California's recall system is really bad

California's recall system is really bad

Slow Boring / by Matthew Yglesias / 1h

(Stephen Lam/San Francisco Chronicle via Getty Images)

Gavin Newsom beat back the California recall effort by a healthy margin, despite (or perhaps because of) all the fuss about the race.

Had I been in California, I would’ve voted no, and as a non-Californian, I am glad that Newsom won. That said, one upside of Newsom losing is that it almost certainly would have prompted a serious effort to fix the state’s incredibly ill-considered recall system. With Newsom victorious and Democrats becoming perhaps excessively confident in their own tactical choices, I’m a bit worried that momentum for reform might evaporate.

This is unfortunate because the current system is so bad that almost any kind of change would be better.

California has traditionally been a national leader in constitutional innovation (such as it exists) in American state governments, but they’ve landed on a bunch of innovations that are worse than the American baseline. It would be kind of cool if, instead, they took some lessons from abroad and built a better system.

California’s recall rules are bad

Here’s how the recall system works in California:

To get a recall election on the ballot, you need to collect a bunch of signatures. That used to be a very high hurdle, but improvements in communications technology and what seems like a structural increase in the extent to which people are engaged with politics have made it easier and easier.

To recall the governor, you need to win a majority of the votes.

In an election run simultaneously to the recall ballot, voters get to pick who will become governor if the recall passes.

The upshot of this is that you could easily have a situation where a mildly unpopular governor loses a recall ballot 49-51 and then is replaced by someone who got 26% of the vote in a nine candidate field. Not only is that stupid, but it would also almost certainly lead to another recall election and an endless spiral of referenda and confusion.

Back in 2003, a cabal of GOP operatives got Arnold Schwarzenegger into the race as a moderate Republican option. We know from states like Massachusetts, Maryland, and Vermont that moderate Republicans can easily win elections in even very blue states under the right conditions. But conventional wisdom in California holds that a moderate can’t win a GOP primary the way they can in the northeast, so the recall was seen as a mechanism to help sneak Arnold in. And it worked.

Perhaps over-learning from that n=1 sample, the initial Republican plan was to re-run that strategy with former San Diego mayor Kevin Faulconer in the role of an electable moderate. But because Republican Party politics has grown incredibly weird, the buzz instead went first to Caitlyn Jenner and then to Larry Elder, a talk radio host who, quite separate from any scandals, is a guy who made zero policy or ideological concessions to the actual views of the California electorate.

Republicans weren’t the only ones basing their strategy on the 2003 outcome. Democrats ran Lieutenant Governor Cruz Bustamante on the second ballot that year, and since that ended with Gray Davis getting recalled, Democrats learned that it was a mistake to engage with the second ballot. The Powers That Be in California politics successfully ensured that none of California’s many, many, many Democratic Party statewide elected officials, members of Congress, or big city mayors ran on the second ballot. They also delivered the message to grassroots Democrats that there was no acceptable choice on ballot two — just leave it blank. Some random YouTube guy ran as a Democrat, but you weren’t supposed to vote for him. Faulconer would’ve made the best governor out of the bunch, but you weren’t supposed to vote for him. The goal was to characterize this as a head-to-head race between Newsom and Elder. And while at one point it looked like this strategy might fail, it ultimately worked, thus “proving” the decision was correct all along.

But was it?

Imagine a closer race

It looks like the recall will be defeated by a margin of about 64-36, which is not very close.

On the second ballot, Elder currently has about 2.3 million votes or 46% of votes cast. His numbers are inflated by the fact that so many Democrats followed party leadership’s directive to avoid voting on the second question, but it seems like about 25% of the people who voted on the first question voted for Elder in the second question. In other words, “yes” ran about 11 points ahead of Elder.

Newsom's huge win was hardly inevitable. Polling earlier in the year showed a close race, which seemingly only opened up because Elder went anti-vax at exactly the moment the Delta variant was tearing through many conservative states. That means it's relatively easy to imagine a world in which Elder is savvier or the Delta wave is timed differently. In that world, perhaps Newsom gets recalled by a 49-51 margin, but Elder still runs 11 points behind yes because some people vote for Faulconer or Jenner. In that world, Democrats are kicking themselves because if you'd put Lieutenant Governor Eleni Kounalakis on the second question, she probably could have eked out a plurality win. That's what Nate Silver had in mind when he criticized California Democrats for adopting this approach, and David Shor, who is normally my guru on these matters, says it was still a mistake in expectation.

But Aaron Strauss, another of my go-to political operators, disagrees, pointing to the fact that this would have been a difficult coordination problem.

Imagine a world in which Kounalakis is on the ballot, and that tempts a more leftwing challenger to run, too. The nightmare scenario would be someone high-profile like Barbara Lee or Ro Khanna, but even someone much lower profile like DSA-endorsed state assemblyman Alex Lee could have thrown his hat in the ring and energized his leftist base. Generating even a small cohort of progressives who actually want to vote yes on Question 1 because they like one of the Question 2 options better than Newsom could have sunk him. And the more it looks like you might go to Question 2, the more incentive there is for other Democrats to get in the race.

The point isn’t necessarily that Kounalakis on the ballot would have doomed Newsom, but that coordination is important. And that the simple “no” is probably the easiest coordination point to find.

I think Shor’s view is still correct. The optimal strategy would have been to coordinate on a single plausible contender for number two, but it all comes down to how much coordination is possible. Which is another way of saying that this is a rotten system. The outcomes are highly unstable and basically depend on parties’ ability to solve difficult coordination problems outside the main constitutional framework.

Some better ideas

I do not love the idea of recall, but one obvious alternative to this shit show would be to have a recalled governor be automatically replaced with the Lieutenant Governor.

In the federal context, this idea of a “backup president” is supposed to (but in practice does not) offer a safety valve in the context of high-level official misconduct. In theory, the question on the table during the first impeachment of Donald Trump was whether or not Mike Pence should become the president. The limited nature of this remedy should have tempered both liberal enthusiasm for impeachment (since liberals would have hated a Pence administration) and conservative inclination to defend Trump. The actual thing Trump was accused of doing had nothing to do with conservative ideology or the GOP agenda, and both of those things could have been easily advanced by a Pence administration.

On a practical level, that doesn’t work. There’s far too much partisan polarization for members of Congress to see the impeachment question as simply being about the elevation of the vice president.

Taking that kind of process out of the hands of the legislature and into the hands of the citizens is a classic progressive-era way to limit the impact of pure partisanship. You could imagine grassroots resistance activists gathering petitions to recall Trump, somewhat over the objections of Democratic Party leaders who quietly regard Pence as harder to beat. And then you could imagine winning a recall vote reasonably handily by getting a fraction of Republicans who really hate the Democrats but have no particular brief for Donald Trump personally to vote for his ouster.

That might also have made sense in California. I think it would be incredibly reasonable for a California voter to have been angry at Newsom over the implementation of very strict Covid protocols (the state kept parks and other outdoor facilities closed in a way that was almost certainly counterproductive) while he was hanging out with lobbyists at the French Laundry. But there was nothing reasonable about the idea that Larry Elder should be governor of California. The whole thing was an attempted right-wing takeover of state government hoping to squeeze through based on public confusion. A recall that didn’t have partisan stakes would be better suited to punishing individual misconduct.

Non-laboratories of democracy

I am not a big fan of America’s presidential system, which I think is markedly inferior to parliamentary government. I strongly recommend Tiago Ribeiro dos Santos’ recent book “Why Not Parliamentarism?” and Lee Drutman’s book-length case for proportional representation.

You can have parliamentarianism without proportional representation and PR without parliamentarianism, but they tend to go well together. Unfortunately, it’s very hard to see how big systemic reform would happen in federal politics. In state governments, though, it should be easier. The California Democratic Party is already so dominant in the state legislature that it doesn’t actually function as a single cohesive bloc on most political issues. And I think parliamentarianism — especially with multi-party coalitions — offers a much better solution than any possible recall or impeachment process to these questions of personal misconduct.

The American system treats continuance in office as the president/governor’s birthright, and either explicitly or implicitly sets an extremely high bar for removal. It’s treated as something akin to a criminal trial, with a demand for unequivocal evidence of criminality.

For lesser cabinet jobs, we don’t see it that way. The Secretary of Whatever’s job is to be a good team player whose work reflects well on everyone. In a parliamentary system, it’s at least possible for the prime minister to have that kind of relationship with the coalition of parties he or she leads. Your job is to make everyone look good and help them do well in the next election. If you embarrass everyone with your French Laundry dinner, you get dumped, not by your enemies but by your own people who prefer a leader who isn’t dogged by scandal.

It’s always disappointing to me that across our 50 states there is so little experimentation with different ways of arranging institutions. And what experimentation there is doesn’t get treated in the spirit of experimentation. I don’t know anyone in California politics who thinks their state’s current system is particularly well-designed, functional, or a good model for elsewhere. But there’s never enough energy or discipline to actually change it. And that’s a shame.

SHARE

VISIT WEBSITE