2018 Elections: What did we learn?
Today is one month after Election Day - the votes have been counted, the races certified, and the narratives solidified. We’ve been pouring over all of the analyses done (most focused on the “R vs D” competition) to draw relevant insights about the context in which new candidates may run. Below are a few we have come to.
The Narrative
The popular question going into the Election Day was, will there be a “blue wave”? On Election Day, Democrats looked like they had picked up ~30 seats in the U.S. House, which seemed modest (a “blue ripple”) compared to expectations. But as the votes continued to be counted over the following weeks, there was a constant drumbeat of close House races called for Democrats, netting an additional ~10 seats, and talk shifted into “blue wave” territory. This was tempered by a few high profile gubernatorial and Senate losses. Overall, though, less than 20 races defined the narrative. We see more nuance as we look further down ballot and with historical context.
Elected Offices
Let’s level set with the universe of elected offices: There are ~520,000 elected offices nationwide. Of those, just 0.1% (537) are federal; 1.5% (~7800) are state-level; and the rest (98.4%) are local. Here’s a simple graph to emphasize the point:
In 2018, there were elections for:
- 35 U.S. Senate seats
- 435 U.S. House seats
- 303 state executive offices
- 6,073 state legislative seats (82% of all)
- ~250,000-~400,000 local seats (no authoritative count has been published, but our educated estimate is ~50%-80% of all)
The Results
We are keen to determine where there were changes from the status quo, what contributed, and what led any candidates to overperform or underperform baseline expectations.
Turnout
Turnout was relatively very high. Turnout this year was estimated to be 50.3% of all eligible voters, which is the highest midterm turnout since women gained the right to vote in 1920, significantly up from 37% in 2014 and 42% in 2010. The last peak was in the 1960’s. (Obviously, there’s a lot of headspace to grow turnout further.)
Both D’s and R’s turnout increased — and this was tied to the President. While the House popular vote (aggregate vote total for each party across all U.S. House seats) was a sizable 8% better for D’s (59.4M-50.4M), R’s were still up 26% from the last midterm’s 40.1M. President Trump successfully rallied voters to view this election as a referendum on his presidency, with~65% voting as “a message about the President”. The numbers show that while the last two years have hardened those opposed to the President and drove significant turnout, the President’s base support remains high and mobilized — the electorate resembles 2016's.
Demographic shifts were significant, but not everywhere. Nationally, women, people of color, millennials, and post-millennials broke heavily for D’s, but the disparity was far smaller in conservative states.
Post-Trump elections will require different turnout strategies. With turnout currently driven by the President, candidates in elections after this President has left office will need to modify their efforts in order to build and maintain turnout, down ballot especially.
Federal: U.S. House
Democrats won. Democrats gained +40 seats, better than +23 avg since 1960 (on par with +38 avg since 1862). Whether this empirically constitutes a “blue wave” is primarily based on what period of time one places this election in. If looking back at all elections since 1918, this election would not be at “wave” level (> +48). But if looking at more recent D performance, this is the party’s biggest single election gain since 1974. The bigger question was, did D’s take over the House (yes) with a significant buffer (yes).
Democrats made the House more diverse. The House took a significant step towards reflecting the diversity of America:
- +17 (+20%) women to now 102 of 435 (23%) (still a ways to go until parity)
- +26 millennials
- +2 Native Americans, +2 Muslims, +4 Hispanic, +5 African Americans…
But the increase in diversity was not evenly split between parties. R ranks saw a net loss of 10 women and 4 people of color.
The competitive general election contests were in moderate suburbs. D’s primarily flipped suburban districts, with the help of suburban women voters and activists. These areas are more socially progressive, but economically conservative, and the candidates who succeeded here used messaging that matched their district. Most of the strongly progressive newly electeds came from already heavily D districts, winning contested primaries before relatively uncompetitive general elections.
Federal: U.S. Senate
Republicans won. R’s had a positive midterm of +2 seats, better than the avg of -2. Given that the selection of seats contested this year was so advantageous to R’s, it’s hard to say which party over/underperformed. The bigger question here was, did R’s maintain control (yes) and increase their buffer (yes) with seemingly more caucus-aligned Senators than before (yes).
No consistency in winning strategies in conservative battleground states. R’s won in IN, ND, MO, FL, and TN, but D’s held/gained in MT, WV, NV, and AZ. D incumbents and challengers generally positioned themselves as moderates, fighting for the middle — which worked some places, but not others. R messaging focused on support for Trump and his policies on immigration, taxes, and jobs. The exception was Beto O’Rourke, who lost but still overperformed in TX, authentically espousing progressive values and focusing on listening to, exciting, and turning out new voters more than persuading traditionally swing voters.
State: Governors
Democrats +7 / Republicans -6. D’s flipped IL, WI, KA, ME, MI, NM, NV, and R’s picked up AK (formerly independent). This is wave territory.
Pragmatic messaging won. D’s battleground messaging was mostly pragmatic. R’s have continued to run strong in the northeast (NH, VT, MA, MD) with moderate, technocratic messaging. D’s GA and FL candidates were the exception, running strongly progressive campaigns that ultimately lost, but were very close, using strategies similar to Beto O’Rourke’s Senate race.
State: Legislature
Democrats contested significantly more seats, picked up many, but less than historical averages. In 27 of the last 29 midterm elections, the party of the President loses seats, an average loss of 415 seats. State legislative general elections had been getting steadily less competitive, with 42% of all seats uncontested in 2016. In 2018, D’s contested significantly more seats (R’s were steady), leading to an overall drop of uncontested races to 33% (12%D + 21%R). R’s also retired in the highest numbers in a decade (723 seats). Despite this, in 2018, D’s gained ~308 seats, underperforming historical averages and far below the +494 demarking a wave. Adding in 2017 races + special elections, D’s still gained under 350 seats.
States consolidated either D or R legislative control. D’s gained 6 chambers, R’s likely gained 1 = net D+5. 7 chamber flips is less than the average of 12 flips and is a 30 year low. The chambers that did flip are among the most frequent to flip. Only 1 state (MN) now has its upper and lower chambers controlled by different parties.
State: Trifectas & Supermajorities
Greater uniform control. Trifectas (where the governor and both houses of the state legislature are under the same party’s control): R’s entered 2018 with overwhelming control of state governments, having 26 trifectas, while D’s had just 8. In this election, D’s netted 6 trifectas and R’s lost 3, leading to R’s having 23 trifectas and D’s having 14 — the other 13 states have divided government. 37 trifectas ties for the most trifectas in recent decades.
One supermajority broken. Supermajority legislatures can override a governor’s veto, which makes the greatest difference in states with legislatures and governors of opposite parties. Going into 2018, R’s had a supermajority in NC with a D governor and D’s had supermajorities in MA and MD with R governors. After 2018, R’s lost their NC supermajority, but D’s maintained theirs.
States are/will be the laboratories for policy. More policy will pass in states than at the divided federal level over next two years, especially in these 39 states with highly aligned governments.
States: New people
Fresh perspectives / Brain drain. Across state governments, there are a lot of new people.
- 19 new governors
- 36 of 99 new leaders of legislative chambers
- 1700 new legislators (there was very high turnover: 23%, rather than the typical 19%)
Greater diversity, especially of women:
- Governors: Women increased +3 to 9/50
- Legislators: Women up to 28% of seats (from between 20–25% over past 15 years). More chamber leaders will be women. NV and CO have female-majority chambers.
Municipal
(we are actively seeking analysis of these races, but little has been compiled and published to our knowledge)
Overall…
State legislative data seem like a better proxy for local elections than the far fewer but higher profile state-wide and federal races. Spurred by recent federal election outcomes, there were far more candidates contesting down-ballot races and far greater voter turnout, but this did not lead to an especially high number of incumbent losses (of either party during the primary or general election). It could be that districts are well represented already or otherwise have become less competitive. Or it could be that the most competitive prospective candidates did not run or that the candidates who ran did not receive the necessary training and guidance to run their best campaign. Or it could be sufficient external support (money, volunteers) was not available to enable challengers to be competitive.
We continue to see the recruitment and training space as having the greatest opportunity for impact, especially when looking beyond elections towards governance. Other national training programs trained ~2,000 candidates who ran during the 2018 cycle (federal, state, and local), representing < 1% of all seats. There is a lot of space for new innovative approaches and partnerships. We are excited to expand our work in 2019.
Help empower thoughtful leaders everywhere to run for office successfully and govern effectively!
