House GOP’s Secret Vote, Deconstructed
It’s been 12 days since House Republicans elected a new majority leader and majority whip behind the closed doors of the House Ways and Means Committee room. And though the ballots and vote totals were a secret, plenty of members and staff think they have an idea. The problem is, they’re probably wrong.
With the exception of the three members who counted the ballots — Shelley Moore Capito of West Virginia, Bill Flores of Texas, and Virginia Foxx of North Carolina — no one definitively knows the vote totals.
Unless, of course, they cracked the safe in conference Chairwoman Cathy McMorris Rodgers’s Cannon office, where the ballots are kept. Those ballots — numbered sheets of paper with candidate names scrawled on each — have not yet been destroyed, contrary to earlier practices, an aide confirmed.
Despite the secrecy of the elections, plenty of members think they know the vote totals.
“What I’m hearing from my House colleagues is that it was in the neighborhood of 156 to 75, give or take,” Rep. Mo Brooks, R-Ala., told CQ Roll Call. That is, Kevin McCarthy of California beat out Raúl R. Labrador of Idaho for the majority leader post. “I have no way of knowing if that is accurate, only the vote counters know for sure.”
That’s a pretty specific tally to be “in the neighborhood of.” Brooks is hearing the same rumor some lawmakers are spreading: A member who was closely watching the vote counters saw one of them tell McMorris Rodgers either “156 or 166.”
That member, speaking privately given the nature of the closed-doors session, told CQ Roll Call the assumption is that the figure was McCarthy’s vote total, which would mean Labrador received 65 or 75 votes.
The lawmaker also noted a few members were conspicuously standing by the vote counters, watching the piles. The two-to-one ratio seemed to jive with their estimates. At one point, someone counted 100 ballots as a sample and found 30 votes being put in the Labrador stack.
Hardly anyone thought Labrador could muster 65 votes.
One vote counter insisted that it wasn’t even close. “The margins for the winners were large, a lot larger than the losers expected,” the lawmaker told CQ Roll Call. More than the 166-65 tally rumor? “It was larger than that,” the lawmaker said.
There actually were 230 members voting, not 231 as the math suggests. The conference knew Mick Mulvaney of South Carolina and Alan Nunnelee of Mississippi were home recovering from health issues, but it turns out James Lankford was home campaigning for his Senate primary in Oklahoma. (He won the next week, and is likely moving across the Rotunda come January.)
CQ Roll Call learned McMorris Rodgers was never even told the actual tally. All three vote counters signed a statement swearing to secrecy.
So where did the estimates come from? The vote counter admitted members could have been observing the size of the stacks over their shoulders, but thinks guessing a count wouldn’t be that easy. Still, there’s enough anecdotal evidence from members spying on the count to suggest Labrador mustered around 50 votes, a stronger showing than most predicted.
“I know some of you in the media keep saying that it was just the same people that I hang out with that voted for me,” Labrador said the day after the election. “That’s not true.”
As for the majority whip race that elected Steve Scalise of Louisiana, the margins between the three candidates were wide, the vote counter told CQ Roll Call.
“The margin between No. 1 and No. 2 on the first ballot, was large, it really was large.” Between No. 2 and No. 3, “Even larger.”
Then the lawmaker revised the estimate: “Well, actually, they were both about the same, the margin between them all was about the same.”
A candidate needed 116 votes for a first-ballot win, so the breakdown under the most generous circumstances for Peter Roskam of Illinois and Marlin Stutzman of Indiana looked something like 116 for Scalise, 77 for Roskam, and 37 for Stutzman.
The vote counter implied, however, that Scalise didn’t simply eke out a first-ballot victory. So a reasonable guess for the election seems to be 127 for Scalise, 77 for Roskam, and 26 for Stutzman.
(Two members opted for write-in votes of candidates not running, according to the vote counter.)
All of this guessing and math, of course, would not be necessary had the conference simply released the total — something they traditionally did, apparently, until Dan Lungren of California challenged John A. Boehner for the minority leader position in November 2008.
Since then, the public and the press have been left with guesswork to determine GOP leadership vote counts.
What’s also been shrouded in some degree of secrecy is the process itself.
The three vote counters divide the stack of ballots into three, roughly equal piles — one for each — and then each vote counter splits the piles by candidate. Each stack is passed to the next counter, ensuring each vote is counted three times.
“We all came to exactly the same number,” the vote counter said.
There was one questionable ballot for whip — thanks to scribbled handwriting. It was errantly put in the Scalise stack but actually was a vote for Stutzman. The vote counters caught the discrepancy and the spectacle sparked memories of the 2000 presidential election and the Florida recount.
“We talked about hanging chads. We were kind of laughing about hanging chads,” the vote counter said.
In the end, the conference agreed after the results were announced to declare unanimous victories for McCarthy and Scalise thanks to their wide-winning margins.
So unless some skilled safecracker attempts to snag the true totals, the votes will be recorded in history simply as unanimous.