(Harry Cabluck/AP Photo)

Despite the favorable political environment for the GOP heading into the 2014 midterms, races in a number of states that were expected to be easy Republican wins earlier in the cycle are proving to be anything but. In this post, I examine two such contexts, Georgia and Kansas, to assess how Latino voters may affect gubernatorial and U.S. Senate outcomes in these states.

Clearly, when one thinks of states with growing Latino populations rarely do Georgia or Kansas come to mind. To be sure, the 2013 U.S. Census estimates of the Latino segments of the Georgia (9.2 percent) and Kansas (11.2 percent) populations are below the Latino national share (17.1 percent). However, between 2000 and 2010 the Latino population in Kansas grew by 59 percent (188,252 to 300,042) and in Georgia the number of Latinos increased by 96 percent (435,227 to 843,689).

Yet, because of the relative youth of the Latino population and its non-citizen component, Latinos constitute four percent of the eligible electorate in Georgia and just over six percent in Kansas. As a consequence of the limited efforts by the parties to engage and mobilize these voters, even fewer Latinos are registered to vote in either state. Specifically, data from the Latino Decisions and L2’s micro-targeting model indicates that 52 percent of Georgia’s and 54 percent of Kansas’ eligible Latino voting populations are registered to vote. Still, as compared to 2000, Latino voter registration has increased by 575 percent in Kansas and by over 400 percent in Georgia and according to data from the U.S. Census, in Kansas 61 percent and in Georgia 81 percent of registered Latinos voted in 2012.

So, what might this mean for the 2014 gubernatorial and U.S. Senate elections in those states? As of late, both Georgia and Kansas have been reliably Republican. In Georgia, the GOP has controlled the governorship since 2003 and the state last elected a Democrat to the U.S. Senate in a 2000 special election. In its history, just five non-Republicans have represented Kansas in the U.S. Senate (three Democrats and two Populists in the late 1880s), but none since 1932. However, in the last 60 years the Kansas governorship has regularly alternated between Democrats and Republicans. More recently, in the 2012 presidential election Romney carried Kansas with 60 percent of the vote and 53 percent in Georgia.

Even with this history, many forecasting models have the Democratic candidates favored, albeit slightly, to win the Kansas governorship and the U.S. Senate race in Georgia, while the Democratic gubernatorial candidate is in striking distance of the incumbent Republican. Meanwhile, long time Republican Kansas Sen. Pat Roberts is in the fight of his life against an insurgent independent candidate, Greg Orman.

Certainly some of the competitiveness of the Georgia races is a function of that state’s changing demography and the presence of two Democratic legacy candidates (Senate candidate Michelle Nunn is the daughter of former U.S. senator Sam Nunn, and gubernatorial candidate Jason Carter is the grandson of a President Jimmy Carter). In Kansas, Gov. Sam Brownback’s woes seem to stem from conservative overreach, while Pat Roberts is the latest example of a long time incumbent being punished for “going native” and losing touch with his home state.

Regardless, all four elections are very close and this, in turn, means that outcomes can be determined by larger than expected turnout of voting blocs that break strongly for one party over the other. Certainly, Latinos in both Georgia and Kansas are potentially one such group, particularly since these voters have received little attention in either state. Indeed, only a handful of the public polls released in Georgia and Kansas include a crosstab for Latinos.

GA_and_KS

Moreover, estimates of predicted turnout and vote choice from Latino Decisions and L2’s micro-targeting model that are presented in the table above suggest a group of voters whose preferences and political engagement are very much inchoate. That is, unlike in states such as Colorado and Nevada, where the Democratic Party has made substantial investments in engaging and mobilizing Latino voters, Latinos in Georgia and Kansas are much less Democratic oriented and most have a low probability of voting.

This is particularly the case in Georgia where independents are by far the modal category and among Latinos with a high probability of voting, there are 55 percent more likely Republicans than Democrats. In Kansas there are roughly twice as many potential Democratic than Republican voters. However, the Democratic advantage among high probability voters is minimal as over 70 percent of likely Democratic voters have a lower probability of voting as compared to half of likely Republican voters.

More generally, in both states the single biggest segment of the Latino electorate is voters who are unaffiliated with either party and who have low probabilities of voting. To this end, the National Council of La Raza and their Georgia and Kansas affiliates are conducting non-partisan get-out-the vote efforts targeting Latino voters who are less likely to vote in a midterm election.

To date, the advocacy of Senator Roberts and Georgia Governor Nathan Deal for policies that are perceived as anti-immigrant by most Latinos have had little consequence. However, if the turnout of Latino voters is larger than expected in Georgia and Kansas in 2014, then that could very well change. There also are the long-term effects that Latino outreach can engender. After all, voting is a habit and among voters with weak partisan attachments, their partisan preferences are formed in response to information that is salient in the immediate political environment that can resonate for years to come.