bottom coalition outnumbers the broad middle. But if the political balance tips, it could be the other way around.
The first politician that I've observed assemble a top-and-bottom coalition was Mayor John Lindsay of New York. He started off as a Republican, a very liberal one, and won the mayoralty in 1965 and 1969 with coalitions of affluent Manhattan whites on the one hand and blacks and Latinos on the other. In both elections, he won with a plurality of the vote and was behind in the four outer boroughs taken together. Lindsay championed soft policies (in the sense of the word in my book Hard America, Soft America) on crime and welfare that produced disaster in New York and in other cities where such policies were followed. Those policies answered the demands of both sides of his top-and-bottom coalition: The top wanted generous policies toward the poor that made them feel good about themselves, the bottom wanted short-term money transfers and leniency toward criminals in their midst. The top paid a small price for the results of these policies; the bottom paid a very large one. Fortunately, soft policies on crime and welfare were abandoned after three decades, thanks to Republicans like Gov. Tommy Thompson of Wisconsin and Mayor Rudolph Giuliani of New York and the many other politicians (mostly Republicans, but including many Democrats) who followed their lead. I'll leave to future speculation the consequences of policies Obama may follow to meet the demands of his top-and-bottom constituency.
Further evidence of this constituency comes from a cursory examination of county election returns, using this nifty Washington Post election map graphic. I took a look at the McCain percentages in major metropolitan counties in target states and compared them with Bush '04 percentages there (a comparison of Obama percentages and Kerry '04 percentages would have shown pretty much the same thing). I used whole-number percentages, on the theory that the 2008 percentages on the map are subject to some change as late returns and absentee votes come in and that any attempt to go to tenths of a point would just represent spurious precision. The results suggest that the Democrats under Obama are moving more and more to a top-and-bottom coalition. Here are the states I looked at.
Pennsylvania. McCain was -7 percent in Chester County and -5 percent in Montgomery County, the most affluent Philadelphia suburban counties (Chester County went Democratic for the first time in ages). But in Allegheny and Beaver Counties (Pittsburgh and its industrial suburbs), McCain maintained Bush's percentage, and in Washington and Westmoreland Counties (west and east of Pittsburgh), he was +2 percent on Bush.
Virginia. Interestingly, McCain's biggest losses from Bush were not in Fairfax County (-5 percent) but in exurban Loudoun (-9 percent) and Prince William in Northern Virginia, in line with the trend in the 2005 governor and 2006 Senate races there. Also McCain, despite his Navy background, ran far behind Bush in Tidewater cities Virginia Beach (-8 percent), Newport News (-9 percent), and Hampton (-10 percent). Increased black turnout could have been a factor there.
North Carolina. McCain's biggest declines were in Charlotte's Mecklenburg County (-11 percent) and Raleigh's Wake County (-9 percent), fast-growing areas that may have been hit by housing price declines. Charlotte is the headquarters of Bank of America and Wachovia Bank.
Florida. McCain's biggest decline in a big county was in Orlando's Orange County (-10 percent), which has a large Hispanic population. Republicans also lost two House seats in the Orlando area. The Republican decline was less in the Gold Coast and Tampa Bay.
Ohio. An interesting case. In northeast Ohio, industrial and Democratic, McCain ran just about as well as Bush—better, actually, in Summit County (Akron) and Trumbull County (Warren, hometown of blogger Hugh Hewitt), and just 1 percent less in Portage County (Kent State), Stark County (Canton), and Mahoning County (Youngstown). As in western Pennsylvania, Joe the Plumber may have struck a chord. McCain ran farther behind Bush in Columbus's Franklin County (-5 percent) and its northern suburbs in Delaware County (-7 percent), where real estate values have been higher and where price declines were probably more conspicuous.
Indiana. McCain's biggest declines from Bush '04 were in Indianapolis's Marion County (-14 percent) and well-to-do suburbs in Hamilton and Hendricks counties (both -13 percent). This looks like a long-delayed move of affluent folks from the Republicans that we saw in many larger metro areas in the 1996-2004 period. In contrast, in industrial Lake County, right next to Chicago, McCain was just 6 percent behind Bush—a drop which, if duplicated statewide, would have left Indiana comfortably Republican.
That's all for right now. I'll be spending much of the weekend looking over the numbers.