Since there should be a focus on GOTV efforts over the next couple of months, I’m posting data on voter registration and turnout. I’ll use the census bureaus terms and data for the most part. Their population measure for this subset counts everyone living in the US over 18 years of age. Most of these residents are citizens. Some portion of citizens are registered to vote. And some smaller portion of citizens reported voting in a particular election. I’m relying on Pew Research and the US Elections Project for graphs and I’ve excerpted some of the raw census data below the fold.

This graph depicts ELIGIBLE voters only.

At a high level, looking at turnout by race, black and white voters have roughly equivalent levels of voter registration and turnout. Now some of this may be due to the fact that Obama was on the ballot in 2008 and 2012. But if you look at the trend on the headline graph, you’ll see black voter turnout was approaching white voter turnout prior to 2008. Assuming black citizens register or turnout at a lower rate than white citizens is a view that is 12 years out of date. In presidential election years, roughly 73% of both black and white citizens register to vote.

The real discrepancy is with Hispanic and Asian citizens, who have far lower levels of registration and turnout. Less than 60% of Hispanic and Asian citizens registered to vote in 2012, and under 50% reported voting. Focusing on Hispanic and Asian voters should result in higher success at registration since there is a larger proportion unregistered. For the 2016 election, Hispanic and Asian voters will form 16% of the eligible voter pool, growing each year (see graph to the right).

As an aside, I should mention that voter registration rates for eligible black citizens are actually in the 80% range. This is because 1 out of 13 black citizens are disenfranchised by felony convictions. When disproportionate felon disenfranchisement is taken into account. black citizens registered, and voted at a higher rate than white voters nationally. I’ll focus on national data, state by state data can differ (links to complete state data on registration/turnout from the census bureau for 2012 and 2014), and often in ways that do not match commonly held assumptions. For example, in Mississippi, 90.6% of black citizens reported they had registered to vote, and 82.4% of them reported voting in the 2012 election. That was 7-8% higher than the rate for white citizens. Of course, none of this should be taken to imply that we should be unconcerned about voter suppression efforts that targets black citizens.

It is worth nothing that registration and turnout rates among women are almost uniformly higher than for men, by 2-5%. Part of that may be due to felon disenfranchisement again. 5.8 million citizens cannot vote due to felony convictions. In any case, more women vote than men.

Another way to interpret this data is that if you’re out there to GOTV, you are twice as likely to run into an unregistered voter if you’re in a Hispanic or Asian neighborhood than if you’re in a neighborhood primarily composed of black or white citizens. There are regional/local differences, to be sure, but at a national level this holds. Part of the reason registration and turnout rates are lower among Hispanic voters is that they tend to be younger. 44% of all eligible Hispanic citizens are millennials. But even among millennials, Hispanic/Asian registration trails significantly. There is no single, clear explanation for this, only speculation. Language barriers may play a role, as might the fact that Hispanic and Asian voters are somewhat concentrated in non-competitive states like New York, California and Texas where voter turnout efforts are uneven. It may also come down to the lack of a “family tradition” of voting in US elections, either because parents are not US citizens, or because they fall into a demographic unlikely to be registered/vote (poorer citizens and those with less than high school education tend to vote at lower levels).

Young Hispanic and Asian citizens are least likely to be registered since they’re at the intersection of two groups that have lower than average registration rates. It’s also important to note that neither Trump nor Clinton appear to have locked up the millennial vote as yet.

This election cycle is likely to engage more Hispanic/Asian voters than in the past. Some of the derogatory statements leveled by Donald Trump’s campaign against Hispanics and recent immigrants have alienated large swaths of Hispanic and Asian citizens. So there may be a significant opportunity to capitalize on voter engagement to register younger Hispanic voters in Texas. If that comes to pass, it should validate my take that Trump is unlikely to win Texas by more than 5%.

Hispanic identification with the Republican party is around 21%, that’s 7% below the already low numbers in 2012. Particularly in states like Florida (where some polls suggest under 15% of Hispanic voters are for Trump), the lack of Hispanic support may prove to be decisive.

The number of Asian voters identifying with the Republican party has fallen as well, to 15% earlier this year. There is one group of Asian-American voters that does appear to support Trump at slightly higher rates, Indian-Americans. That is partly is driven by Indian-Americans who identify with Hindu-nationalist politics in India and see echoes in Trump’s rhetoric (including his anti-Muslim statements).

More detailed data/tables can be found at census data for 2012 election (Reported Voting and Registration by Sex, Race and Hispanic Origin):

As most people know, voter registration and turnout falls off in mid-year, non-presidential elections.

In particular, you can see fewer black citizens registered to vote in 2014 (63%), as opposed to white citizens (68%). That is possibly driven by the fact that Obama wasn’t on the ballot along with relative age (older voters tend to be more reliable in mid-term elections). Hispanic and Asian citizen registration remained lower, at or under 50%. Turnout rates also dropped remarkably, with 45% of white citizens turning out while roughly 40% of black citizens did. Hispanic and Asian turnout was below 30%.

Census data for 2014 election (Reported Voting and Registration by Sex, Race and Hispanic Origin, for States).