Monday, November 5, 2018

Some thoughts on what to look for in NC's Election evening for the NC General Assembly contests

After 2016, I thought it best to stick to "explaining" why elections and their results happened, rather than "predicting" beforehand what could happen on Election evening. But as I'm teaching a course on Congress this semester, and we just wrapped up the section on mid-term elections, I thought it might be fun to post some thoughts on elections and what political scientists call the "fundamentals" when it comes to campaigns & elections, especially at the state level for 2018.

As many North Carolinians will be focused on the state's 13 congressional districts, it will be important to watch the state's General Assembly contests for both the state house and state senate in the Old North State. All eyes are on the question: can Democrats reduce the Republicans' seats in the state senate (GOP controls 35 seats) and the state house (GOP controls 75 seats) to below 'supermajority' status (3/5's of the 50 senate seats and 120 of the house seats) and thus break the power of the legislature to override Democratic Governor Roy Cooper's veto? If so, that would pose an interesting divided government scenario for the state for the next two years.

Thus, the key numbers for Democrats to pick up is 6 in the state senate (to reduce Republican seats down to 29) and 4 in the state house (to reduce Republican seats down to 71).

Many observers will be looking at a lot of different dynamics Tuesday night for these districts. Two good analysts are Jonathan Kappler at the NC Free Enterprise Foundation and his analysis of both the house and senate districts, and Darren Janz* over at Politics NC blog (house & senate), but I tend to focus on the 'fundamentals' of a district, such as: President Trump's 2016 performance in the district; party registration percentages; voters' racial percentages; percentage in central city urban voters/suburban voters in urban county/surrounding suburban county voters/rural county; and whether it's an open seat contest or an incumbent vs. challenger contest.

As explained in an early post, Trump's 2016 performance is closely related to the Republican state legislative candidate's 2016 performance in the district: in the state senate, Trump's 2016 vote percentage explains 95 percent of the Republican state senate candidate's vote percentage:

While in the state house, Trump's 2016 vote percentage explains 82 percent of the Republican state house candidate's vote percentage:

With this close relationship between Trump's percentage in the district, I think that's a key fundamental, due to the fact that mid-term elections tend to be referendums on the party in power, and that's usually the White House and Congress--and this year, it's the Republicans. How much connection there will be between the national and state-level dynamics may be tested, especially if the president's approval is sitting in the low to mid-40s.

Another key fundamental for state legislative districts are the percentages of party registration within the districts. While North Carolina is home to a legal contest over partisan gerrymandering, the power of these districts based on party dynamics is a key fundamental to watch. The true wildcard in this mix is the growing influence, and uncertainly, of unaffiliated voters. Typically these unaffiliated are not 'pure independent voters' and have a partisan bent (they just don't like the labels), but if these voters move against the party in power, it could help us to understand after the fact the results from Tuesday evening.

Another key fundamental, especially in Southern politics, is the composition of race among the voters within a district. Many Southern white conservatives have become strong Republican voters, while the overwhelming majority of black Southerners are staunchly Democratic in voter loyalty.

Some of these districts are considered 'open seat' contests, in which there is no incumbent (and thus no incumbency advantage) for the party controlling the seat. Open seat elections tend to make the district much more competitive, in light of other fundamentals (but some open seat elections will simply transfer the seat from one member of the party that controls it to another candidate of the same party because of the district's partisanship and voting behavior). But in conjunction with other fundamentals, open seats can be ones that pose the greatest risk and ones to watch during inhospitable electoral seasons.

Finally, the dynamic of 'regionalism' is having a strong influence in the Old North State's voting behavior, and if the national trend of this mid-term being a 'suburban' voter revolt against the President is evident, we should see if play out in districts with a strong suburban voter tilts, especially in areas outside of the central city in an urban county and in the surrounding suburban counties to that urban county. 

A spreadsheet with all of NC's state senate and house districts can be found here. The columns represent different pieces of fundamental information on each district:
  • A: Current incumbent party
  • B: House district #
  • C: Trump's 2016 Vote Percentage in the 2018 Senate/House Districts
  • D: Republican Candidate Percentage in Tuesday's Election (to be filled in)
  • F: Major county that is in the district
  • H: Registered Democratic Percentage, based on 10/27/18 data from the NC State Board of Elections
  • I: Registered Unaffiliated Percentage
  • J: Registered Republican Percentage
  • L: Registered White Voter Percentage
  • M: Registered Black/African-American Voter Percentage
  • N: Registered All Other Races Voter Percentage
  • P: Percentage of Central City Voters in Urban County in the District, based on this classification from the Office of Management & Budget
  • Q: Percentage of Suburban Voters in Urban County (but outside the Central City) in the District
  • R: Percentage of Surrounding Suburban County Voters in the District
  • S: Percentage of Rural County Voters in the District
  • U: Republican Candidate
  • V: Democratic Candidate
  • W: An open seat contest?
  • Y: District #
  • Z - AO: the actual numbers from October 27, 2018 NC State Board of Elections data that creates the percentages for: party registrations; voter race; and 'regions' for districts. 
So, while these aren't necessarily predictions, I will be watching some of the following districts to see how the fundamentals line up and what the final results are, if anything can be gleaned out of what I think has been a very strange election, most notably, the impressive increase in early votes.

In the State Senate:

Senate District 1 (Camden): This seat may have been a hard seat to pick up, due to Trump at 59%/GOP senate candidate at 59% in 2016, but the new 2018 district is now 55% Trump. Current registration is 42% Democratic, but it’s 80% rural; most observers of the state's politics know that rural, typically white, older registered Democrats are Republican voters. But this contest also presents another key fundamental: typically, open seats are most likely to swing against party in a wave year, and this is an open seat.

Senate District 9 (New Hanover): Another Trump 51%/GOP Senate candidate 57% in 2016; the 2018 district is 51% Trump. Again, watch for Trump disapproval swing effect and any ‘drag’ on GOP candidate. This district is a unique but becoming all too familiar one in the state, due to the fact that a plurality of registered voters are unaffiliated (37%), and the district is 51% central city-urban/49% outside central city but inside urban county (aka, suburban).

Senate District 19 (Cumberland): Trump got 50% of the district’s 2016 vote, while GOP senate candidate got 56% (most political analysts would consider the range of 45-55 percent to be competitive, so the GOP state senate candidate was barely outside competitive status). In the new district, Trump would have gotten 48%. If there’s a swing against Trump and his 2016 performance, how deep is it and does it cut into GOP candidate’s performance?

Senate District 27 (Greensboro): Trump got 50% of district vote in 2016, while the Republican senate candidate got 53%. In 2018, it’s again 50% Trump. Both performances are competitive, and this one, with an environment of urban/suburban moving against GOP due to Trump disapproval, could be one to watch for early dynamics of any wave.

Senate District 41 (Mecklenburg): this disgruntled district over I-77 tolls continues to reverberate in its hostilities about the toll lanes. Trump got 47% of the 2016 vote; in the new district, Trump would get 45%. This district may be a prime pick-up test case if Democratic suburban wave and Trump's disapproval are in combination, but local forces may have greater impact on whether the Republicans can hold this seat or not. 

In the State House:

It's likely that 2 seats will flip from R to D (House Districts 8 & 61) and 2 seats will flip from D to R (Districts 66 & 7). So, a wash in terms of where both parties may begin the real contests. 

House District 1 (Bertie): it's an open seat, with Trump in 2016 at 64% and GOP candidate was 64%; but the 2018’s district is now 52% Trump. Among the party registration, 53% registered Democratic and 38% black, which equates to a dangerous combination for Republican to hold on.

House District 25 (Nash): 2016 Trump was 61% and GOP candidate was 68%, but the 2018 Trump is 45%. Party registration shows that it's 53% Dem, but 50% white to 44% black. Another open seat contest, so one to see how any wave may push this from an R to a D. 

House Districts 35-37 (all in Wake): in these three seats, 2016 Trump's vote was 49, 45, and 51% (respectively); 2018 Trump for all districts is 49 to 50%. All three have GOP registration right at 34-35%, typically a key indicator in the state for a district to swing Republican. But these are heavy suburban districts inside an urban county, and if NC's suburbs revolt, watch these and 3 in Mecklenburg (103-105).

House District 98 (Mecklenburg): 2016 Trump was 48% but the GOP candidate was 56%; 2018's district benefited a bit, with Trump at 52%. However, it's a 100% “suburban” voters in urban county, with 38% Unaffiliated to 36% GOP to 25% Dem. Again, this could be a test case to see what kind of suburban revolt there is and how deep it could be Tuesday night. 

House Districts 103-105 (Mecklenburg): 2016 Trump %s were 49, 43, 46, while GOP candidates were 55-56%; 2018 Trump is now 52%, 44%, & 47% respectively. GOP registration between 36-38%, so a bit of a cushion, but Districts 104 & 105 are inside urban city limits, while 103 is 60% suburban voter in urban county. One incumbent candidate's recent news may not help his chances in the 104, but again, the question of the suburban momentum to any Democratic wave will be key to watch.

Again, no predictions in what I've described as a strange and weird election, but for those of us who study politics and campaigns and elections, there are some potential fundamentals that could shape how the elections play out.

* The Politics NC ratings were originally credited to Thomas Mills; this was in error, as the ratings are from Darren Janz. I apologize for the error.