Court Under Roberts Is Most Restrained in Decades

Last Sunday, the NYT published a lengthy article by Adam Liptak, titled “Court Under Roberts Is Most Conservative in Decades,” arguing that the Supreme Court, under Chief Justice Roberts, has become “the most conservative one in living memory.”  I was in Yellowstone National Park at the time, and did not have the opportunity to comment.  Yet as this question is a recurring interest of mine (see, e.g., here, here, here, and here), I thought I would now.  The post is long, so I’ve hidden it below the fold, but here is a super-short (and overly simplified) summary: Liptak’s article overstates the purported conservatism of the Roberts Court and, based on the data presented, could just as easily labeled the Roberts Court the “most restrained” or “least activist” Court in living memory.  In sum, there’s less to the article than meets the eye.

The article’s central claim is based upon political science research showing that the Roberts Court has been more likely to reach “conservative” decisions than its predecessors.  Liptak reports:

In its first five years, the Roberts court issued conservative decisions 58 percent of the time. And in the term ending a year ago, the rate rose to 65 percent, the highest number in any year since at least 1953.

The recent shift to the right is modest. And the court’s decisions have hardly been uniformly conservative. The justices have, for instance, limited the use of the death penalty and rejected broad claims of executive power in the government’s efforts to combat terrorism.

But scholars who look at overall trends rather than individual decisions say that widely accepted political science data tell an unmistakable story about a notably conservative court.

Elsewhere in the article, Liptak writes “the pace of change has been dizzying” since Justices Roberts and Alito joined the Court, but also that “the rightward shift is modest.”

Assuming that the definition of what constitutes a “conservative” or “liberal” decision is accurate (more on that below), this methodology does identify decisional trends, but does it necessary identify which courts are more “conservative”?  It depends on what one means.

A court that adopts a conservative result more often does not necessarily move the law to the right more than one with a more liberal record (a point Liptak acknowledges).  This is because a “conservative” decision may be one that overturns or modifies precedent to move the law in a rightward direction, but it may also be no more than a decision which refuses to embrace a liberal change in the law.  A court in which the latter sort of decision predominates, what we might call a conservative “minimalist” court, could be identified as more “conservative” court than one which issues a greater percentage of liberal decisions but, when it issues conservative decisions, is more likely to overturn precedent and change the law.  Yet the more conservative court in this instance would do less to make the law more conservative.  As I have noted before, there is also an important difference between a judicial decision that, say, identifies a new constitutional limitation on legislative power and one that adopts a restrained statutory interpretation, as the latter is far more deferential to the political branches and easier to correct.

This distinction is important because the data presented by Liptak suggests that the Roberts Court is such a “conservative minimalist” court.  Indeed, it appears to be the most restrained – or least “activist” (if “activism” is defined as willingness to overturn federal statutes or prior precedents) – Court since World War II.  According to the data presented with the article in this chart, the Warren, Burger and Rehnquist Courts overturned precedents at an average rate of 2.7, 2.8 and 2.4 per term, respectively.  The Roberts Court, on the other hand, has only overturned an average of 1.6 precedents per term.  The record on striking down laws shows a similar pattern.  The Warren, Burger, and Rehnquist Courts struck down an average of 7.9, 12.5, and 8.2 laws per term, whereas the Roberts Court has only invalidated an average of 3 laws per term.  Liptak acknowledges this data at the close of his article, but downplays it with his description:  “The Roberts court is finding laws unconstitutional and reversing precedent — two measures of activism — no more often than earlier courts.”

So, while the majority of the Roberts Court’s decisions are “conservative,” the data Liptak summarizes does not appear to have resulted in a more “conservative” legal regime, as the Roberts Court has done relatively little to change the law (at least thus far) compared to its predecessors.  This is important, because it effectively refutes claims that there is anything particularly radical or “activist” about the Roberts Court, even if one accepts that it is notably “conservative.”

Had Liptak engaged in a substantive examination of the Roberts Court’s decisions, he would have found a similar pattern in most areas, some aspects of criminal procedure being a notable exception.  As I discussed in this article, in many high-profile areas the substance of the Roberts’ Court’s opinions is no more conservative than that of the early Rehnquist Court.  On abortion, for instance, Gonzales v. Carhart is more “conservative” than Stenberg v. Carhart, but Stenberg arguably represented a leftward shift from Casey and Carhart merely shifted the law back.  On race, Parents Involved and Ricci may seem more conservative than Grutter, but they are not clearly more restrictive than Croson and Adarand.  There is no evidence as yet that the Roberts Court is as willing to challenge federal power as the Court was under Burger (National League of Cities v. Usery) or Rehnquist (Lopez, Morrison, Boerne).  There are exceptions, such as some of the Court’s Miranda decisions – which have certainly made the law less protective of criminal suspects and defendants – and Citizens United, but these exceptions are balanced by aggressive liberal opinions in areas like executive power and the death penalty.  In sum, even if most of the Roberts Court’s decisions are “conservative” a substantive analysis of the Roberts Court’s decisions does not reveal a significant rightward shift in the law.

The article attributes the Court’s rightward shift to changes in personnel, in particular the replacement of Justice O’Connor with Justice Alito.  There is little question that, on many issues, Justice Alito is noticeable more “conservative” than was Justice O’Connor at the time of her retirement.  This is significant because, for many years, Justice O’Connor was the Court’s median justice.  Replacing her could not help but shift the Court.  And yet the addition of Justice Alito did not make the Court meaningfully more conservative than it had been a decade earlier, largely because Justice O’Connor’s jurisprudence “evolved” in a strikingly more liberal direction during her tenure on the Court (as did Justices Kennedy and Stevens, as shown by the work of some of the very scholars Liptak relied upon for his article).  As a consequence, the Rehnquist Court drifted leftward even while its personnel did not change – and even more leftward when Justice Ginsburg replaced Justice White.  So to say the Court is now more conservative than it was in, say, 2001, tells us very little about its overall ideological trajectory.

Interestingly enough, any rightward shift has not made the Court more out of step with the American public.  As Liptak reports:

While the court is quite conservative by historical standards, it is less so by contemporary ones. Public opinion polls suggest that about 30 percent of Americans think the current court is too liberal, and almost half think it is about right.

On given legal issues, too, the court’s decisions are often closely aligned with or more liberal than public opinion, according to studies collected in 2008 in “Public Opinion and Constitutional Controversy” (Oxford University Press).

Liptak points to abortion and affirmative action as areas where the Court is “in sync” with public opinion.  He could also have talked about executive power, the death penalty, and other areas in which the Court is, if anything, still to the public’s left.

What about the underlying methodology?  Can we really describe all opinions as “conservative” or “liberal”? Here’s how Liptak describes the methodology:

votes favoring criminal defendants, unions, people claiming discrimination or violation of their civil rights are, for instance, said to be liberal. Decisions striking down economic regulations and favoring prosecutors, employers and the government are said to be conservative.

About 1 percent of cases have no ideological valence, as in a boundary dispute between two states. And some concern multiple issues or contain ideological cross-currents.

But while it is easy to identify the occasional case for which ideological coding makes no sense, the vast majority fit pretty well. They also tend to align with the votes of the justices usually said to be liberal or conservative.

This approach is certainly easy enough with most contemporary cases, but there are areas in which this is difficult.  Think of the sentencing cases.  Which side is “conservative”?  What about preemption?  We usually think of preemption as something advocated by business groups, and might be coded as conservative, but would this apply to the cases involving Arizona’s immigration laws?  And what about the dormant commerce clause?  Is the “conservative” position one that clears away state obstructions to interstate commerce? Or that which adopts a more restrained view given the doctrine’s uncertain textual provenance?

Cases that clearly have an ideological valence at the time, may seem less clear over time.  Ricci v. DeStefano was an “anti-employer” and “anti-government” decision favoring “people claiming discrimination,” and yet we would recognize it as “conservative” by contemporary standards.  But a political science grad student twenty years hence might not recognize it as such.  The same may be true for Comstock, in which the Court upheld federal power against a challenge by federal convicts.  Does that make it conservative?  I would think not.  My point is not that coding cases is indeterminate, but that trying  to maintain a consistent ideological metric over time through simple coding rules could be very difficult, and I am skeptical that such analyses give us a clear sense of whether, say, Justice Hugo Black was more or less “conservative” than Justice Kennedy.

A final note: To say the Court is more “conservative” or “liberal” is not to say anything about the quality of the Court’s work or the correctness of the Court’s decisions.  For myself, I think the Court is too liberal in some areas, too conservative in others, and just right in still others.  My interest in this area stems from a frustration with the rush to characterize the Court’s work in sweeping ideological terms and what I perceive as a fairly persistent effort to label the Court as ideologically extreme generally, and “conservative” in particular when such labels do not (yet) accurately describe the work of the Court.