2012 Elections Conference: Presidential Campaign Politics

Panelists: Ben Bishin (University of California at Riverside), Gary Jacobson (University of California at San Diego), Drew Kurlowski (University of Missouri), and John Petrocik (University of Missouri).

Dr. Petrocik discussed trends in presidential vote intentions over the course of the 2012 campaign. Two events clearly moved the daily averages in October: the first debate produced a decline in Obama support, the series was static until Hurricane Sandy, and then the series shifts back in favor of Obama. This calls into question the widely-held notion that undecided break in favor of the challenger.

Click image to enlarge

Next, Dr. Petrocik discussed the differences in trial heats for registered and likely voters. As per usual, polls of registered voters were more favorable to the Democratic candidate than samples of likely voters. In 2012, likely voters also oscillated more dramatically than the registered voters. This is suspect, since public opinion is almost certainly not that fickle. The registered voters series is much more stable and ends much closer to the election outcome. It may be that polling outfits simply have little idea of who the likely voters are. Dr. Petrocik’s suspicion is that the differences between registered voters and likely voters is closer to a difference between likely voters and the likeliest voters.

Click images to enlarge

Dr. Petrocik’s argument is that, in the sweep of American politics, turnout is a mostly inconsequential factor in presidential elections. As seen in the figure below, there is little correlation between turnout rate and Democratic vote in presidential races:

Click image to enlarge

Dr. Petrocik closes with an analysis of issue ownership in the 2012 elections. Romney did better on economic issues: improving the economy, creating jobs, reducing deficit, and taxes. Obama was rated more highly on most social welfare (Medicare, education) and foreign policy issues. However, none of the differences are very substantial (double-digit). Dr. Petrocik concludes that 2012 was a “normal” election: most individuals voted their party with only marginal deviations on the economy or issue concerns.

Click image to enlarge

Next, Dr. Jacobson focuses on presidential effects in the 2012 congressional elections. Dr. Jacobson finds that this was the most president-centered set of congressional elections in 60 years: House/presidential vote were correlated at 0.94 and Senate/presidential vote correlated at 0.81 in 2012:

Click image to enlarge

Indeed, based only on district partisanship (measured with Cook PVI), we can correctly predict 94% of the winners in House races. Incumbency increases vote share by only 4.8%, the lowest amount in any election since 1952:

Click images to enlarge

Looking forward, Republicans appear to have a important structural advantage in House elections. This is because Democrats win by very large margins in urban areas, while Republicans are more efficiently distributed across suburbs and rural areas. This is an important advantage when party-line voting is high, and small partisan advantages are more important in deciding electoral outcomes.

Click image to enlarge

Dr. Kurlowski looks at the impact of rules in the 2012 GOP presidential primary race, and which rules changes mattered and which did not. His findings suggest that changes between proportional and winner-take-all delegate allocation schemes had little impact. Instead, the GOP Delegation Allocation Speed slowed down considerably from the 2008 rules, and less delegates were awarded early. This was the main culprit in dragging out the 2012 presidential primaries. In addition, the 2012 Republican presidential primary does not appear to be particularly divisive. Santorum voters, for instance, did not stay home in the general election. Finally, Dr. Kurlowski suggests that it may be worth looking into the intended versus actual effects of “sore loser” laws, which are designed to protect parties from spoiler candidates who lost in the primary. For example, Missouri Republicans were stuck with Rep. Todd Akin, since sore loser laws prevented his primary competitors for running as independents.

Finally, Dr. Bishin discusses Latino voting behavior in the 2012 election. Most importantly, Latinos swung even further towards the Democrats and comprise yet-larger shares of the electorate. In tandem, these trends are very beneficial to the Democratic Party.

Click image to enlarge

In 2012, there is something of a controversy about the level of Republican support among Cuban-Americans in Florida. Estimates range from 48-64%, but it is clear that a drop-off occurred:

Click images to enlarge

Moreover, the amount of Cuban-Americans as a percentage of the Latino vote in Florida is declining, and will continue to drop as younger Mexican-Americans and Puerto Rican-Americans enter the electorate:

Click images to enlarge

How can the GOP reverse the trend of declining Latino support? The most important issue data indicates that immigration doesn’t rank among the top three policy concerns of Latinos. Instead, the issues that Latinos consider most important are quite similar to those of non-Latinos: the economy, education, and health care. Dr. Bishin also notes that even Marco Rubio secured only 55% of the Hispanic vote in 2010.

Click images to enlarge

This entry was posted in 2012 Elections Conference. Bookmark the permalink.

Comments are closed.