Turkey holds a runoff in its presidential election on May 28, which would not normally be our concern except that it gives us an opportunity to:
a) analyze the voting trends in the Kurdish precincts in southeastern Turkey.
b) speculate about the political direction of the nation on NATO’s eastern flank.
c) look at Virginia’s history, present and, perhaps, future.
I’m guessing most of you would prefer (c) so let’s go with that.
Turkey’s runoff comes after President Recep Tayyip Erdogan failed to win a majority in what turned out to be the first-round election. He took 49.5% in a three-way race. His main challenger, Kemal Kilicdaroglu, took 45%. Turkish law requires a majority. That’s not an unusual provision. France has the same rule. So do Georgia and Louisiana. Twice in the past two congressional elections (2022 and 2020), Georgia’s Senate race has required a runoff to determine the winner — and, ultimately, control of the closely divided U.S. Senate. Ten states — almost entirely in the South — require runoffs in primaries where the winner falls short of a majority. The origins and motives for these runoffs vary, but all the Southern ones are rooted in the fact that Democrats once were the dominant party there — and the primary was tantamount to election. (Now it’s typically the Republican primary that is tantamount to election.) Arkansas enacted a runoff in the 1930s to prevent Klan candidates from winning a multicandidate primary with a small plurality. In other states, runoffs were seen as a way to box out more liberal candidates who might be able to win a plurality but not a majority.
Virginia used to be one of those. (I’m indebted to University of Virginia political analyst Larry Sabato for much of this history, which he wrote about in his classic 1977 work, “The Democratic Party Primary in Virginia: Tantamount to Election No Longer.”) Virginia Democrats first started holding primaries in 1912. There was no requirement for a majority, because there was no need for such a rule. Other Southern states had multiple factions within the Democratic Party. In Virginia, though, there were only two: the “organization” (later the Byrd Organization or the Byrd Machine, named after Gov. and later Sen. Harry Byrd), and those who were “anti-organization.” Virginia being Virginia, the organization — often written in upper case, The Organization — always prevailed.
Then came 1949, one of the most important election years in Virginia ever. The organization, er, Organization candidate was John Battle. Two other candidates with Organization ties decided to run, though. So did a fourth candidate: an anti-Organization candidate named Francis Pickens Miller. Miller, who had grown up in Rockbridge County and is buried there, scared the bejeebers out of the Organization. A decorated military veteran who had served on Dwight Eisenhower’s staff in World War II, he posed the most serious threat that the Organization had ever seen. (One of his main issues: State funding for school construction, an issue that endures today.) Miller came close to winning. He might have, too, if Byrd hadn’t intervened to rally support for the beleaguered Battle. Battle took 42.7% of the vote, Miller 35.4%, with the other two candidates at 15% and 6.9%. Battle went on to win the governorship and co-opted Miller’s school construction plank; that’s why the 1950s saw a wave of state-funded school construction — many of those buildings are still in use today.
That’s policy, though. Here are the politics: The Byrd Machine wanted to make sure no future anti-Organization candidates could win in a multicandidate race, so they passed a runoff law. It was only used twice — both in 1969, in the Democratic primaries for governor and attorney general.
The headliner that year was a three-way contest for the Democratic nomination that ended with Bill Battle at 39.6%, Henry Howell at 37% and Fred Pollard at 23.4%. Those results showed how much the Democratic Party was changing. Pollard was the classic Organization candidate; Battle a more modern candidate but still part of the so-called establishment (he was son of the former Gov. Battle); Howell was the classic anti-Organization candidate.
I’ll let Sabato tell the story from here since he took the time to type out a long, thoughtful email on the subject:
The Organization had frayed after Byrd died in 1966, and the Voting Rights Act of 1965 resulted in tens of thousands of new Black voters, changing the nature of the Democratic primary. Nonetheless, observers expected Battle to consolidate the Pollard vote and win the runoff handily, as Byrd & Company had designed it back in 1952. Nope. Battle won a narrow victory over Howell, 52% to 48%. Afterwards, Battle never consolidated the Democratic vote, and went on to lose in November to Linwood Holton.
The death of the runoff was inevitable at that point. The next year (1970) Harry Byrd Jr. — appointed to his father’s seat in 1965 after Senior’s resignation (he was ill) — came up for election again. (He narrowly won the Democratic nomination in summer 1966 for the four years remaining on his father’s term.) Byrd correctly assessed that he was very vulnerable in the primary, so he left the Democratic Party and announced for reelection as an Independent.
Meanwhile, ex-Del. George Rawlings and Del. Clive DuVal faced off in the Democratic primary. The turnout was miserable, and Rawlings edged DuVal out, 46% to 45% (just 700 votes separated them). A third candidate took 9%, leaving Democrats facing a runoff. But DuVal read the handwriting on the wall, and rather than continue the fight for a probably worthless nomination, he threw in the towel, waiving his right to a runoff. Byrd went on to win a solid majority (53%) in November against Rawlings and Holton’s Republican candidate, Del. Ray Garland of Roanoke.
Shortly thereafter, the runoff was abolished, never to be seen in these parts again.
Our story could end there, but I prefer to think it’s really only the beginning. What if Virginia hadn’t abolished the runoff? For starters, we’d have had runoffs in five Democratic nominations for lieutenant governor, two Democratic nominations for attorney general, one Republican nomination for lieutenant governor and one Republican nomination for attorney general. Democrats dominate the list not because Republicans are less fractious but because they are less inclined to hold primaries, often preferring conventions instead.
Here’s that list of the runoffs for lieutenant governor and attorney general that we missed:
- the 1977 Democratic nomination for lieutenant governor (won by Chuck Robb with 39% with Richard “Major” Reynolds second at 33%)
- the 1977 Democratic nomination for attorney general (won by Ed Lane with 35% with Shad Solomon second at 23.2%)
- the 1997 Republican nomination for attorney general (won by Mark Earley with 36% with Jerry Kilgore second at 24.6%)
- the 2001 Democratic nomination for lieutenant governor (won by Tim Kaine with 40% with Alan Diamonstein second at 31.4%)
- the 2001 Democratic nomination for attorney general (won by Don McEachin with 34% with John Edwards second at 29.5%)
- the 2005 Democratic nomination for lieutenant governor (won by Leslie Byrne with 33% with Viola Baskerville second at 26.1% )
- the 2017 Democratic nomination for lieutenant governor (won by Justin Fairfax with 49% with Susan Platt second at 39.1%)
- the 2017 Republican nomination for lieutenant governor (won by Jill Vogel with 43% with Bryce Reeves second at 40.0%)
- the 2021 Democratic nomination for lieutenant governor (won by Hala Ayala with 38% with Sam Rasoul second at 24.3%)
This list of nine winners includes two future governors (Robb and Kaine), one future attorney general (Earley) and one future lieutenant governor (Fairfax), all of whom won their initial primaries with less than a majority vote. How might a runoff have altered things? Could they have put together a majority in a runoff? Some surely would have, but I suspect Lane, who was one of the last Democratic conservatives, would have had a hard time against Solomon in 1977 if liberals had united around the Bath County legislator. How might Solomon have fared against Marshall Coleman in the general election that fall? Coleman that fall ran to the left of Lane; he wouldn’t have run to the left of Solomon. Likewise, how might McEachin might have fared in a runoff against Edwards in 2001? McEachin lost badly that November, the only Democrat on the statewide ticket to lose. If Edwards had won in a runoff, might he have gone on to defeat Jerry Kilgore in November?
If the runoff law had continued, we’d have had three runoffs in gubernatorial primaries:
- the 1989 Republican nomination for governor (won by Marshall Coleman with 36.8% with Paul Trible second at 35.1%)
- the 2009 Democratic nomination for governor (won by Creigh Deeds with 49.8% with Terry McAuliffe second at 26.4%)
- the 2017 Republican nomination for governor (won by Ed Gillespie with 43.7% with Corey Stewart second at 42.5%)
For what it’s worth, all three of those primary winners went on to lose their fall contests. Two of the three weren’t close, but Coleman against Douglas Wilder went to a recount. Feel free to speculate what a Republican runoff that year might have done.
We’d have had only one Senate runoff — the 2018 Republican primary was won by Stewart, with 44.9% of the vote to 43.9% for Nick Freitas. Earl Jackson peeled off 12%; it would have been fascinating to see what a straight-up Stewart vs. Freitas contest would have meant. Mind you, that’s just my eye for history, not an endorsement of policy.
So far we’ve dealt with history and hypotheticals. What I’ll introduce next is history, as well, but more recent history: In 2015, state Sens. Charles Carrico, R-Grayson County, and Ryan McDougle, R-Hanover County, introduced separate bills to require runoffs in any general election in Virginia where the winner didn’t get 50% of the vote. McDougle’s bill was rolled into Carrico’s and that bill passed the state Senate 22-16 on nearly a party-line vote, with Republicans in favor, Democrats against. The bill died in a Republican-controlled House committee.
This bill came two years after Democrat Terry McAuliffe won the governorship without a majority vote — he took 47.7% in a three-way race against Republican Ken Cuccinelli (45.2%) and Libertarian Robert Sarvis 6.5%. It also came two years after Democrat Mark Warner won reelection with just 49.1% in 2014 against Republican Ed Gillespie’s 48.3%. How might a McAuliffe-Cuccinelli runoff have worked out? Or a Warner-Gillespie runoff in what was a distinctly Republican year nationally?
The argument for runoffs is that winners ought to get a majority; the argument against is that it requires voters to come out a second time. Historically, runoffs result in a lower turnout than the first round (although those 1969 Democratic runoffs for governor and attorney general were exceptions). Democrats, who believe they do best with a higher turnout (a somewhat debatable proposition but that’s the belief), aren’t keen to roll the dice on runoffs.
Fun fact: In last year’s Georgia Senate runoff, Democrat Raphael Warnock, who finished first in the first round, received fewer votes in the runoff than Republican Herschel Walker, the second-place candidate, had received in the first round. The problem for Walker is that his vote total went down even more in the runoff. The same phenomenon was true for the 2020 Georgia Senate runoff between Democrat Jon Ossof and Republican David Perdue. A lower turnout in those runoffs certainly didn’t help Republicans.
In general, runoffs work against any candidate who can win a plurality but not a majority, be they on the left or the right. In some situations, they might make it more difficult for minority candidates to be elected, although the most famous example of a runoff election is very much on the opposite end of the spectrum. In 1991, Klansman David Duke came close to leading the first round of balloting for Louisiana governor — Democratic Edwin Edwards took 33.7%, Duke took 31.7%. Without a runoff, it wouldn’t have taken much change for Duke to win. In the runoff, though, Edwards won handily, 61.2% to 38.8% (which still seems a lot more votes than a Klansman ought to get, but that’s beside the point).
If that Carrico-McDougle bill had gone through, McAuliffe would have surely vetoed it — and there weren’t enough votes to override that. We also haven’t had a statewide election since then that was won by a candidate with less than a majority — although we always could. All you need is a close race with a third-party candidate taking even just a few votes.
We end on an ironic note: In Virginia, and perhaps other states, it’s conservatives who have wanted runoffs and liberals who have opposed them. But in Turkey, it’s conservatives supporting Erdogan who are rueing the runoff provision and liberals backing Kilicdaroglu who are hoping for different results.