Who Voted in 2016?
We now have the breakdown from the Secretary of State’s office on who voted during the 2016 general election. I took the information I compiled on who voted in 2014, and updated it to reflect this year’s results. In short, voting by blacks and whites is down, while voting by Asians and Hispanics is up. Within each group, a larger percentage of voters went to the polls for all groups except for blacks, which declined slightly. Two years ago, I examined how different ethnic groups voted compared to 2010. I’ll do roughly the same this year, except comparing the results to 2012, the previous presidential election.
First, a comparison of the percentage the vote of each racial group contributed to the total, over the last four elections:
| Year | Asians | Blacks | Hispanics | Other | Unknown | White |
|---|---|---|---|---|---|---|
| 2016 | 1.6% | 27.6% | 2.1% | 1.1% | 6.6% | 61.0% |
| 2014 | 0.8% | 28.7% | 1.0% | 0.7% | 5.2% | 63.5% |
| 2012 | 1.0% | 29.9% | 1.3% | 0.9% | 5.5% | 61.4% |
| 2010 | 0.6% | 28.3% | 0.7% | 0.7% | 3.4% | 66.3% |
The biggest gain from 2012 was in the Hispanic vote, which increased by 0.8%, followed by the Asian vote, which grew by 0.6%. The share of the black vote decreased by 2.3%, and the share of the white vote was down 0.4%. One reason for a lower black vote would be that Barack Obama was not on the ticket. For whites, the decrease is partially due to over 77,000 fewer registered white voters, and because minorities generally raised their share of the vote.
The next table shows the relative strength of each racial group over the past four elections, based on the number of registered voters for each group. Again, the number of registered Hispanic voters increased by almost 36,000, increasing that group’s share. The Asian share also increased, while the percentage of registered black voters remained almost the same. 77,295 fewer white registered voters contributed to that group’s reduction in the share of registered voters by 2.5%. Also worth noting, is an increase of the number of voters whose race and/or sex is unknown. Many believe that most of those in the Unknown category are minorities.
| Year | Asians | Blacks | Hispanics | Other | Unknown | White |
|---|---|---|---|---|---|---|
| 2016 | 1.8% | 30.1% | 2.3% | 1.2% | 7.8% | 56.7% |
| 2014 | 1.4% | 30.1% | 1.8% | 1.0% | 7.7% | 58.0% |
| 2012 | 1.4% | 30.0% | 1.7% | 1.1% | 6.7% | 59.2% |
| 2010 | 1.3% | 29.2% | 1.5% | 1.0% | 5.2% | 61.8% |
For each group, what percentage of its registered voters actually showed up at the polls? The next table holds that information:
| Year | Asians | Blacks | Hispanics | Other | Unknown | White |
|---|---|---|---|---|---|---|
| 2016 | 65.7% | 68.6% | 67.2% | 65.6% | 63.6% | 80.4% |
| 2014 | 28.8% | 47.8% | 28.0% | 36.3% | 34.1% | 55.0% |
| 2012 | 54.8% | 72.6% | 56.4% | 61.3% | 59.8% | 75.7% |
| 2010 | 26.2% | 50.4% | 25.5% | 34.5% | 34.2% | 55.9% |
Perhaps due to the unusual nature of this election, a higher percentage of registered voters in each group showed up at the polls compared to 2012, with the exception of blacks. Asians and Hispanics were each up around 11%, while an amazing 80.4% of whites voted, up almost 5 points. Black participation was down 4 points statewide, to 68.6% of RVs. In the three counties that flipped from Republican to Democrat, however, a substantially higher percentage of blacks went to the polls, while the number of whites were just over the statewide average. In Cobb County, 73.0% of registered blacks voted. In Gwinnett County, it was 73.5%, and in Henry County, it was 76.7%.
Looking at statewide averages tells us a lot, but looking at changes in individual counties can tell us more. This table shows how voting patterns have changed over the last six years in Douglas, Gwinnett, and Ware counties:
| Asians | Blacks | Hispanics | Other | Unknown | White | Douglas County | |||||||
| 2016 | 406 | 0.7% | 23,514 | 41.0% | 1,237 | 2.2% | 682 | 1.2% | 4,128 | 7.2% | 27,427 | 47.8% | |
| 2014 | 152 | 0.3% | 15,271 | 39.6% | 404 | 1.1% | 301 | 0.8% | 2,211 | 5.7% | 19,842 | 51.5% | |
| 2012 | 293 | 0.5% | 21,913 | 39.5% | 780 | 1.4% | 504 | 0.9% | 3,495 | 6.3% | 28,507 | 51.4% | |
| 2010 | 130 | 0.3% | 13,839 | 36.5% | 324 | 0.9% | 251 | 0.7% | 1,776 | 4.7% | 21,618 | 57.0% | |
| Gwinnett County | |||||||||||||
| 2016 | 16,510 | 5.1% | 84,934 | 26.1% | 18,739 | 5.8% | 7,903 | 2.4% | 27,254 | 8.4% | 170,041 | 52.3% | |
| 2014 | 4,903 | 2.4% | 50,866 | 25.4% | 5,692 | 2.8% | 3,623 | 1.8% | 13,362 | 6.7% | 122,021 | 60.9% | |
| 2012 | 9,915 | 3.3% | 76,107 | 25.6% | 11,127 | 3.7% | 6,809 | 2.3% | 20,311 | 6.8% | 173,234 | 58.2% | |
| 2010 | 3,927 | 2.0% | 44,917 | 22.7% | 3,868 | 2.0% | 3,014 | 1.5% | 11,328 | 5.7% | 130,540 | 66.1% | |
| Ware County | |||||||||||||
| 2016 | 45 | 0.4% | 2,579 | 21.5% | 71 | 0.6% | 44 | 0.4% | 280 | 2.3% | 8,977 | 74.8% | |
| 2014 | 13 | 0.2% | 1,430 | 20.9% | 15 | 0.2% | 11 | 0.2% | 101 | 1.5% | 5,264 | 77.0% | |
| 2012 | 35 | 0.3% | 2,939 | 24.5% | 51 | 0.4% | 32 | 0.3% | 186 | 1.6% | 8,749 | 73.0% | |
| 2010 | 16 | 0.2% | 1,406 | 19.5% | 19 | 0.3% | 15 | 0.2% | 47 | 0.7% | 5,690 | 79.1% | |
Whites make up only 47.8% of the registered voters in Douglas County, and that number tracks pretty well with its percentage of the actual vote. While the number of Asian and Hispanic voters increased since 2012, the increase of 1,600 black voters and a decrease of over 1,000 white voters was enough to bring the white vote share below 50%.
In Gwinnett, the shares of Asian and Hispanic votes each rose above 5% of the total, a sign of the growing diversity in the state’s second largest county. Despite having only 3,193 fewer white votes than in 2012, the share of the total votes from whites declined by almost 6%. Yet with whites being only 48.4% of registered voters, they still had a majority of those that showed up on Election Day.
In South Georgia’s Ware County, the number of registered black voters actually declined by just over 200. That was reflected in the share of the black vote, which declined by 3 percentage points. With the number of Asian and Hispanic votes increasing by a negligible amount, whites were able to increase their percentage share of the vote by almost two points.
I updated the tables I made two years ago to reflect the 2016 numbers for the different groups.
Loading...
Reload document 
Open in new tab And finally, the disclaimer that there’s now way to tell from the data provided by the Secretary of State’s office how people actually voted. Nonetheless, it’s clear that the changing voter demographics will impact whether Georgia remains a red state, turns purple, or becomes blue.
About Author
Jon Richards
Add a Comment
You must be logged in to post a comment.
Interesting figures….Clinton was able to get to within 5 points of Trump here in Georgia even with the reduced black turnout, suggesting the closer margin (compared with 2012 Romney/Obama) was due to GOP defections from Trump, not because of higher black turnout.
Also, while Jon Richards a few weeks ago provided a link to a website with Trump/Clinton numbers by congressional district, I haven’t seen any similar for our U.S. Senate race, so here we go below (99% complete, mainly just missing some hard to calculate split precincts in Cobb):
Isakson Barksdale Buckley
CD 1 (Buddy Carter) 60% 36% 4%
CD 2 (Sanford Bishop) 47% 50%+ 3%
CD 3 (Drew Ferguson) 67% 29% 4%
CD 4 (Hank Johnson) 26% 70% 3%
CD 5 (John Lewis) 20% 76% 4%
CD 6 (Tom Price) 58% 37% 5%
CD 7 (Rob Woodall) 56% 39% 5%
CD 8 (Austin Scott) 66% 31% 3%
CD 9 (Doug Collins) 78% 18% 5%
CD 10 (Jody Hice) 64% 32% 4%
CD 11 (Barry Loudermilk) 66% 29% 5%
CD 12 (Rick Allen) 60% 37% 3%
CD 13 (David Scott) 30% 67% 3%
CD 14 (Tom Graves) 74% 21% 5%
Clearly the most polar opposite districts in Georgia are CD 5 (John Lewis—76% Barksdale) and CD 9 (Doug Collins—78% Isakson). In his 30 years in Congress, Lewis only once has dipped below 70% against a Republican (“held”, if you could say so, to “only” 69% in 1994).
One thing that struck me with the congressional results above was how close Bishop’s district was–Barksdale just barely clearing a majority in that majority-black district. A Democrat isn’t likely to win statewide if he or she struggles to win the 2nd District. And the closeness there could be explained by drop in D presidential voting—in 2008, Obama took about 154,000 votes in what is now CD 2 (the district of course had different lines back in 2008, prior to 2011 redistricting). Clinton, however, only got about 136,000 there this time—a near 18,000-vote decline from Obama 2008—while Trump’s totals in CD 2 were virtually the same as what McCain got there in 2008.
Yeah, you go county by county in Rural Georgia and you can see the Rural Black Dem drop off this year. For example:
Hancock County:
2012: D: 3308 R 769
2016: D 2701 R 843
Taliaferro County:
2012: D 636 R 323
2016: D 545 R 349
So that’s how you ended up with Clinton winning CD 2 55-43 while Obama won it 59-41 in ’12. Not so much a lot of crossover, or hidden Rs (though some turned out more; a return of the kind of sporadic ‘Darn the system!’ Perot voter’), but Dem turnout dropping while the Rs increasing only a little bit. It’s one reason Trump was able to ride out losing the Atlanta Metro by almost 160,000 votes: He did better in the out-state in raw margin than even Bush ’04, winning the rest of the state by almost 369,000 votes. Obama, buoyed by rural black turnout, only lost the rest of the state by 279,650 in ’12 and 253,000 in ’08. Almost 40,000 Democratic votes disappeared in the out-state from ’12 to 16. That wouldn’t have been enough to swing the state, obviously, but if she’d done as well in the outstate margin as Obama 2012 she’d have lost the state by a little under 3%–closer than North Carolina.