We’ve shown several times, both on this blog and on our sister blog, the Illinois Supreme Court Review, that there are two key insights one can gain from oral argument: who is likely to win, and how each Justice is likely to vote.
Is it possible to refine the model further to where we can make a reasonable inference as to the likely winner merely by counting a single Justice’s questions?
Here’s the theory: step one, count the questions of each Justice and the entire Court. The party which gets the most questions is more likely to lose – as the margin increases, the likelihood of the more heavily questioned party winning falls. If a particular Justice asks significantly more questions of one side than the other – particularly if he or she substantially exceeds their trend level of questions – then the Justice is likely voting against the more heavily questioned party, and if he or she is well above trend, the Justice is likely writing an opinion.
Step two: once we determine whether a Justice is likely voting against the appellant or respondent, we look at the percentage of the Court’s divided opinions in which that Justice has voted with the majority. If the Justice appears to be voting against the appellant and is highly likely to be in the majority of a divided opinion, then we can infer that the Court may well be affirming. If the reverse, then the Court may well be reversing. So let’s review the yearly data on how often each Justice was in the majority of divided decisions in civil cases from 1990 to 2018.
A few ground rules: in order to keep the math relatively simple, a Justice is defined as being in the majority if he or she agrees completely with the judgment – even if the Justice files a separate opinion saying “I agree with the result, but not with any of the majority’s reasoning.” So if the majority affirms and an individual Justice votes to affirm in part and reverse in part, that’s not counted as being in the majority. Second, in order to avoid making these numbers artificially high (since the Court is typically unanimous in anywhere from 55 to 65% of its cases), we limit the data to non-unanimous decisions. Because we’re trying to answer a question about the individual Justices as opposed to the evolution of the Court, contrary to our usual practice, we arrange the tables by Justice to make it easier to see trends.
In Table 764, we report the data for five of the fifteen Justices who served on the Court between 1990 and 1997: Arabian, Baxter, Broussard, Brown and Chin. Justice Arabian was closely aligned with the majority of the Court during these years. With the exception of 1993, when he voted with the majority in divided civil cases 77.78% of the time, his rate was in the eighties four times (1990, 1992, 1994 and 1995) and 100% in 1996 (we don’t attribute that much significance to the numbers from Justices’ first and last years on the Court, since these data points are typically based on relatively few cases). Justice Baxter started out our study period closely aligned with the majority too, voting in the majority more than ninety percent of the time from 1991 to 1993 and over eighty percent the next two years. But with the exception of a blip from 1999 to 2001, his rate drifted downward after that – scoring in the seventies in 1996, 1998 and 2002 and in the sixties in 2003 and 2004.
Justice Broussard was far less often in the majority of divided cases during his two final years on the Court. In 1990, he voted with the majority 52.17% of the time, and in his 1991 part-year, he did so only 23.53%. Justice Brown was a relatively frequent dissenter as well, ranking in the fifties in 1997, 1999 and 2003, and in the sixties in 2002 and 2004 (although she joined the majority 72.41% of the time in 1998, eighty-four percent in 2000 and 71.43% in 2001). Justice Chin’s rate of siding with the majority was comparatively high – in the seventies in 1997, 1999 and 2002-2004 and in the eighties in 1998 and 2000.
In the next table, we review five more Justices – Eagleson, Justice (and later Chief Justice) George, Kaufman, Kennard and Lucas. In his final year on the Court, Justice Eagleson voted with the majority in divided civil cases 86.96% of the time. Chief Justice George’s rate was very high throughout – in the seventies twice (1995 and 2000), in the eighties three times (1993, 2001 and 2003), in the nineties six times (1992, 1994, 1996, 1998-1999 and 2002), and 100% three times (his part-year in 1991 ,1997 and 2004). Justice Kaufman’s rate was 100% in his final divided civil cases in 1990. Justice Kennard’s rate went through two phases – from 1991 to 1997, she was in the fifties five times (1992-1994 and 1996-1997) and in the sixties once. But after 1997, she was in the seventies four times (1998-1999, 2001 and 2004), in the eighties in 2002 and the nineties in 2003. Chief Justice Lucas was nearly always in the majority in divided cases between 1990 and 1994 (in the nineties in 1990 and 1992-1994 and 100% in 1991), but he drifted down a bit in the final two years on the Court (79.31% in 1995 and 75% in 1996).
Finally, we review the 1990-2004 data for Justices Moreno, Mosk, Panelli and Werdegar, plus the collective data for pro tem Justices.
Let’s review Justice Mosk and his successor, Justice Moreno, together. With only occasional exceptions, Justice Mosk was comparatively seldom in the majority in divided civil cases as the period started – 39.13% in 1990, fifty percent in 1991, in the forties 1992-1994, 1996 and 1998, but then his trend numbers drifted up, reaching the sixties in 1995, 1997 and 1999-2000 and seventy percent during his final year of 2001. Justice Moreno, on the other hand, was in the majority of divided civil cases 84.21% of the time in 2002, 88.24% in 2003 and 95% in 2004. Justice Panelli’s numbers were similar – 91.3% in 1990, 90.91% in 1991, 89.66% in 1992, 85.19% in 1993 and 80% in 1994. Justice Werdegar’s rate was typically quite high during these early years, with relatively frequent dips – between the high seventies and low nineties in 1994-1995, 1997-2000 and 2002-2004, but 69.23% in 1996 and 47.62% in 2002.
That brings us to the pro tems. On the one hand, since pro tems typically sit for only one case (there are occasional exceptions when a recently retired Supreme Court Justice sits, technically as a pro tem, in multiple cases which were largely worked up in the closing months of his or her tenure), one might speculate that the permanent Justices know them less well, and that might affect the pro tem’s ability to bring the others around to his or her preferred result. On the other hand, since pro tems tend to dissent less often than the permanent Justices, one might expect their rate to be quite high. Or, since our pro tem category encompasses a fairly large number of individuals, one might guess that the numbers will be all over the board.
In fact, the result doesn’t entirely match any of these speculations. With the exception of 1992, pro tems had a quite high majority rate from 1990 to 2000 – 100% in 1993, 1995-1996 and 1999, in the eighties in 1991, 1994 and 1997-1998 and 71.43% in 1990. But then, it dipped sharply – 33.33% in 2001, only 20% in 2003 and fifty percent in 2004 (there were no pro tems in 2002).
Join us here next time as we turn our attention to the data for the years 2005 to 2018.