January 5, 2021, was an excellent night to be a Democrat. A historically conservative state, in the heart of the Deep South, sent a Jewish filmmaker and an African-American pastor to the United States Senate, effectively giving Democrats control of the chamber. Many election analysts were shocked that this could happen, given the historical perception that Democrats do poorly in Georgia runoffs.
But a minority of analysts such as Lakshya Jain as well as Elections Daily‘s own Kraz Greinetz argued that black voter turnout, often described as lax relative to white voter turnout, typically punches above its weight in runoffs – and moreso than rural white turnout, who are often the undiscussed low-propensity voters.
Taking Ideas from Alabama
The simplified summary of why Democrats won goes as follows: black turnout in urban and rural areas did not slip as much as rural white turnout. Where has this pattern been seen elsewhere in recent history? The answer lies in Georgia’s next-door neighbor, the state of Alabama. In 2017, the special election to fill the Senate seat of then-Attorney General Jeff Sessions drew national attention. Republican nominee Roy Moore proved to have enough skeletons in his closet that one would think him a grave robber. Ultimately, Democratic attorney Doug Jones would win by just over 20,000 votes in a stunning upset.
Jones flipped a number of highly racially polarized counties in the middle and southern parts of the state. In fact, he improved virtually everywhere. The first federal statewide election won by a Democrat in over two decades, it was a photo finish as heavily black counties in central Alabama pulled Jones over the line.
Political consultant Matthew Isbell’s map of the turnout comparison between the high-turnout presidential race in 2016, and the low-turnout special in 2017 (which in terms of the dynamics of the race, is very similar to the Georgia runoffs of 1/5) is pictured below. It may be easy to think that this was due to Jones’ strength as a candidate, which was indeed real, but Isbell’s map below shows that it was largely a product of reduced turnout among key Republican areas.
Looking at the areas which punched the most weight, they were:
- Jefferson County (Birmingham)
- Madison County (Huntsville)
- A large swath of west-central Alabama, where the bulk of the state’s rural black population resides.
Where were the biggest turnout drops? North-central Alabama, the heavily white, heavily rural, and heavily conservative swath of land that roughly corresponds to the state’s 4th congressional district. This is one of the most Republican areas in the country, let alone the state. It clocks in as one of the few districts in the nation to give over 80% of the vote to the Republican.
Granted, Roy Moore was a dumpster fire of a candidate who alienated tens of thousands of core Republican supporters, but his antics were another reason for an already difficult-to-persuade group of rural white voters to not turn out and vote. Suboptimal Republican turnout in the southeastern Wiregrass region can also be blamed for Moore’s loss too, in this exact same manner.
Old Habits Die Hard in Georgia
In Georgia, the turnout patterns followed a similar pattern. The first map shows the turnout dropoff by congressional district between the November presidential race (a high-turnout election) and the January runoffs (a comparatively lower-turnout election), as mapped by Georgia political analyst Niles Francis.
All districts saw a drop, but where were the highest drops located? Districts 8 (Warner Robins, a conservative military city and a swath of archconservative southern Georgia), and 9 and 14 (ultraconservative north Georgia) had the biggest dropoffs. These districts are all heavily rural and heavily white. The lowest dropoffs were located in the Atlanta metro, specifically in districts 5 and 6, which have large concentrations of black voters and increasingly urban liberal white voters.
The second map shows this on a precinct and county level, mapped by Naperville Politics on Twitter, and the turnout lag is staggering. The color is darkest in the middle of the state where most of the state’s rural black voters lie, and lightest in the far northwest and the far southeast, home to some of the most conservative Republican areas of the state. Since race, demographics, and education are such strong predictors of party vote preference in an inelastic turnout-based state like Georgia, analysts like Lakshya Jain were able to piece together that Democrats had banked a virtually insurmountable early-vote lead through their turnout operations – one that piecemeal Republican counteraction efforts would not be able to overcome.
These same general principles of disparity in racial voting patterns in Georgia runoffs have held up historically as well. The last crucial runoff Senate election of this importance was in Georgia in 1992, when incumbent Democratic Senator Wyche Fowler lost in a runoff to Republican Paul Coverdell, after winning a plurality less than 50% in the first round.
In 1992, Georgia Democrats were far more reliant for their coalition on rural conservative whites than they are now. As such, they suffered the same propensity issues that Republicans in the state now do. Rural areas noticeably got more Republican between the general and the runoff, all the while turnout cratered. Where did it crater the most? Once again, we find our worst turnout dropoffs in the far southeast and in the north. The highest-propensity areas are in the middle of the state, an area that even now has a high black population, but was appreciably less white then. The lightest shades of green on the second map are in the southwestern part of the state, another heavily black area of the state. Once again, our turnout patterns hold.
The difference between Ossoff and Warnock’s wins and Fowler’s loss is how crucial rural Dixiecrat-legacy conservative whites were to the Democratic coalition. Fowler, in losing his turnout edge, did not have anywhere else to make up ground against the ascendant high-propensity Republican strength in the Atlanta suburbs. Rural whites too, were not the 90% Democratic voting bloc that they were in the 1970s; around the south Georgia city of Tifton, you can see a circle of counties swinging Democratic from the general to the runoff, this is an indication that a lot of low-propensity voters there would have leaned Republican.
Near the Savannah River in northeast Georgia, and the Satilla and Altamaha Rivers in southeast Georgia, the red swings indicate lots of Democratic-leaning low-turnout rural whites. Where local identity and parochialism ruled partisan politics almost 30 years ago, now it is solely race. The propensity edge thus has appeared to shift towards the Democrats, the more liberal party, in an ironic twist of fate that would befuddle the original creators of the runoff system in the state.
The Dynamic of Runoffs
Runoffs were implemented post-civil rights era to protect the political interests of white conservatives (who at the time were virtually entirely Democratic) against an ascendant black political class. At the time in Georgia, race (white/black for simplification) and party (Democrat/Republican) were a binary set of axes, with the center of gravity concentrated in favor of white conservatives. The runoff was designed so that in the event of a white-aligned candidate facing a black-aligned candidate, the majority of racially-polarized Georgians who were and are white would unite behind the white-aligned candidate. At the time, when the Republican Party was virtually non-existent in the state, this made sense to keep political power in the hands of conservative white Democrats.
Fowler’s loss in 1992 though marks a shift, in that an ascendant high-propensity suburban Republican coalition made its appearance in Georgia politics. Especially in midterms, Republican turnout in Atlanta’s suburbs and exurbs, combined with waning rural Democratic support, slowly shifted the state into the red column. Knife-fight elections for Governor and Senate in 1994, 1996, 1998, were combined with a killing blow to Georgia Democrats in 2002. They lost the governorship, a Senate seat, and the legislature for the first time in a century to mark the supremacy of the new suburban-based coalition.
The Trump Realignment
Conservative whites still controlled the state, only this time, they were Republicans. The center of gravity was still suburban-based, though, so through numerous cycles, such as 2008, 2010, and 2014, Georgia Republicans managed to sweep statewide offices. The vote-rich Atlanta suburbs had slowly been moving leftward, but Republicans managed to lock down the region, until the worst nightmare of suburban Georgia Republicans stumbled onto the political stage: Donald Trump.
Trump blew Georgia politics wide open because he only won the state by five points with sky-high rural turnout that shattered previous Republican margins. Numerous rural counties in white parts of Georgia posted Republican vote shares above 80 and 90 percent. At the same time, suburban Cobb and Gwinnett Counties revolted, voting Democratic for the first time in decades. Georgia Democrats smelt blood, and in the 2018 midterms, with pieces of the suburban coalition in place, nominated Stacey Abrams for governor in a marquee debut of the new strategy.
While Abrams came up short, this was due to Trump-level rural turnout among white voters. This turned most of outstate Georgia even redder. Former Rep. John Barrow’s failed Secretary of State bid performed better among rural white voters, but it still didn’t match the raw suburban turnout and voter engagement in the Atlanta area to come close enough. Even with vote-rich Atlanta suburbs juiced for turnout, the region still got more Democratic.
Now, the binary axes of race and party are one. White voters in the state are almost perfectly Republican, and black voters in the state are almost perfectly Democratic. Due to Republicans virtually shedding their suburban support, they are almost fully based on supercharged rural white turnout, and thus have lost the propensity edge they once had in normal and lower-turnout elections. We saw this in the runoffs.
High-propensity rural black voters, urban liberal white voters, and voters of color powered Ossoff and Warnock’s wins, inverting the original propensity design of the runoff system. Democrats broke 60% in Gwinnett County in a federal race for the first time in decades. Rural white voters all across the state did not vote in the numbers they needed to to elect their preferred candidate. This was clear on election night when county returns were consistently three to five percent more Democratic than November’s results.
This is not to say that Georgia Republicans will never win another statewide election again. Georgia is a highly inelastic, turnout-based state. If for some reason the high-propensity Democratic coalition falters, such as suburbanites reverting somewhat in a red midterm, it is likely that a Republican victory could occur. But this belies the point: the current Democratic coalition that powered three statewide wins in two months is virtually unstoppable in the state if held together.