Election 2016 Forecast / Identity Politics: Hispanic Community
The Hispanic community has proven to be one of the Democrat’s most valued assets. In comparison to the African-American community, Hispanics tend to vote Democrat at slightly lower rates but at approximately 70% they still produce an overwhelming advantage for Democrats. Hispanic turnout is consistently lower, however, than other major demographic groups.
As for 2016, Hispanics should, judging from the headlines, rush towards the Democrats this election to an unprecedented extent. Though immigration has been a topic in previous elections, it has taken center stage by many accounts in 2016. A superficial analysis would likely conclude that turnout should skyrocket and the percent vote within this demographic for Democrats should hit records. However, as we will see in this post, there is little evidence to support this superficial conclusion. In fact, Hispanics are not showing high relative interest in this election which does not bode well for an assumed upward spike in turnout. Also, various polls point to Trump actually doing slightly better than Romney with this demographic. In short, this does not appear to be the election that the Hispanic community pushes Democrats into the White House.
Let’s start by looking at the Hispanic community’s historical voting pattern.
Chart 1: Adjusted Percent of Hispanic-Americans Voting for Democratic Presidential Candidates
Source: Cornell University
Note: this data is adjusted so that we can compare it across elections. We are most interested here in comparing the two major parties and how different groups voted for them. The adjustment is to take the actual percent voting for Democrats and dividing by the sum of percent voting for Democrats plus Republicans. This takes out the impact of third parties.
Judging from the previous chart, it looks like Hispanics could lean towards either party in 2016 as they seem to have been, in 2012, around the middle of the demographic’s historical range. Again, going back to the headlines we can assume that the Democrats would benefit and that the percent voting Democrat could move towards the upper end of the range.
In a highly unusual occurrence, the polls actually show that the Hispanic community is not supporting Clinton any more than they did Obama. In fact, taking an average of four polls (PPP, Economist/YouGov, McClatchy-Marist, and Quinnipiac University) shows that 69% of the Hispanic community intends to vote for Clinton, which is slightly above 2008 levels but below 2012 levels. In other words, it does not seem like the Democrats will get a major boost this election from the Hispanic community increasing their vote share towards them.
Backing up this conclusion is data from Pew Research. It shows intention to vote Democrat levels from the Hispanic community during the summer of election year.
Table 1: Pew Research, Hispanic Community Intention to Vote Democrat versus Republican
| Intend to Vote Democrat |
Intend to Vote Republican |
|
| 2008 |
66% |
23% |
|
2012 |
69% |
21% |
| 2016, two way race |
66% |
24% |
| 2016, three way race |
58% |
20% |
Source: Pew Research
This data shows that the situation in 2016 is just about the same as in the previous two elections at approximately the same periods. If anything, the Republicans look slightly better in 2016 in relative terms, especially when including the Libertarian Party into the mix. In short, this data does not show the pro-Clinton blow-out that many assume. It shows that even with all of the volatile headlines, the Hispanic community’s voting intentions more or less replicate those of previous elections. Furthermore, the Pew data tends to support the conclusion from the other four cited polls that indicate a slightly improved Republican outlook this year.
Another interesting point from the PEW research that I do not see very often is how the Hispanic community tends to vote differently depending on what language is dominant. For those whose dominant language is “Bilingual/Spanish” the Clinton/Trump split is 80% to 11% whereas if the dominant language is “English” the split is only 48% to 41%. This clearly implies that recent immigrants tend to heavily favor Clinton whereas Hispanics in the US for longer periods of time tend to be more evenly split. A different report from PEW shows that the percent of foreign-born Hispanics in the US who “mainly use English” is only 5% whereas for a 3rd generation or higher Hispanic in the US this same figure goes to 76%. The main point here is that though generally clumped together, the Hispanic community is very diverse with time in the US perhaps being one of the more important indications of voting preference.
Though we will not explore this topic further at this stage, perhaps a better way to analyze the Hispanic community is to do so by integrating language preference and/or time in the US as they seem to greatly influence party selection. For Democrats in 2016, the real push should be turning out the vote not for all Hispanics but for targeted groups of Hispanics. Again, assuming Democrats focus on Identity Politics, it is a little shocking that a Hispanic was not chosen as the VP candidate as this could have really improved election chances via Hispanic turnout.
As for general Hispanic community voter turnout, it does not seem like there will be a significant increase. Given the headlines, it would seem like a surge in turnout would make sense. However, publicly available data does not show an expected surge.
Chart 2: Percent of Hispanic-Americans Voter Turnout in US Presidential Elections by Gender, 1984 – 2012
Source: Cornell University
The previous chart shows that Hispanic voter turnout for both men and women hit record highs in 2008. As with many demographic groups there was definitely a feel-good factor at the time which brought many to the polls. Although relatively high by historical terms turnout still appears fairly low as compared to the African-American and White communities. Looking at the data superficially, you might expect there to be significant upside, meaning that given the right conditions Hispanic voter turnout could move much higher to be more in-line with other groups.
Lower voter turnout could be somewhat cultural though which would make it harder to increase. This analysis might be stretching things a bit, but in Latin America voter turnout in countries where voting is not mandatory is about at the same level as the Hispanic community in the US.
Table 2: Voter Turnout in Latin American Countries during Presidential Elections with no Mandatory Voting Laws
|
Voter Turnout in Last Presidential Election |
|
| Columbia |
52% |
|
Guatemala |
50% |
| Chile |
46% |
| Average |
49% |
Source: International Institute for Democracy and Electoral Assistance
These figures show that the average voter turnout during the last presidential election was 49%. If we take the average the last four elections, the result is a slightly lower turnout, but essentially in the same general range. Considering that the Hispanic community turnout for the US election in 2008 was approximately 50% and in 2012 was 48%, it seems like a Latin America average (for countries that do not have mandatory voting laws) of 49% can explain a lot.
Additionally, voter interest and enthusiasm within the Hispanic community does not seem to be that high this year in comparison to the rest of the populace, though in absolute terms it is. In fact, by every metric cited here, Hispanics appear less interested and enthusiastic than the general populace.
Table 3: Net Interest and Enthusiasm Levels of General Public and Hispanic Community regarding 2016 US Presidential Election
| General Public | Hispanic Community | Hispanic “Enthusiasm Deficit” | |
| (Net) Thought given to election | 61% | 37% | 24% |
| (Net) Election really matters | 53% | 40% | 13% |
| (Net) More interested in politics | 44% | 42% | 2% |
| (Net) Closely following election news | 71% | 35% | 36% |
| Average | 57% | 39% | 19% |
Source: PEW Research
Note: For each of these a “Net” calculation was used. For the “Thought given to election”, the calculation was to take the “Quite a lot” response and subtract the “Some/little/none” response to get to a “Net” answer. The same thing was done with the other questions such that the negative response was subtracted from the positive response.
The previous table shows that the US is generally very interested in and enthusiastic about the 2016 election in that there are many more people thinking about it, believing the outcome really matters, viewing it as interesting, and following it closely than those taking the opposite point of view. However, the Hispanic community is less so than the rest of the populace. This is a little surprising considering that immigration has emerged as such a hot topic. Regardless, the level of interest should provide a good indication for voter turnout. An average Hispanic “Enthusiasm Deficit” of 19% indicates that the expected voter turnout will likely be significantly below the rest of the populace. As the actual voter turnout for the Hispanic community in 2012 was roughly 16 percentage points below the aggregate voter turnout, an “Enthusiasm Deficit” of 19% does not seem to indicate a rapid increase in relative turnout.
This is very much based on a back-of-the-envelope type analysis, but from what I have seen it should be more or less accurate. In other words, if the Hispanic community’s 2016 voter turnout was to spike higher, it would follow that we would be able to detect significantly higher voter interest as compared to other demographics. It seems like there is a considerable enthusiasm deficit for the Hispanic community in comparison to other demographics. So, until there is stronger evidence implying a higher voter turnout, turnout is forecast to remain near the same as the last few elections.
Summarizing, this does not appear to be the election that the Hispanic community will come out in unusually high numbers and help to push the Democratic candidate over the top. Voter enthusiasm within the demographic appears to support turnout somewhere around historical levels and current polls tend to show that Democrats will receive approximately the same, or even slightly lower, percent of the community’s vote.