Electoral Math
Reality-Based^TM Political Numbers from Nicholas Beaudrot

| Home | Yeah, That's About Right

The Electoral Math of Social Security, or, Karl Rove's fuzzy math, part #870
Now Playing: Franz Ferdinand / Franz Ferdinand / Jacqueline

After many weeks of slacking, I've finally managed to draw meaningful crosstabs out of the 2004 exit polls. My hope was to find a useful way of exploiting Bush's "plan" for Social Security unreform to win seats in the House and Senate for the 2006 midterms. It appears there are no real surprises: the best potential Republican voters for Democrats to woo to their side are older voters, particularly older women.

Before looking for new voters, I want to touch briefly on those voters that may abandon Democrats because they support Bush's "plan". The first group is voters under 30. This group is in a unique position, since their approval of the President is the lowest of any age group, but they are also the group most open to privatization. Therefore I think the most effective critique to target at young voters would try to "bleed" Bush's poor favorable ratings into the Social Security debate. Something like this: "Social Security might need fixing, but you can't trust Bush to get it right." The second core Democratic constituency that Bush has tried to target is African-Americans, but Bush's fraudulent claims that Social Security is a bad deal for blacks based on his inability to interpret statistics (if you're feeling generous), or his willful misrepresentation of them (if you're not) ... well, let's just say that no one's buying it.

Now let's move on to the more interesting question: where should Democrats look for Republican voters that will be sympathetic to the Democratic goal of preserving Social Security? I'll start trying to answer this question by looking at results from the 2004 election broken down by both age and race. The number in parentheses is the percentage of voters that this demographic makes up, which is followed by Kerry's vote totals, followed by Bush's:

[Source: National Election Pool 2004 Exit Polls]

From this table two things become clear. First, there are few new votes to be had by persuading the small number of African-Americans who supported Bush to change their mind. Second, the traditional Democratic advantage among seniors is gone. Take a look at the historical results for voters age 60 and up:

[Sources: CNN 1996 Exit Polls, ABC News 2000 Exit Polls]

In 1996 Clinton won voters age 45-59 by a 48-41 margin, so the problem is not that a huge number of fifty-something Republicans are moving into the 60+ age bracket. This suggests that the over-60 voters can be brought back into the fold with relative ease. Recent polling data supports this hypothesis; recent polls show that even though seniors are aware that Bush's "plan" does not affect their Social Security check, they are still strong opposed to privatization. It's also worth mentioning that voters close to age 55, but not yet there, ought to be sympathetic to Democrats as well. These folks would be taken to the cleaners under Bush's Social security unreform; they pre-funded their retirement under the Greenspan reforms, but would see a large cut in guaranteed benefits.

Now that we've taken a look at the effect of race and age on voting patterns, let's examine at gender as well.

Note that the gender gap is almost perfectly consistent across all races, with the exception of elderly Hispanics, though it's quite possible that this is due to sampling error. But what is interesting is that the 4-point gender spread in the last election is small by historical standards. How did this happen? Well, in part it is due to Kerry's ability to pull in almost all male Nader voters (or a number of voters that's roughly the same). But it's also due to Bush's impressive showing among women in 2004:

And there, folks, are your security moms: the four or five percent of women who voted for Gore in 2000 but moved to Bush in 2004. As with seniors, these voters who recently defected from the Democrats will probably be the voters that can be brought back to the fold most easily. Thankfully, most polling shows that women are more sympathetic to building a strong social safety net than men are, so this should help as well.

At this point, the most promising demographics to target look something like this:

1. White women age 56+
2. Hispanic women age 45+
3. White women age 45-55
4. White men age 56+
5. White men age 45-55

We've done a decent job of narrowing the universe of target voters already, but I think we can do better. Coming next, I'll try to refine the results further by looking at income, education, and church attendance.

| | technorati

Home | Mail  | RSS

Last updated by Nicholas Beaudrot on 09:40 16 April 2005