Democrats Can and Should Expand the Map to Georgia and Arizona in 2020

While still putting the Midwest first

Ruy Teixeira & John Halpin | March 1, 2019

As the field of Democratic presidential candidates continues to grow, the outstanding strategic question for the party remains, “What is the best path forward for victory in 2020?”  

Any realistic assessment of President Trump’s geographic and demographic strengths leads to the conclusion that Democrats need to focus first and foremost on Trump’s backstop of white working class voters in the upper Midwest—particularly, the trio of Pennsylvania, Michigan, and Wisconsin that, along with Florida, tipped the Electoral College in his favor in 2016. Trump is a deeply unpopular president nationally, but he retains a hold on many voters in these states.  Any Democrat hoping to pull off the difficult task of unseating an incumbent president—even one as divisive as Trump—must be able to mobilize base Democratic voters and cut into his margins with white non-college educated voters in these trio of states.

Alternative strategies have been floated, contending that Democrats can essentially ignore these states and these voters and instead turn to a coalition of voters of color and white liberals in emerging battlegrounds in the south and Sunbelt regions. Although these emerging voting blocs are essential parts of the Democratic coalition, this advice makes little sense strategically.  It would be irresponsible for the eventual nominee to concentrate resources in states that Democrats have not won in recent presidential elections only to ignore states like Pennsylvania, Michigan, and Wisconsin (plus Florida) that Barack Obama successfully captured both times he stood for president.

But Democrats at this point should know how to walk and chew gum at the same time, particularly with the chance to recapture both the presidency and the U.S. Senate.  The party and its eventual nominee will almost certainly raise more than $1 billion in the upcoming cycle, and with campaigning already beginning, nearly two years before the actual 2020 election, Democrats have the time and the resources also to expand the map beyond those four key states and thereby stretch Trump thin in 2020. 

So where should they look and who should they look towards to do this? 

The 2018 gubernatorial election in Georgia provides an excellent model for how Democrats can expand the map in ways that put more states in play without conceding the upper Midwest. Although Democratic candidate Stacey Abrams did not win the election last year, losing narrowly to Republican Brian Kemp by only 1.4 points, she did outperform Hillary Clinton in the state by almost 4 percentage points. How did the Abrams campaign get so close to winning a state that just a few years ago observers had thought was lost to Democrats for the foreseeable future?

According to Catalist Analytics data, one key to Abrams’s performance was high turnout among strong Democratic-leaning constituencies, particularly young voters and nonwhite voters. The turnout level of black voters, which actually exceeded 2016’s vote share (31 percent vs. 30 percent), is particularly noteworthy. And young (18-29 year old) voters nearly matched their 2016 vote share (13 percent vs. 15 percent) and significantly exceeded their 2014 share (8 percent).

Besides excellent turnout performance, there were large shifts in Abrams’ direction among groups that had already swung towards the Democrats in 2016: young voters, women voters and suburban voters, especially white voters who live in the areas surrounding Atlanta. The biggest margin shifts toward the Democrats compared to the 2016 tallies came among whites under 45: an 11-point shift among 18-to-29-year-old whites and a 12 point shift among 30-to-44-year-old whites.

Interestingly, the margin shift toward the Democrats was actually larger among white non-college voters (+6 points) than among white college voters (+1 point). This was even true in suburban areas. Thus, despite the popular image of improved Democratic performance in 2018 as overwhelmingly attributable to white college graduate voters, it appears that, in Georgia at least, Democratic gains among white non-college voters were also quite important. The significance of this is underscored by the fact that the white non-college share of Georgia’s voters in 2018 (40 percent) far exceeded the white college share (24 percent). In fact, the overall contribution of white non-college voters to Democrats’ improved margin in 2018 over 2016 was literally 10 times the size of white college voters’ contribution.

It is worth noting that, despite very strong black turnout, Abrams’s popular vote margin among these voters appears to have actually been slightly less than Clinton’s in 2016. The same was true for Latino voters: strong turnout for a midterm election, but a slightly smaller Democratic margin than in the 2016 presidential contest. 

Looking ahead, Democrats’ performance in 2018 in Georgia does suggest a path to victory in 2020, albeit not an easy one. Clearly, high black turnout must be maintained and, ideally, pro-Democratic margins must equal or exceed where they were in 2016. Abrams’s race also shows how important young voters, especially whites, are likely to be for the Democrats in 2020. And there is little room for backsliding on the white non-college vote, since this is still an unsympathetic constituency for the Democrats—a fallback to the 63 point margin by which Trump bested Hillary Clinton in 2016 could prove fatal to their chances in 2020; conversely, continued improvement would substantially improve their chances. Finally, the failure to improve white college margins by much in 2018 must be reckoned a disappointment. A reasonable target would be to move the white college deficit from its -36-point level in 2018 to under 30 points in 2020.

Although Arizona is a state with a radically different demographic profile than Georgia, it too provides good insight into how best to expand the map. One particularly promising result for the Democrats in their very good 2018 election was capturing a Senate seat in Arizona that had long been held by Republicans. Kyrsten Sinema won by 2.4 points in a state that Hillary Clinton lost by 3.5 points in 2016.

So how did Sinema do it when Clinton couldn't? Relying again on the Catalist Analytics data, we note relatively strong nonwhite turnout in 2018 that helped keep the Arizona midterm electorate closer to the presidential electorate than is normally the case. Latinos, who loom particularly large in Arizona, were 14 percent of voters, just 2 points under their 2016 share of voters. 

That was helpful for Sinema's cause—but by far the biggest factor was a strong shift toward the Democrats among white voters (75 percent of Arizona voters in 2018). Compared to 2016, Arizona whites shifted toward the Democrats by 11 points. That included a pro-Democratic shift of 13 points among white college voters (Sinema came close to breaking even with these voters) and an also-impressive shift of 10 points among non-college whites. There were also big shifts among whites by age groups, with Sinema actually carrying all whites under 45.

The regional dimension is also interesting. Sinema not only cleaned up in urban areas but also managed to narrowly carry suburban areas, where two-thirds of Arizona voters reside (as Maricopa County epitomizes). Her success was driven primarily by white voters in suburban areas, who swung a solid 11 points toward the Democrats from their 2016 performance.

Based on these patterns, Arizona could definitely be in play in 2020 and would perhaps be an easier lift for the Democrats than Georgia. The key for 2020 will be replicating Sinema’s performance among whites, which may be summarized as coming close to carrying white college graduates, keeping the deficit among white non-college voters under 20 points, and carrying whites under 45. This may not be easy with Trump on the ballot and seeking to make immigration a central issue.

It will also be necessary to maintain high mobilization among nonwhites, particularly Latinos. Moreover, the margin among Latinos for Sinema was actually smaller than for Clinton in 2016 (Sinema had a +38-point margin, Clinton, a +43). Returning that margin to at least the Clinton level in 2020 will be key for the Democratic candidate.

Texas is another state frequently mentioned as a target for Democrats.  But for a variety of reasons, mostly related to the cost of campaigning there and the more conservative nature of the white electorate, Texas probably should not be a major focus for the 2020 presidential nominee, although it certainly could be in the post-2020 future and in the battle for the U.S. Senate.

That said, Democratic candidate Beto O’Rourke’s performance in the 2018 Texas Senate race was impressive. He came amazingly close to unseating incumbent Republican Senator Ted Cruz, losing by only 2.6 percentage points, a considerable improvement over Clinton’s 9-point loss in the state in 2016. How did O’Rourke manage to get so close? 

He certainly benefitted from unusually high turnout for a midterm. In a relatively high turnout election nationally, Texas stood out with an 18-point increase in turnout over 2014, the sixth highest increase in the nation. The number of votes cast in Texas in 2018 was roughly 8.3 million, not too far off the nearly 9 million votes cast in 2016. 

Looking at the demographics of O’Rourke’s support is tricky, since Catalist has not yet released its statewide data for Texas. That means we have to rely on the exit polls, which are far less reliable, and useless for such comparisons as white college vs. white non-college due to changes in methodology. With these caveats in mind, it does appear that O’Rourke benefitted from strong Latino support, carrying this group by 29 points, somewhat better than Clinton’s 27-point margin in 2016. Even more impressive, exit polls indicate that Latino voters comprised a larger share of the state electorate in this midterm (26 percent) than they did in the 2016 Presidential election (24 percent).

But probably the most salient factor to O’Rourke’s relatively strong performance was a substantial improvement in support among white voters. He lost these voters (56 percent of the electorate) by just 32 points, compared to Clinton’s deficit of 43 points. 

Could the Democratic candidate in 2020 shave this deficit further? That will not be easy. The 48-point deficit O’Rourke suffered among white non-college voters is very large indeed and may not be moveable. White college voters could be a better bet, since O’Rourke only lost his group by 11 points. That would be consistent with the geographical results from the 2018 election that show suburban counties outside Austin, Houston, and Dallas/Fort Worth moved toward the Democrats. 

In addition, of course, Democrats in 2020 will have to maintain and increase the levels of mobilization exhibited in Texas in 2018, particularly among Latinos. Combining that with continued shifts in the white vote is a big challenge, but it is not out of reach, if perhaps more realistic for another cycle down the road.

Democrats face a nerve-wracking election in 2020.  They desperately want to get rid of Trump in any way possible.  The cautious approach would be to focus only on the upper Midwest states while the risky approach would be to ignore these states to focus on emerging southern tier targets.  Neither is correct. The data from 2018 clearly indicate a smart money approach: play in all of these states by creating an open and inclusive campaign that appeals to a wide variety of voters from different backgrounds, with a strong message of economic uplift, political reform, and basic honesty and decency in the White House.  

This “both-and” approach is the obvious way forward for the Democrats in 2020. And, despite the endless debate within the Democratic Party about the Rustbelt vs. the Sunbelt paths, the obvious “both-and” way, in this case, is the right way.

President Obama showed how to do this in 2008 and 2012.  Now the next generation of Democratic leaders, as exemplified by Stacey Abrams and others, can add to this by reaching out to more people in more states with an agenda of common purpose.  Democrats must expand the map and put the interests of working- and middle-class voters of all races at the forefront of national politics as they reject Trump’s divisive actions to pit Americans against one another for his own personal needs for adulation.


Read the article in The American Prospect >