Policy Preference Change on the SCC: New paper looks at whether we can predict how appointees will rule

Consider asking yourself several questions: Is the process that has prevailed for much of the Supreme Court of Canada’s history of Prime Ministers making appointments with informal consultations with a number of interested and knowledgeable parties a good process? Should there be a more open appointments process for Supreme Court of Canada justices? If so, what precisely should this “more open” system look like? Should we adopt U.S.-style confirmation hearings of nominees to our top court? Why or why not?

If you’re like most participants in debates surrounding these types of questions, you have a strongly-held opinion about the appointments process and how it might be reformed. You may, for example, believe that appointments to the Supreme Court are largely politically motivated and, as such, ought to involve more transparency, public vetting and input. You may, on the other hand, believe that the appointments power has not been abused by Prime Ministers in the past, and that the current members of the bench, whatever your possible disagreements with them in individual cases, are a professional and competent group of jurists who have been selected for the most part on the basis of merit. If this is your view, you may fear that reforming the appointments process to allow for more public input through, for example, confirmation hearings, will lead to an overtly politicized Court, as appears to be the case in the United States.

If you have an opinion on these questions — and it’s likely that you do — it’s interesting that you seem to have been able to reach an opinion without having interrogated the available data in a systematic way. For this very reason, in a recent paper, Andrew Green and I seek to do precisely that — place the debate on a firmer empirical foundation.You shouldn’t feel too badly about leaping to your conclusions. For one thing, you have lots of company. It’s what just about everyone else who has waded into the debate has done. People formulate opinions and vigorously argue that this or that should be changed about the process without actually going to the data (or at least all the data) to check whether the imagined deficits in the prevailing processes are real or are simply that, imagined. This is a weak justification, however. It’s much better to actually take a look at what the data suggest with an open mind (to the extent this is possible).

In our working paper, we ask and try to answer the following questions: Are justices’ policy preferences correlated with the party of the Prime Ministers who appoint them, or with newspaper descriptions of their dispositions at the time of appointment? Is a justice’s voting behaviour more predictable in his or her first year on the bench than in later years? Do justices’ policy preferences change over time? If so, is there a common or predictable pattern as to how justices change?

What do we find?

Overall, there is a modest relationship between the party of the appointing Prime Minister and a justice’s subsequent voting record. In some specifications, the result was statistically significant, and in others it was not. Even where the results were significant in statistical terms, however, the substantive difference was relatively small (e.g. a 5% difference in “liberal” voting behaviour). Even among justices appointed by the same Prime Minister, there are dramatic differences; for example, we find that Brian Mulroney appointed both the most “liberal” justice and the most “conservative” justice. Similarly, Pierre Trudeau appointed the second most extreme justices on the Supreme Court according to the same methodology. Thus, the overall results of a slight correlation disguise a huge amount of variation among the individual justices.

We also find that there is surprisingly little correlation between the predictions of judicial policy orientation as contained in newspaper editorials from the time of a justice’s appointment (as reported by Ostberg and Wetstein in Attitudinal Decision Making in the Supreme Court of Canada, UBC Press, 2007) and a justice’s subsequent voting behaviour. Journalist expectations of how nominees will vote just do not match up well with how they do in fact vote.

It bears noting that there is little evidence that justices are more predictable in their first year on the Supreme Court than they are in later years. It is not that justices are predictable at the start of their careers and then become less so over time. A justice, if at all, is only slightly predictable in his or her first year on the bench in criminal matters. On top of this, judicial policy preferences unpredictably shift over time — sometimes significantly. Some justices become more “conservative”, some become more “liberal”, some stay constant, and some exhibit other patterns. For example, Justice (now Chief Justice) McLachlin appears to have started out as a “liberal” justice when she first joined the Supreme Court in 1989, became more “conservative” beginning in the early 1990s, and has returned to a more “liberal” approach in recent years.

The net result of the empirical analysis is this: despite some notable exceptions, most particularly Justice L’Heureux-Dubé, the justices of the Supreme Court of Canada do not, by and large, vote in ideologically predictable ways either in the short-term or on a long-term basis. In addition, their policy preferences are in a constant state of change; at any given time, some justices will be changing little, others will be moving in a more “liberal” direction, and others will be tending towards a more “conservative” approach. We show that justices who served from 1982-2004 do not appear to have been particularly ideologically driven, especially in comparison with their colleagues on the U.S. Supreme Court.

So what is the payoff for the appointments process debate? It is important to note that justices appointed under another system or in another time might well exhibit other traits or inclinations. All we have done is fill the lacuna relating to the voting behaviour of Supreme Court justices in the post-Charter era. In making prescriptions for changes to the process, what’s missing is counterfactual information relating to how other appointments processes would affect the composition of the Supreme Court and the conduct of its justices. At least for now, opinions formed through impressionistic reasoning and comparative conjecturing still have an unavoidable — and unfortunate — role in the appointments process debate.