Have the “Republicans” on the Supreme Court become less ideological–You might be surprised

Eric Posner, a professor at the University of Chicago Law School, has an interesting analysis about whether the Republicans on the Court are becoming less ideological. You might be surprised by his conclusion that “Roberts and Kennedy are more likely to agree with the Democrats than with the other Republicans. I don’t think this has happened before.” On the other hand, “The Democrats and Thomas follow the ideologically polarized pattern of prior years.” Interestingly, if I am reading Posner correctly, Justices Scalia and Alito are nearly neutral.

Here is the table* that Posner presents.

Justice Same-Party Agreement, % Opposite-Party  Agreement, % Propensity to Same-Party Agreement
Roberts 74.5 82.3 0.91
Kennedy 73.8 84.8 0.87
Scalia 76.5 74.3 1.03
Alito 77.3 75.3 1.03
Thomas 73.5 62 1.19
Breyer 92.7 82.5 1.12
Ginsburg 91.3 75.2 1.21
Sotomayor 91.3 74 1.23
Kagan 90.7 77.3 1.17

For a variety of reasons, a sophisticated empiricist might dismiss the foregoing as meaningless “fun facts.”  But not all “fun facts” are meaningless. Sometimes they can, precisely because of their simplicity, illustrate something important. In my opinion, this is one of those occasions.


*”The numbers in the second and third columns are averages derived from Scotusblog (scroll down to the bottom). The last column is the second column divided by the third, so 1 means greater agreement with co-partisans.”

H/T How Appealing.

13 responses

  1. The fly in this ointment is that the Republican party has veered so far to the right in the last couple of decades (Democrats have moved to the right also but not as far). Roberts and Kennedy perhaps represent a more middle of the road version of Republicanism. David Souter, appointed by Bush I, was at one time thought of as a centrist. If he were still a Supreme, Souter would probably fall somewhere Ginsburg and Breyer on the current left – right spectrum.

  2. “Democrats have moved to the right. . . . .” HA! IMHO both parties have moved farther from the middle, the donkeys to the too far left and the elephants to the too far right. Clinton was a bit of an exception and governed closer to the middle because he is a more realistic, political savy fellow. What we have now is a stubborn ideologue in the White House and a herd of asses populating the halls of Congress.

  3. It would be interesting to see those data points spread out on type of case rather than party affiliation (eg 1st Amendment cases, 4th Amendment cases, etc.).

  4. Dredd,

    Indeed. That is one of the reasons that one might call this analysis nothing more than a generator of “fun facts.” The more interesting research question, it seems to me, is where they shake out in the high profile cases dealing with hot button social issues. That said, Posner’s analysis, albeit superficial, does give us a good overall perspective.

    All the best.


  5. By treating all cases as having an equal political valance the study does not show anything other than that it shows nothing. On the other hand the problems of weighting are very difficult. On the other hand, while party affiliation is a fun club to use, done it myself, I do not think it is terribly relevant to SCOTUS except as a place holder for other divisions, nature of federalism, executive power, and what have you Votes on where is the capital of Israel is exemplary, doubt either of the D Senators from NY were thrilled with result.

  6. repentinglawyer,

    Yes, but if you consider that any case that gets to the Supreme Court is important the analysis is helpful. It may confirm what many of us suspect, and that is that for many cases in the Supreme Court ideology is not so important.

    All the best.


  7. Judge, no question that a lot of the business of SCOTUS is not highly divisive or at least not devise along strong ideological lines that would interest non lawyers, but then the public pays no attention. Doubt a lot of folks really got excited over the limits of Bankruptcy Courts as nonArticle III Courts and Abacrombie vs the headscraf though an important case on statutory protection of religious exercise did not excite many.. SCOTUS for most is only interesting when it confirms or rejects their preconceptions in cases where divisions in the Court mirror or appear to reflect partisan or ideological divisions.

  8. One potential statistical problem is that opinions ignores the issue of deciding which cases to hear. Part of the grant decision involves counting to 5. If the dynamics of the cert process lead to a large percentage of cases that split “conservative” vs. “liberal,” you would expect the four “liberals” (all Democratic appointees) and the three “conservatives” (all Republican appointees) to stick together with the two “moderates” (both Republican appointees) swinging back and forth between the two sides. The stats seem to bear this up other than suggesting that Scalia and Alito or more moderate than many folks believe (or at least that they both have some issues on which they are more moderate).

    p.s. Because of the nature of many Supreme Court cases, I am not convinced that Republican vs. Democrat or Liberal vs. Conservative are always relevant distinctions (e.g. what is the difference between these philosophies on an intellectual property case).

  9. tmm, your cert analysis is exactly correct. The four “liberals” can always cause the Court to take hot button cases, and then they need to “turn” only one “moderate” to move the Court toward the “progressive” end.

    Moreover, I particularly agree with you regarding your last point. That is, in many cases politcal or philosophical leanings don’t matter.

    Thanks for writing. All the best.


  10. A couple of things struck me: 1) The “liberal”/Dem Justices appear to be more ideologically driven than the “conservative”/GOP Justices, with Scalia, as you noted, the unexpected “neutral” exception. 2) Statistics show that many statistics don’t do anything more than show whatever someone wishes to show. Like others above, I’m not too sure how one classifies many of the cases taken by political party. And then, what about the rejected cases and who voted how on those? What about GVRs? Rehearings? Relists (and how many times, and what ultimate resolutions)? And what about the grant/deny/GVR/rehearing/etc. ratio by circuit and state? And what about…

  11. I would phrase it slightly differently. If you have a 5-4 split, the five can pretty much take any case that they want to review (knowing that they have the votes regardless of whether they can gain any of the four). As such, they do not need to look for a case that splits the minority. The four, on the other hand, need to find cases in which the lower court went too far for the tastes of the “moderates” or issues that split the majority.

    If the justices thought this way, you would expect that neither the majority nor the minority would look for cases that split the minority (resulting in the minority having a high rank for partisan loyalty when liberal vs. conservative matches the partisan affiliations of the judges, a fact that has been generally true since Stevens and Souter retired), but that the minority will get some cases heard that split the majority (resulting in a lower rank for partisan loyalty among the majority).

  12. Well, one fun fact is that Posner is no statistician. The data on Scotusblog are a mish-mash, of which Posner makes more of a mish-mash, and then draws conclusions about ideology, without any warrant.

    As for the comments about whether the Dems or Reps in general have moved to the right or to the left, the data have nothing to do with how well the justices agree with the politicians, but how often they agree to some extent with each other in their opinions. You can draw from that no conclusions about how much the opinions of the justices conform to any political idea or ideology. Unfortunately, that is what Posner does, or at least, suggests.

    It may well be that the four Dems on the Court tend to agree with each other and disagree with Thomas. But if so, why that may be the data do not indicate.

  13. Oh, yes. The comments that the choices the Court makes about which cases to hear introduce selection bias are right on. 🙂

%d bloggers like this: