by Alan Abramowitz, Alben W. Barkley Professor of Political Science, Emory University
With only five weeks left until the 2006 midterm election, political analysts remain divided about the Democrats’ chances of regaining control of the House and Senate. While some indicators such as the generic ballot continue to show a strong Democratic advantage and the overwhelming majority of competitive races involve GOP seats, President Bush’s approval ratings have climbed a bit in recent weeks, giving Republican strategists and candidates renewed hope. In addition, many analysts believe that Republicans have an ace in the hole going into the final days of the campaign—their party’s well-oiled get-out-the-vote (GOTV) machine. Despite polls showing Democratic voters more energized than Republican voters, these analysts believe that the Republicans’ vaunted “72-hour program” will give the GOP a critical edge on Election Day.
But is the Republican GOTV machine really as good as it’s made out to be? Claims about the effectiveness of the GOP’s 72-hour program are based largely on the results of the 2004 presidential election, and especially on what happened in the large battleground states. And no state is cited more frequently to illustrate the effectiveness of the Republican 72-hour program than Ohio.
Does the evidence from Ohio actually support these claims about the effectiveness of the Republican 72-hour program? It is true that the number of Republican votes in Ohio increased dramatically between the 2000 and 2004 presidential elections, going from 2.35 million in 2000 to 2.86 million in 2005. That’s an increase of more than 500,000 votes, or 21.7 percent. Sounds impressive. But wait a minute—the number of Democratic votes in Ohio rose from 2.19 million in 2000 to 2.74 million in 2004. That’s an increase of more than 550,000 votes, or 25.4%. So which party’s GOTV program was more effective?
Okay, but everyone knows that the real key to a GOTV campaign is getting out your base. So how effective were Republicans and Democrats at getting out their base voters in Ohio? Well, one way to gauge this is to focus on each party’s strongest counties. In Ohio there were 14 counties that George Bush carried by a margin of at least 10,000 votes in 2000 and 6 counties that Al Gore carried by a margin of at least 10,000 votes. Between 2000 and 2004, the number of Republican votes in the 14 strongly Republican counties increased by 163,000, or 23.4%. That’s pretty impressive. But the number of Democratic votes in these 14 GOP counties increased by 113,000, or 26.4%! So while the Republican plurality in these counties increased by about 50,000 votes, the percentage increase in Democratic turnout was actually greater than the percentage increase in Republican turnout.
.
And what happened in the six strongly Democratic counties? The number of Democratic votes increased by 193,000, or 24.9%, while the number of Republican votes increased by 97,000, or 21.7%. So not only did the size of the Democratic plurality increase by almost 100,000 votes in these six counties, but the percentage increase in Democratic turnout was greater than the percentage increase in Republican turnout.
One of the most difficult things to accomplish in a GOTV campaign is to mobilize your own party’s supporters without also mobilizing the opposing party’s supporters. Not only did Democrats do a better job of turning out their own voters in Ohio, but they also did a better job of not turning out opposition voters. Based on the actual turnout data, it appears that the GOP’s vaunted 72-hour program was actually less effective than the Democratic Party’s GOTV effort in Ohio. On November 7th, that Republican ace-in-the-hole may turn out to be a joker.
Latest Research from:
tds-wwc-roundtable
Thanks for finding good empirical data that bears on this issue.
I’ve heard, though, that one of the strengths of the GOP turnout machine is in microtargeting. They (it’s said) can effectively turn out the Republican voters in mixed or Democratic leaning areas. Looking at heavily D / heavily R areas as a sign of the effectiveness of D and R turnout machines may not be illuminating with respect to that.
You might be right about the absolute number of additional voters the Democrats were able to add in the most favorable counties compared to the Republicans, but you must remember that the Republicans were generally working with counties with much smaller population bases. In other words, their success lies in the ability to increase the percentage of favorable votes that fall into their column. It is no mean feat to accomplish that in some of the counties involved in Ohio – generally pretty rural and not easy to contact potential voters. They’ve clearly crafted a technique which plays to their base.
“n Ohio there were 14 counties that George Bush carried by a margin of at least 10,000 votes in 2000 and 6 counties that Al Gore carried by a margin of at least 10,000 votes.”
Since many of Bush’s strongest counties would be rural with small populations (and Gore’s strongest would be the heavily urban ones like Cuyahoga), is this really a relevant way to measure performance in each candidate’s strongholds? How would these numbers look if we examined counties that each candidate won by 10% or more?
Interesting post, with lots of data that I hadn’t seen before. But just to play devil’s advocate, this post only focuses on Ohio in 2004. Since then, the outside groups like America Coming Together (ACT), which were pretty much relied upon by the Democrats to provide the GOTV, have disbanded. ACT spent over $85 million on the 2004 election with multiple offices and dozens of paid canvassers in every swing state. Without ACT in 2006, the Democrats can’t heavily rely upon 527s to provide the GOTV operations. Plus, the DNC under Howard Dean is spending more money on the “50 state strategy,” and less on targeting specific states and races for GOTV.
In addition, there have been instances since Ohio in 2004 that are pointed to by those who believe that the GOP GOTV machine is effective. Examples include the special election in Duke Cunningham’s CA-50 district earlier this year, where the GOP managed to hold on to the disgraced congressman’s seat. Another example is the recent Senate primary in Rhode Island, where the RNC and NRSC dumped a great deal of resources into protecting Lincoln Chaffee from being upset by the conservative Steve Laffey. Since Laffey was far too conservative to get elected in Rhode Island, the RNC and NRSC needed to keep Chaffee as the candidate, and they succeeded by a far larger margin than expected.
It is hard to tell whether the GOP GOTV machine is “a joker” as the post suggests, or something Democrats should fear, as is suggested in the recently published book “One Party Country: The Republican Plan for Dominance in the 21st Century.” The answer is probably somewhere in the middle.