Can McCain and Obama Win by Targeting Specific Age Groups in Swing States?

We had greater differences between the age groups in the Democratic presidential primaries than any I can recall seeing, and we are seeing significant differences between age groups in general election polls. The ABC/Washington Post poll has Obama ahead among the under-30s by a whopping 66 percent to 30 percent, while McCain leads among over-65s (technically, I should say "65 and overs," but "over-65s" is more succinct) by 45 percent to 40 percent. Quinnipiac has Obama leading among under-35s by 63 percent to 31 percent, while McCain edged Obama 45 percent to 44 percent among over-55s. The CBS/New York Times poll showed Obama leading among under-30s by 48 percent to 36 percent, while McCain led among over-65s by 42 percent to 40 percent. In general, Obama's current lead in recent polls is due entirely to his lead—by as much as 2-1—among young voters.

But will they vote? The ABC/Washington Post (see page 2) poll shows young voters significantly less likely to say they would vote than they were in March. The Obama campaign is sending in organizers to register and turn out young voters—a good use of its copious resources, I think.

To see where young—and old—voters are concentrated, I took a look at the Census Bureaus 2007 estimates of the states' populations by age. Nationwide, under-30s were 22.2 percent of the 18-and-over population, while over-60s were 23.0 percent. But the numbers in the 50 states (not 57, as Obama said in May) are somewhat different.

What states have unusually large percentages of under-30s? Here's the list of those with 24 percent or more:

• Utah (32 percent). Utah has the same demographics of the America of the 1950s, with lots of children and young families and comparatively few old people. Of course it's also the nation's most Republican state. You can bet the Obama campaign is not sending staffers into Salt Lake City.

• District of Columbia (28 percent). The nation's most heavily Democratic constituency is becoming increasingly yuppified. Since D.C. whites voted 81 percent for John Kerry in 2004 (while Hispanics there were only 66 percent for Kerry), there's not going to be much organizing here.

• North Dakota (26 percent). This was a surprise to me. North Dakota has been losing population this decade, but it also has a very low unemployment rate, and apparently 20-somethings are staying in the state—a goal of its politicians for a long time. Or perhaps this number reflects the fact that sparsely populated North Dakota has a couple of Air Force bases, with several thousand young men stationed there (which also explains why North Dakota's black men have very low rates of crime; I suspect most of its blacks are in the Air Force). In any case, a recent poll showed Obama tied with McCain in the state, which voted heavily for George W. Bush. Registering young voters could pay off for the Democrat.

• Texas (25 percent). I gather that the Obama campaign is sending organizers into Texas in an attempt to force the much less affluent McCain campaign to spend money in what has been a safely Republican state since 1980. The under-30 population here is disproportionately Hispanic, which could help Obama narrow the McCain margin, but it's hard to see how he wins.

• Louisiana (24 percent). Not as heavily Republican as Texas has been but still a stretch for Obama.

• California (24 percent). Again, the under-30s here are heavily Hispanic. But California is so heavily Democratic even without their votes that an organizing campaign would hardly be in order.

• Idaho (24 percent). This is one of the most heavily Republican states in the nation. Obama won 80 percent of the votes in the Democratic caucus here, in which 21,221 people participated. But that's only 4 percent of the 598,447 Idahoans who voted in November 2004.

• Oklahoma (24 percent). Another heavily Republican state and one in which Obama was shellacked in the Democratic primary.

• New Mexico (24 percent). A target state in 2000, 2004, and again this year. Registering young Hispanics should be a no-brainer for Obama, and I'm sure Democrats are already busy doing so.

• Mississippi (24 percent). The good news for Obama is that this is the state with the highest black percentage of the population, 37 percent. The bad news is that very few Mississippi whites are going to vote for Obama (the surprise Democratic winner of the 1st District House race, Travis Childers, has been shunning him) and that not even the most fabulous registration drive is going to result in a majority-black electorate.

So the pickings are a little slim for the Obama campaign: among the states with unusually high percentages of under-30s, only North Dakota (3 electoral votes) and New Mexico (5 electoral votes) are likely to be target states.

What about states with unusually large percentages of voters 60 and over? Here is a list of those with 25 percent or more:

• Florida (29 percent). McCain has been running relatively well in polls here, better than he has in Virginia. (Times change: in 1976 Virginia was one of only two Southern states not carried by Jimmy Carter, the other being Oklahoma.) Obama seems weak among older Jewish voters in south Florida: Hillary Clinton beat him 57 percent to 33 percent in Broward County and 61 percent to 27 percent in Palm Beach County in the primary. The McCain campaign seems to have nothing like the organizational capabilities of the Bush 2004 campaign. But it would be sensible of the McCain team to make efforts to turn out older Republican voters in much of the state and to make a major persuasion effort among Jewish voters on the Gold Coast.

• West Virginia (27 percent). This was Hillary Clinton's second-best primary state, after Arkansas, and Obama seems to have no chance here. Appalachian voters seem ill disposed toward him.

• Pennsylvania (26 percent). A target state in which Obama has been running ahead in polls since he clinched the nomination. The obvious target for the McCain campaign is all those old people in small towns clinging to guns and God.

• Maine (26 percent). The Bush campaign thought this was a target state in 2000, but it slipped off the list in 2004 and seems far out of reach for McCain now.

• Arkansas (26 percent). See West Virginia.

• Iowa (26 percent). Obama has been running very well in this state, and Democrats have done a great job over the last several years registering their voters and getting them out on Election Day. It's not clear that older Iowans are a demographic for McCain, who skipped the caucuses here in 2000 and dropped in only occasionally in 2008.

• South Dakota (25 percent). A recent poll showed McCain ahead by only 4 percent here. Older voters here are, I suspect, more inclined to McCain than are their neighbors in Iowa.

• Montana (25 percent). Democrats here hold both Senate seats, the governorship, and majorities in the legislature, and Obama has been running ahead in some polls. Memories of the depredations of the Clinton-Gore environmental policies have faded. I think it would be a mistake to leave Montana off the list of target states.

• North Dakota (25 percent). A recent poll showed the state even. I suspect that the dynamic in the Dakotas and Montana, where young people seem to have been staying in the state in this decade, is that younger voters are breaking heavily for Obama, while older voters are staying with McCain.

• Alabama (25 percent). I think Obama has no chance here.

• Delaware (25 percent). I was surprised to see this state on the list. Are retirees clustering in the beach communities of this state with no sales tax? There's no polling here, but Obama is assumed to be well ahead.

If it makes sense for McCain to work to turn out older voters—a big if, and this group usually turns out pretty well anyhow—then there are more targets among these heavily over-60 states for McCain than there are in heavily under-30 states for Obama: Florida (27 electoral votes), Pennsylvania (21 electoral votes), Iowa (7 electoral votes), South Dakota (3 electoral votes), Montana (3 electoral votes), and North Dakota (3 electoral votes). Not what I would have guessed before I ran the numbers. Of course, in close states it may make sense for both campaigns to organize their target demographics even if they are relatively sparse as a percentage of eligible voters.