« Parse This Paragraph and Win a Prize! | Main | Zogby: Military Wants Us To Leave Iraq »

February 28, 2006


on unions giving money to pro-labor (read: Dem) candidates, see also this piece in Times today. Punchline is that despite loss of 4 unions' membership and 25% decrease in budget, AFL-CIO is increasing its campaign spending to $40M (from $35M in 2002).

If I read it right though that is not money going to 527s, that is money directly spent organizing union members (?): The money will not be used for contributions, but to mobilize union members.

Yeah, the link in this piece is about the same thing. And the money they're talking about is the soft money, in this case from union dues. It can be used to say "vote for" or "vote against" or variations of the same when the communications are directed at union members.


Could you address the role of the marketing-style GOP targeting that was so successful last time? It seems to me they did a lot based on demographic analysis, using marketing data rather than just knocking on doors. And particularly since these white suburban swing districts are also harder to walk (or the newfangled young communities with lots of locked in apartment and condo developments, where you CAN'T walk them), I'm curious how much of the IDing was marketing data (the old Mercury Mariner owners are the most reliable Republican voters in the book, though I'd bet they're ignoring all non-White Mercury Mariner owners, which is a significant demo) and how much was old-style IDing.

Gotta run--was just shutting down--but briefly, that was a big part of how they made their GOTV efficient and avoided Dems. And that stuff was:
A. Done with hard money
B. Begun well before the election, even in early 2003, and funded through the RNC

It made their IDing more efficient and less likely to turn up a Dem. And they did some registering based on that data (no doubt many of them the socially conservative Evangelicals Rove kept talking about). ACT, on the other hand, if they did a registration drive, couldn't avoid anyone who appeared likely to be a Dem. Thus, it made ACT much less effective in swing areas; in fact, ACT mostly stayed out of swing areas, and couldn't effectively do GOTV of Dems in Republican areas.

Typo: ACT, on the other hand, if they did a registration drive, couldn't avoid anyone who appeared likely to be a Repub.

Isn't another problem for the Dems the custody of the info after the election? The RNC builds a database, but the 527 GOTV info can't be shared with the DNC. And if a presidential candidate won't share his database either, then the Dems pretty much have to start from scratch each election, no?

Yeah, but GOTV lists aren't anywhere near as valuable as fundraising and volunteer lists. You have to update the voter files constantly, especially in the high dem areas (which tend to have lots of students, renters, transients, etc.)

And one of the things Dean is doing that's almost completely below the radar screens of the DC chattering classes is to build the capacity of state and local party orgs to use, maintain and build upon their own voter file data. We had already gotten a head start on that in Michigan, but in a lot of states they didn't have anything, and that's the type of thing emphasized in the grants from the DNC to the party organizations.

Check this out:
Bill Clinton Urges Governors to Help Change Nation's Eating Habits

We had already gotten a start in MI.

Except for one lonely county that refused to play by the rules and work with the state database.


Florida Dems are getting access (for a price) to a new voter file system that's supposed to be phenomenal. It's actually a website that precinct captains will be able to log on to and update immediately after canvassing. Our county hasn't bought it yet, but will soon. Friends of mine live in a neighboring county which has already purchased the file and will begin training their volunteers soon.

To clarify: my friends will actually be teaching other volunteers how to use the system.

DHinMI interested in your take on the trustworthiness of the Federal Election Commission. McCain has been hammering that it is corrupt. Bush just made Vivica Novak's husband a recess appointment to it. Bush's other recess appointment to it was some kind of GOP election "hit man" IIRC with respect to either Florida, Ohio, or both.

John, isn't FEC bi-partisan?

As we've seen with the DOJ staff attorneys who objected to the TX redistricting plan, I suspect that for the most part the staff at the FEC is generally sound. Where the Bush appointees come in to play are the advisory decisions. It was a big concern in 2004 whether the FEC would screw the Dems on the decisions about whether they would allow the Dem-aligned 527's to do what they ended up doing. There may be similarly important decisions pending, but I'm not sure of what they are.

I'm currently doing some training of folks who plan to do para-party activities in the next couple of election cycles -- community groups that work parallel to Dem party campaigns and the unions. Some of them are going to be good enough to do the high Dem precincts with very little regular Party input, especially in urban communities of color. This kind of GOTV is not rocket science.

The good news is that some Dem levels are getting into microtargetting -- crossreferencing consumer data with voter data to find the Dem needles in some pretty confusing haystacks. The current Campaigns and Elections has a terrific, descriptive account of how Virginia Governor Kaine used microtargetting in the DC exurbs to figure out what issues resonated with which demographics (white housewives in their 30s with education, for example) and managed to win counties that Bush carried in 2004. Unfortunately the article requires subscription.

This is the sense in which McEntee is right: the 527 structure in 2004 did not lend itself to finely targetted GOTV. I worked in New Mexico and watched the Dem 527s falling over each other, flailing really. Because I was doing local candidates in the exurbs, I also saw the Rep effort, very carefully working targetted new voters using volunteers from the churches. The Dems looked like a happy fractious mob; Reps looked like dedicated, boring members of the solid middle class. We saw the results.

republican vote inducing efforts may have been more effective than democratic vote inducing efforts in the 2004 elections


i personally would be skeptical of any analysis of this issue that did not take into account the possibility,

if fact, i personally have come to think

the liklihood

that voting machines were manipulated to provide support for republican candidates in excess of actual republican voting.

i consider this the most serious problem facing democrats in the coming congressional elections.

the decision to use diebold in california may be perfectly reasonalbe

but to me it smells of anticipated manipluaton of voter totals.

DH's analysis is well taken, as far as it goes, but the parties (and their allies) are in an evolutionary arms races, and the other side has the sharper, pointier sticks.

We invested in generations of ID/GOTV strategies. They invested in generations of persuasion/conversion strategies. GOTV is minimally cumulative. ID is weakly cumulative. Persuasion is strongly cumulative, cycle over cycle.

GOP doesn't microtarget just for turnout. When they win, they win followers -- not just elections. When they lose, they still win followers (in 50 states and 435 districts).

A secondary effect of Dem's turnout emphasis is that we put most resources into districts with fewest conversion proospects, forgoing even fortuitous conversions.

The comments to this entry are closed.

Where We Met

Blog powered by Typepad