Who is Brett Kavanaugh, and how will he vote if he becomes America’s next Supreme Court justice? There is no shortage of speculation about his personal or political worldview — but it’s not especially uniform. In the days after he was announced, court-watchers mined Kavanaugh’s old opinions and emerged with conflicting accounts of his ideology. His abortion rulings are simultaneously too liberal and too conservative. He’s both opposed to and supportive of the Affordable Care Act. He’s most similar to Chief Justice John Roberts and most similar to Justice Clarence Thomas.
It seems like there should be a better, more systematic way to evaluate Kavanaugh’s political ideology than hunting for clues in past rulings. It’s easy enough to do this for justices once they’re seated on the Supreme Court, but for relatively obscure federal appeals court judges like Kavanaugh, the ideological signals are dimmer, and quantitative political scientists must get creative.
They’ve found indicators about ideology in judges’ votes, hiring decisions, political donations, the text of judges’ rulings, and the language used to talk about them. It’s impossible, of course, to predict with certainty how Kavanaugh will vote if he ends up on the Supreme Court. But taken together, these tools can provide us with a general guide for where Justice Kavanaugh might land. Will he be another Roberts? Another Thomas? Or something in between?
Using judges’ votes as a proxy
The gold standard for measuring the ideology of Supreme Court justices is by looking at judges’ votes on cases. Its measurements are called Martin-Quinn scores, and they place the justices on a familiar left-right spectrum.
It is not that simple for other federal judges like Kavanaugh, who sits on the District of Columbia Circuit Court of Appeals. “Right now, we have nothing perfect for the lower courts,” said Maya Sen, a political scientist at Harvard. There are a few thorny reasons why.
A big part of the problem is that lower court judges’ ideology isn’t always as visible in their rulings. For one thing, lower court judges aren’t supposed to contradict the Supreme Court. “If you’re a circuit judge, you might think a Supreme Court precedent is horribly wrong, but you don’t have the ability to do anything about it,” said Josh Blackman, a professor of law at the South Texas College of Law Houston. This means that circuit court judges who want to adhere closely to Supreme Court precedent may frequently need to set their ideology aside when considering cases. And while the Supreme Court takes cases only when there is a genuinely unresolved legal dispute, lower court judges mostly deal with dry, straightforward legal questions where the answer may not tell us much about a judge’s ideology.1
One alternative to trying to code lower court judges’ rulings as liberal or conservative is to look at a simpler measure — how often they disagree with the majority. According to economists Elliott Ash and Daniel Chen, Kavanaugh cast a dissenting vote in 7 percent of the published cases he heard on the D.C. Circuit between 2006 and 2013. By contrast, only 3 percent of the overall votes during his tenure were dissents. And Kavanaugh was especially likely to dissent when he was overruled by two of his Democrat-appointed colleagues. (Circuit court judges mostly decide cases on three-judge panels.)
Tying these findings to ideology, though, is tricky. At best it just shows that Kavanaugh disagreed with his colleagues more often, not how he disagreed with them.
Using politicians’ ideology as a proxy
One of the oldest and most widely used methods for measuring judicial ideology is based on a simple idea: A judge’s views on key political issues are likely to be fairly similar to those of the people who appointed him. In 2007, a group of political scientists and legal scholars proposed that lower court judges’ ideology could be gauged using a combination of factors: the ideology score for the president at the time of the judge’s appointment and the ideology scores for the senators from the president’s party in the state where the judge resides. (If both home-state senators share the president’s party, their scores are averaged; if neither senator is from the president’s party, the president’s score alone is used.) This latter piece draws on the norm of “senatorial courtesy,” where senators generally do not vote for nominees who are opposed by senators who share the president’s party and are from the nominee’s home state.
These Judicial Common Space scores give us a rough sense for how a lower court judge like Kavanaugh might vote relative to the current justices if he or she were confirmed to the Supreme Court. According to the JCS scores, Kavanaugh would land far to the right, just to the left of the arch-conservative Thomas.
Because Kavanaugh was appointed to the District of Columbia Circuit Court of Appeals, he has no home-state senator for the purposes of JCS, which means that his score is based entirely on the ideology of his appointing president in the year he was confirmed. In Kavanaugh’s case, that’s George W. Bush in 2006. It is a thin basis on which to project a judge’s ideology.
But Andrew Martin, one of the creators of the Martin-Quinn scores and the JCS scores, remains optimistic about their usefulness. He pointed out that there is a high correlation between the JCS scores of circuit court judges who were subsequently appointed to the Supreme Court — such as Ruth Bader Ginsburg, Stephen Breyer and Clarence Thomas — and their Martin-Quinn scores in their first year on the court, suggesting that the JCS scores have been fairly accurate predictors of ideology.2
Using money and clerks as a proxy
Another approach is to follow the money — specifically, campaign donations by judges. For this, Sen and other political scientists and legal scholars have mined an enormous data set of political contributions by public figures, including federal judges. According to these scores, Kavanaugh sits roughly in the middle of the ideological distribution of conservative judges.
The problem is that there are limits to how much a political donation can tell us about any person’s political views. “One interesting thing about Kavanaugh is that he’s actually donated to Democrats and Republicans, which is unexpected,” Sen said. It gets even more interesting. According to Sen, the one Democrat whom Kavanaugh has donated to is Richard Cordray of Ohio. Cordray, like Kavanaugh, was a law clerk for Justice Anthony Kennedy, whom Kavanaugh has been nominated to replace. But Cordray was also the director of the Consumer Financial Protection Bureau, which was subjected to a scathing dissent written by Kavanaugh this year.
While writing a paper on campaign contributions and judicial ideology, Sen and her co-author, Adam Bonica, debated whether to remove this Democratic contribution from their ideology calculations. Sen told us that with it, Kavanaugh appears fairly moderate. Without it, he appears “quite conservative.” They decided to leave it in.
In an effort to get a different predictor of ideology — with a bigger data set to draw on — Sen and her co-authors also took a look at the political leanings of judges’ clerks. According to Sen, Kavanaugh’s clerk-based ideology score is less ambiguous than his personal donation-based score. His clerks might be diverse, but they are very conservative, she said.
This method is helpful in several key ways. The first is that, unlike the JCS scores, a clerk-based metric can capture how a nominee’s ideology has changed over the course of a judge’s career because as a judge hires more clerks, the scores can be updated. The score is also based on a much larger pool of data. Judges hire multiple clerks each year, enabling the data set for each judge to grow over time.
As innovative as this method is, though, there are downsides. Although many judges look for at least some clerks who are ideologically similar, some may also hire “counter-clerks” — people with opposing ideological views who can challenge and sharpen the judges’ arguments. And lower-level judges, including district court judges, may simply care less about their clerks’ views because so few of their cases hinge on contested legal questions.
Using judges’ language and evaluations as a proxy
So what if we stop looking at the people in a judge’s personal orbit and instead try to get as close to their ideology as we can by looking at reports from the courtroom or the language of their decisions? A new approach examines the text of reports written about judges by an assortment of lawyers who have argued in their courtrooms. These are compiled in a legal almanac with entries that read like a Zagat guide for judges, with snappy quotes about their friendliness and hostility to a host of legal issues.
Two political scientists, Adam Feldman and Kevin Cope, created separate dictionaries with words and phrases associated with liberal and conservative rulings and used them to analyze the reports and create ideology scores for each judge. They’re still in the process of conducting their analysis, but Feldman said that according to their method, Kavanaugh falls in the middle of the D.C. Circuit’s conservative bloc. (They don’t have a comparison yet for the Supreme Court.)
The methodology’s biggest downside, Feldman said, is its reliance on the editors of the legal almanac. “We can’t tell exactly who was surveyed to create these reports or how they were compiled,” he said.
Other scholars have used text-based analysis in a different way, by comparing the kinds of language judges use. Do they frequently quote conservative precedents? Or use economics-based language? Ash and Chen compared Kavanaugh’s writing style to the sitting Supreme Court justices and found that it’s most akin to Alito’s.
Some remain skeptical of text-based measurements for judges. They might work well in Congress, Sen said, where members use colloquial language in floor speeches that is readily identifiable as conservative or liberal — “death tax” and “estate tax,” “fake news” and “alternative facts,” or “pro-life” and “anti-choice,” for example. Legal language, however, is jargony and typically avoids these colloquialisms, making it much harder to tease an ideological signal out of the noise.
All of these approaches add up to one general conclusion: If confirmed, Kavanaugh is likely to be a very conservative justice in the mold of Alito or Neil Gorsuch. Predicting exactly where he’ll fall is harder to do, but given the court’s composition, it doesn’t matter whether he’s the second- or third-most conservative member of the court — the question is whether he’ll join the solid bloc on the right-most side of the court or stake out territory in the center-right, like Roberts.
For some court-watchers, though, the clue-hunting method remains more appealing than cobbling together the results of several clever — but imperfect — quantitative analyses. Blackman, the law professor, said he viewed Kavanaugh as most similar to Roberts. Since we’re natural empiricists here at FiveThirtyEight, we asked why. “You won’t like my answer,” he said. It was simply that he’d been following Kavanaugh’s career closely for over a decade.