Supreme Court agreement


We tend to think of the US Supreme Court as being sharply split along ideological lines. But as this article points out, that is because the media focuses on a few high profile cases which have an elevated profile precisely because they are split that way. In actuality, there is a remarkable degree of agreement among the justices in their rulings, with even the lowest level being 66% between Ginsburg and Thomas

The Ginsburg, Kagan, Sotomayor bloc shows a slightly higher level of cohesion than the Scalia, Thomas, Alito bloc.

US Supreme Court votes

The justices are one of the lasting legacies of the presidents who appoint them, and you might expect ones appointed by the same president to vote together particularly often. This is certainly true of the four newest justices. The ones appointed by President Obama, Justices Sonia Sotomayor and Elena Kagan, have agreed 94 percent of the time. The members of the court appointed by President George W. Bush, Chief Justice John G. Roberts Jr. and Justice Samuel A. Alito Jr., have agreed 93 percent of the time.

The tendency is less marked among justices on the bench longer. President Bill Clinton’s nominees, Justices Ruth Bader Ginsburg and Stephen G. Breyer, were less likely to agree, at 88 percent. And President Ronald Reagan’s appointees, Justices Antonin Scalia and Anthony M. Kennedy, have agreed just 82 percent of the time.

People’s views do change with time and I suspect that this is true for Supreme Court justices as well.

Some people have been urging justice Ginsburg (who at 81 is the oldest person on the court) to retire while president Obama is still president and can name her replacement without risking a Republican president to take office in 2017. But the situation in Washington now is so viciously obstructionist that it is possible that any Obama nominee will filibustered and force the court to limp along with just 8 people until a new president is sworn in. Furthermore Obama seems to not have the inclination to push through a strong progressive voice anyway.

Comments

  1. StevoR : Free West Papua, free Tibet, let the Chagossians return! says

    Don’t worry, the next US president will almost certainly* be a Democratic party one too – most likely Hillary Rodham Clinton**, you read it here first!

    (Look at the mess the Repubs are in! Sure its still a few years off but I really can’t see ’em getting their act together or putting up someone electable for 2016.)

    * Just be sure, memeo to Nader & any other third party – stay the flip outta it & don’t spoil things! (Tea party excepted – not that they’d listen to sanity anyhow.)

    ** Who should be the current POTUS really but still. Hillary Rodham Clinton beating Obama and McCain-Palin in 2008 is alternative history now. Wonder how much better off we’d all be?

  2. StevoR : Free West Papua, free Tibet, let the Chagossians return! says

    Actually given the fact that POTUS’es generally get two terms unless something exceptionally bad happens under their watch and sometimes even then (cough, Bush Jr, cough) I’m gonna call the election after 2016 for Hillary Clinton too and say I’m pretty confident (say 85 %)that she’ll get 8 years in the White House as POTUS too. Won’t predict any further ahead than that mind.

  3. Marshall says

    Hi Mano,

    Lawrence Sirovich (a mathematician with whom I’ve collaborated) wrote a PNAS paper back in 2003 about this exact issue. He ran Principal Component Analysis on the rulings of two Supreme Courts, and determined that, in general, the Supreme Courts act in the vicinity of 5 distinct judges (“eigenjudges”).

    You can check out the paper here: http://www.pnas.org/content/100/13/7432.long and the NY Times article about the paper here: http://www.nytimes.com/2003/06/24/science/a-mathematician-crunches-the-supreme-court-s-numbers.html.

  4. Chiroptera says

    Mano Singham: In actuality, there is a remarkable degree of agreement among the justices in their rulings, with even the lowest level being 66% between Ginsburg and Thomas

    I guess I’m not terribly surprised by this. I would expect that the majority of cases are either fairly cut and dry (relying on interpretation of legislation, not the Constitution) or involving fairly straight forward issue where ideology may not be a factor.

    But as this article points out, that is because the media focuses on a few high profile cases which have an elevated profile precisely because they are split that way.

    It could be that these cases are significant precisely because ideology plays a significant role in how one will view it.

  5. lpetrich says

    Very interesting work, Marshall.

    I did a cluster analysis on the OP’s matrix of levels of agreement. Some cluster-analysis algorithms can work with distance matrices, and I used them. Here is a typical one: UPGMA, a rather simple algorithm sometimes used in phylogeny. It finds the closest two, then replaces them with their average. It then repeats this operation until there is only one left. It keeps track of these replacements as a sort of family tree. Here is what I’ve found with it:
    Sonia Sotomayor — Elena Kagan — 6
    (SS,EK) — Ruth Bader Ginsburg — 8.5
    ((SS,EK),RBG) — Stephen Breyer — 11.75
    John Roberts — Samuel Alito — 7
    Antonin Scalia — Clarence Thomas — 9
    (JR,SA) — (AS,CT) — 10.75
    ((JR,SA),(AS,CT)) — Anthony Kennedy — 15
    (((SS,EK),RBG),SB) — (((JR,SA),(AS,CT)),AK) — 24.3906
    This algorithm’s results agree with more subjective estimates of the court’s alignments, though the results also contain some surprises. Antonin Scalia and Clarence Thomas vote much alike, but John Roberts and Samuel Alito also do so. Those four form a conservative bloc with Anthony Kennedy a loosely-affilated member of it. There is also a liberal bloc of four Justices, with Elena Kagan and Sonia Sotomayor voting the most alike, Ruth Bader Ginsburg voting much like them, and Steven Breyer voting a bit less like them.

  6. lpetrich says

    Looking at the PNAS paper that Marshall linked to, I checked on Table 3, and I found from it:
    The strongest component, at 0.571, is all the Justices voting the same.
    The next component, at 0.219, is liberal vs. conservative. Scalia and Thomas are the strongest, with Rehnquist not far behind, and with Kennedy and O’Connor weakly in their direction. In the other direction, Stevens was the strongest, followed by Ginsburg, and then by Breyer and Souter.

    I’ve also done that paper’s calculations for the most recent court. The information-theory number of Justices is 4.9, and PCA on the Justices’ votes revealed two strong components: one for everybody voting much alike, and one for liberal vs. conservative. Of the conservative ones, Kennedy voted only weakly like the others in that bloc. He is indeed a “swing justice”.

Leave a Reply

Your email address will not be published. Required fields are marked *