We’ve had four House special elections this year. Democrats have lost all four. All four were tough races to replace Republicans from very safe districts who were appointed to posts within the Trump administration. In 2016, the Republicans were all incumbents, and their Democratic challengers were:
- Mike Pompeo (60.7%) won against Daniel Giroux (29.6%) in KS-4
- Ryan Zinke (56.2%) won against Denise Juneau (40.5%) in MT-AL
- Tom Price (61.7%) won against Rodney Stooksbury (38.3%) in GA-06
- Mick Mulvaney (59.2%) won against Fran Person (38.7%) in SC-05
In KS-04, an Independent and a Libertarian won almost 10% of the vote in 2016.
First for the good news, Democrats improved their margins significantly, losing by single digits races they had lost by 15%, 20% or even 30%.
3rd party candidates didn’t play a significant role in the special elections except in Montana. I’m ignoring CA-34 for obvious reasons. FWIW, Chaffetz’s seat in UT-03 is up on November 7, Sessions’ Senate seat is also up on Dec 12.
Two of these special elections rank among the most expensive house races. The Georgia race was the most expensive ever, with estimates putting total spending at $60 million. The DCCC dropped $5 million into the Ossoff race, which was an enormous bet that they could win a deeply Republican district in the Atlanta suburbs. The DCCC and other Democratic groups spent far less on the other three races.
D Party Spend
R Party Spend
* The numbers above aren’t complete, Q2 filings aren’t in, I link to my sources.
The amount of spending does seem to correlate with turnout. The two expensive races, MT-AL and GA-06 had pretty high turnout for an off-year election. For a special election, it was sky-high. The low spend races had much larger drops in turnout.
But the spending didn’t result in better margins. The improvements over the 2016 Democratic candidate and Clinton’s vote share were lowest in MT-AL and GA-06. The largest gains over 2016 tallies were in KS-04 and SC-05. Those races were largely ignored by the national media and Democratic donors, until the very last minute.
Distaste for Trump alone won’t automatically translate into victory. I listed the candidate’s stances on single-payer as a proxy for whether or not they’re progressive, so you can consider whether or not policy positions played a role. It’s not clear it made a difference, but if you look at average improvement over the 2016 Presidential and Congressional candidate, even Rob Quist did better (3.6% + 8.2%) than Jon Ossoff (9.8% + 1.3%). There is still some question as to whether any of the candidates have run a robust “anti-Trump” campaign.
That said, every candidate improved on the vote-share gained by the presidential or congressional candidate in 2016. They all improved on Obama’s vote-share in 2012.
If improvement over 2016 is the measure, then James Thompson’s campaign was the best run. But he lost by the widest margin. If loss by smallest margin is the measure, then Archie Parnell ran the best campaign. And Parnell is literally a lawyer for Goldman who ran on a centrist message.
The loss margin has shrunk with each race. Perhaps that’s because the GA, SC candidates ran better races. Or it may be because Trump’s approval ratings have dropped even further in a month. The only woman in the four races was a Republican, Karen Handel in GA-06. All the candidates were white.
There are many different ways to read the data. Personally, the two things that stand out to me is that the sleeper races were tighter, and that fundraising/spending drove turnout.
It is very difficult to claim the enormous sums spent on the GA-06 race were a wise allocation of resources. Yes, on its face, Clinton outperformed Obama’s 2012 tally in GA-06, so it’s reasonable to believe that district had a high dislike for Trump. Yet Ossoff didn’t improve much over Clinton’s total. Should that silence those strategists who believe dislike of Trump married to a carefully crafted centrist message delivered by a polished candidates will carry Democrats to victory (at least in the suburbs)? It probably won’t, because there are other reasons the party contested Ga-06 as if it were and existential threat:
If Ossoff were to win, it would be taken by a lot of Democrats as a huge validation of the centrist/barely-left-of-center approach to 2018.
But yeah, if he comes up short, it’s a new opening for the left of the party to argue that only a more pointed liberal message will fire up the voters the party needs for 2018.
That was the biggest concern mainstream Democrats had going into tonight. — NY Times
The biggest question I’m left with is this. Would it have made more of a long-term difference if $20 of the $28 million spent in GA-06 had been spent on a voter registration campaign? We will never know.
PS. If this isn’t enough, and you’re still searching for answers as to why Jon Ossoff lost in GA-06. Eve Peyser over at Vice has collected 51 reasons, including this gem:
Ossoff was too skinny. America likes fat leaders.
— Judah Friedlander (@JudahWorldChamp) June 21, 2017