pageok
pageok
pageok
Measuring Ideological "Activism" in Supreme Court Justices' Decisions to Overrule Regulatory Agency Actions:

Cass Sunstein is a top-rank scholar, and I always learn a lot from his work, even when I think he is ultimately wrong. Unfortunately, however, I fear his recent study (coauthored with Thomas Miles) of judicial "activism" in Supreme Court justices' votes on cases reviewing federal regulatory agency decisions doesn't tell us as much about the subject as one might like. There are three major problems: the way they analyze agency decisions, the failure to consider the possibility that there are centrist ideologies as well as "liberal" and "conservative" ones, and their interpretation of Justice Stephen Breyer's record (the justice they claim is the most "restrained").

Essentially, Miles and Sunstein first code each agency decision as "liberal" or "conservative" (e.g. - a decision to reject an environmental group's claim that a regulation isn't broad enough is "conservative"), and then try to determine how "activist" each justice is by seeing how often they vote to overrule agency decisions. Lastly, they measure how ideologically "partisan" the justices are by determining how often each one voted to overrule "conservative" decisions versus "liberal" ones. Justice Kennedy comes out as the most ideologically "neutral" justice because he voted to overrule conservative agency decisions about as often as liberal ones, while Breyer comes out as the most "restrained" (because he was the least likely to vote to overrule agencies overall).

Here are my three reservations about this framework:

I. What about the Distribution of Agency Errors?

When the Supreme Court reviews agency actions, what they are trying to do is determine whether the agency erred in its interpretation of the federal statute that grants it the regulatory authority in question. Miles and Sunstein implicitly assume that agencies are equally likely to err in a conservative direction or a liberal one. Thus, if Justice Thomas votes to overrule liberal decisions far more often than conservative ones, that shows his ideological "partisanship" in a conservative direction. However, if agencies are more likely to err in a liberal direction than a conservative one - then perhaps his voting pattern simply reflects the distribution of agency errors, not Thomas' ideological biases. Maybe Thomas is the only one fully able to set aside his own biases and focus only on the "true" pattern of agency behavior. Miles and Sunstein's approach can't rule this possibility out. And the same goes for Justice Stevens' apparent tendency to overrule conservative decisions far more often than liberal ones; perhaps that's just a reflection of the agencies' pattern of errors (which may tilt more in a conservative direction), not Stevens' biases. I won't go into detail here, but there are many theories of agency behavior that posit that liberal errors are more common than conservative ones or vice versa.

II. What about Centrist Ideologies?

Miles and Sunstein's framework assumes that there are liberal and conservative ideological biases, but doesn't consider the possibility of centrist biases. Thus, Justice Kennedy is seen as ideologically "neutral" because he votes equally often to strike down liberal and conservative agency decisions. But let's assume for a moment that Kennedy is an ideological centrist, and all he cares about is enforcing that ideology against agency decisions that deviate "too much" from his centrist preferences. He's willing to overrule both deviations in a conservative direction and those that go in a liberal one. Commitment to imposing ideological centrism could explain Kennedy's voting record just as readily as ideological "neutrality." I don't myself know which of these explanations is accurate (perhaps neither is). But the Miles-Sunstein methodology can't distinguish between them.

III. The Special Case of Justice Breyer.

Miles and Sunstein praise Justice Breyer as "the champion of modesty and restraint" because he is the least likely to vote to strike down agency actions overall. Perhaps Breyer deserves this praise. However, it's important to remember that Breyer has a strong ideological commitment to regulatory agency autonomy. In various writings, such as his 1993 book Breaking the Vicious Circle, Breyer has argued that regulatory agencies staffed by expert bureaucrats should be given greater autonomy in order to insulate them from what he regards as harmful pressure from the democratic process. He also believes (for similar reasons) that they should to a large extent be insulated from judicial review. Breyer is a leading defender of the "rule of experts" theory of governance - at least when it comes to regulatory policy. This belief is no less an ideological commitment than is a commitment to conservatism, liberalism, or centrism.

Is Breyer's apparent reluctance to overrule regulatory agency actions driven by "modesty and restraint" or by his ideological commitment to governance by expert regulators relatively insulated from outside control? I honestly don't know the answer; quite possibly both factors are at work. Or maybe neither is. But Miles and Sunstein's methodology can't distinguish between the two.

Mark Field (mail):

if agencies are more likely to err in a liberal direction than a conservative one - then perhaps his voting pattern simply reflects the distribution of agency errors, not Thomas' ideological biases. Maybe Thomas is the only one fully able to set aside his own biases and focus only on the "true" pattern of agency behavior. Miles and Sunstein's approach can't rule this possibility out. And the same goes for Justice Stevens' apparent tendency to overrule conservative decisions far more often than liberal ones; perhaps that's just a reflection of the agencies' pattern of errors (which may tilt more in a conservative direction), not Stevens' biases.


I can see this as a theoretical problem, but surely it can't be true that during the same time frame agencies demonstrate a pattern of both liberal and conservative bias. Thus, assuming a constant time frame, one of the following should be true: (a) Thomas is the most partisan; or (b) Stevens is the most partisan.


there are many theories of agency behavior that posit that liberal errors are more common than conservative ones or vice versa.


I seriously doubt that any theory positing this could be coherent, adjusting for time and the relevant administration. The terms "liberal" and "conservative" lack sufficiently well-defined meanings to make such a "study" meaningful.
10.24.2007 4:31pm
PLR:
Reservation I: This seems to require a judgment as to whether the agency was "right" in the first place, which will be just as problematic to determine as whether the justices are "right" in upholding or invalidating.

Reservation II: I don't see how Kennedy's personal motivations affect the numerical analysis. Coding a centrist position would be pretty hard, since virtually all Supreme Court appeals involve two competing positions, not three.

Reservation III: I am pretty sure that the authors are not using "champion of modesty and restraint" in anything close to a literal sense. See also II above re personal motivations..
10.24.2007 4:31pm
Ilya Somin:
I can see this as a theoretical problem, but surely it can't be true that during the same time frame agencies demonstrate a pattern of both liberal and conservative bias. Thus, assuming a constant time frame, one of the following should be true: (a) Thomas is the most partisan; or (b) Stevens is the most partisan.

That is perhaps true. But note that Miles and Sunstein claim that BOTH Stevens and Thomas are highly "partisan." I'm pointing out that, depending on the underlying pattern of agency error, maybe only one of them is.
10.24.2007 4:38pm
Ilya Somin:
I don't see how Kennedy's personal motivations affect the numerical analysis. Coding a centrist position would be pretty hard, since virtually all Supreme Court appeals involve two competing positions, not three.

It's true that there are usually only 2 options (uphold or strike down the agency action), though the Court could also strike down part of the agency's decision and uphold part. But there are more than two possible ideological motivations for making the decision. Miles and Sunstein claim to be analyzing the reasons for the decision.
10.24.2007 4:40pm
Ilya Somin:
there are many theories of agency behavior that posit that liberal errors are more common than conservative ones or vice versa.



I seriously doubt that any theory positing this could be coherent, adjusting for time and the relevant administration.


Sure it could. For example, standard public choice theory holds that agencies want to expand the scope of their authority, which implies a bias towards "liberal" decisions in Miles and Sunstein's terminology. By contrast, Ralph Nader and others claim that agencies are systematically captured by business interests - even in Democratic administrations. This implies a stable bias towards "conservative" outcomes.
10.24.2007 4:42pm
frankcross (mail):
Ilya, your point about agency directional bias would matter if cases came to the Supreme Court randomly. But they do not. These are self-selected by the same justices rendering the decision and presumably involve the most difficult legal questions. Where it is harder to see a systematic error in the cases reviewed.
10.24.2007 4:48pm
Mike S.:
You cannot meaningfully measure activist decisions unless you have something like an operational definition of what one is. That is, a definition that tells you how to tell whether a given decision is activist or not. No one has proposed that, as far as I know. Saying a decision against that demanded by the Constitution boils down to "one I disagree with" unless you have some algorithm for determining what that is without needing a Supreme Court to tell you. After all, the Justices on both sides will tell you that their opinion is the one demanded by the Constitution (or the statutory language.) Unless you have some object measure for telling which is right (in which case any Justice who ever voted wrong should be impeached) you can't sensibly measure who is activist.

By the way, if you mean by activism imposing your policy preferences on the Constitution, even if you can't measure it, upholding an error by an agency is as just as activist as striking down a correct action, isn't it?
10.24.2007 4:51pm
Dilan Esper (mail) (www):
I agree with Prof. Somin that if activism is supposed to involve respect for democratic decisionmaking, Breyer's the worst possible poster child. Not only does he love regulatory expertise, but he loves expertise in other areas— he thinks the sentencing commission is more qualified than criminal juries to decide how long someone should be thrown in prison for, the federal election commission is better qualified than the public to decide what political speech is constructive to the debate, the FDA is more qualified to decide what medical treatments we should take than patients and their doctors, etc.

It's not that I always disagree with Breyer— far from it— but rather that this sort of deference to "experts" is far from the sort of respect for the decisions of private individuals, citizen juries, and the market that "restraint" is supposed to embody.
10.24.2007 5:01pm
JohnAnnArbor:

Breyer has argued that regulatory agencies staffed by expert bureaucrats should be given greater autonomy in order to insulate them from what he regards as harmful pressure from the democratic process. He also believes (for similar reasons) that they should to a large extent be insulated from judicial review.

In other words, shut up and take what your betters dish out, peon. That that is considered a good thing by some professors is scary.
10.24.2007 5:09pm
Mark Field (mail):

For example, standard public choice theory holds that agencies want to expand the scope of their authority, which implies a bias towards "liberal" decisions in Miles and Sunstein's terminology. By contrast, Ralph Nader and others claim that agencies are systematically captured by business interests - even in Democratic administrations. This implies a stable bias towards "conservative" outcomes.


I think your two examples support my point rather than your own.
10.24.2007 5:12pm
Adam J:
Ilya- I'm not sure I understand your point about Breyer- so because he has an ideological belief that calls for restraint he's not really using judicial restraint? Simply because his ideological beliefs coincide with judicial restraint doesn't seem to me to make it any less judicial restraint when he chooses not to overturn an agency action.
10.24.2007 5:18pm
Dan Simon (mail) (www):
Breyer is a leading defender of the "rule of experts" theory of governance - at least when it comes to regulatory policy. This belief is no less an ideological commitment than is a commitment to conservatism, liberalism, or centrism.

It's not an ideological commitment at all, any more than "judicial activism" or "originalism" are. It's a meta-political position, regarding how governmental processes should work, not a political position regarding what the outcomes of those processes should be.

But the Breyer example does underline the flaw in the Miles/Sunstein methodology, at least with respect to measuring judicial activism. They define "activism"--quite correctly, in my view--as the vigorous exercise of the Supreme Court's power to override the elected branches of government, replacing those branches' decisions with the Court's. But when a Supreme Court justice votes to overturn an executive branch decision, citing language in a relevant statute, the justice may in fact be exerting his or her judicial power to override the elected branches of government--or he or she may simply be defending the prerogatives of one elected branch (the legislature) against encroachment by another (an executive failing to accede to the legislature's expressed intentions).

It's only when a justice votes to overturn a statute on Constitutional grounds--after it has been approved by both elected branches of government--that we can be sure that he or she is trying to use the Supreme Court's powers to thwart the wills of the elected branches.
10.24.2007 5:24pm
NickM (mail) (www):
The difference in the degree of control the President has over an agency can allow some agencies to be conservative while others are liberal at the same time.

Nick
10.24.2007 6:03pm
DCraig:
I'd also like to point out the problem that I had with the study. I understand that there is a view that many people have of the agencies in the executive branch of governments are completely responsive to the office of the executive. However, this isn't necessarily the case in practice. As a really great book by Heclo points out, a fairly large number of secretaries or directors rely upon the knowledge base and skill set of civil service careerists to actually be responsive. Without getting into the whole Wilsonian politics vs. administration dichotomy, I think it is kind of intellectually dishonest to label all agenciy actions as either "liberal" or "conservative." There are many agency careerists who are themselves centrists or independents and don't have any motivation in deciding agency decisions other than, at worst, the agency's best interest, and at best, the public's. To code every agency decision as motivated by one of these two ideology paints the bureaucracy in very broad strokes.
10.24.2007 6:54pm
frankcross (mail):
The study isn't about the motivations of the agency action, just its ideological direction. So if an environmental agency adopts a new regulation that is challenged by industry, an affirmance would be liberal (and deferential) while a reversal would be conservative (and not deferential).

The study implies nothing about agency intent. Though it is interesting to look for effects like liberal decisions from conservative Administrations and vice versa
10.24.2007 8:28pm
Ilya Somin:
I'm not sure I understand your point about Breyer- so because he has an ideological belief that calls for restraint he's not really using judicial restraint? Simply because his ideological beliefs coincide with judicial restraint doesn't seem to me to make it any less judicial restraint when he chooses not to overturn an agency action.

It depends on what one means by "restraint." If restraint is simply choosing not to overrule an agency for whatever reason, then the reason doesn't matter. If "restraint" is defined as following the law rather than one's ideological preferences or as "modesty" (to use Miles and Sunstein's term), then the motive does matter. "Restraint" that consists of efforts to empower political actors whom you favor is not truly judicial restraint in most ordinary usage senses of the term.
10.24.2007 9:19pm
Ilya Somin:
Breyer is a leading defender of the "rule of experts" theory of governance - at least when it comes to regulatory policy. This belief is no less an ideological commitment than is a commitment to conservatism, liberalism, or centrism.

It's not an ideological commitment at all, any more than "judicial activism" or "originalism" are. It's a meta-political position, regarding how governmental processes should work, not a political position regarding what the outcomes of those processes should be.


Much ideology in fact consists of "metapolitical positions" about how government processes should work. Usually such positions are based on a belief that these processes are more likely to produce good outcomes as defined by the ideology in question. Ideology enters into assessments of processes no less than assessments of results.
10.24.2007 9:21pm
Adam J:
Ilya- So apparently it's Breyer's motives that you find fault with, but isn't the whole reason that restraint is virtue is because it "empowers [more democratic] political actors," and because judges are not creating laws that reflect their ideological beliefs.

As you said, Breyer is empowering political actors (that are at least marginally more democratic). Also, Breyer doesn't favor agencies because they create laws that reflect his ideology, he defers to them because they're the experts and he's a mere generalist. I suppose you can argue that this choice is ideology, but under such a broad definition a choice to be deferential itself is ideology- heck under this definition "modesty" is ideology.

On the other hand, if you want to argue that Sunstein is cherry picking, because Breyer's restraint only exists for agencies, but not to the states... that's an argument that I'm open to.
10.24.2007 10:06pm
frankcross (mail):
Much ideology in fact consists of "metapolitical positions" about how government processes should work. Usually such positions are based on a belief that these processes are more likely to produce good outcomes as defined by the ideology in question. Ideology enters into assessments of processes no less than assessments of results.

Logically, one would think this is so. But it hasn't been established, at least not on the Supreme Court level. I.e., you would think that federalism would produce results here. But they are pretty weak. The justices tend to vote their political ideologies more than federalism. Of course, they say that in some cases federalism principles aren't really applicable. But those tend, remarkably, to be cases where the ideological implications of the particular case seem contrary to those of the justice. Of course, this is not absolute, but still statistically significant.
10.24.2007 11:29pm
Dan Simon (mail) (www):
Much ideology in fact consists of "metapolitical positions" about how government processes should work. Usually such positions are based on a belief that these processes are more likely to produce good outcomes as defined by the ideology in question. Ideology enters into assessments of processes no less than assessments of results.

If that were so, then every school of constitutional interpretation would be just another ideology--that is, a preference for government processes that follow the constraints defined by the interpreter's construction of the Constitution. Thus every decision by a Supreme Court justice would be inherently ideological, and every Supreme Court justice an ideologue by definition.

Alternatively, we can choose a non-vacuous definition of "ideology", one that excludes preferences for specific governmental processes irrespective of those processes' outcomes.
10.25.2007 1:33am
Ilya Somin:
As you said, Breyer is empowering political actors (that are at least marginally more democratic). Also, Breyer doesn't favor agencies because they create laws that reflect his ideology, he defers to them because they're the experts and he's a mere generalist. I suppose you can argue that this choice is ideology, but under such a broad definition a choice to be deferential itself is ideology- heck under this definition "modesty" is ideology.

Yes, ideology does underly all theories of judicial interpretation at some level. After all, there is no completely nonideological theory of such concepts as the rule of law, the authority of the constitutional text, etc. However, a belief in "rule of experts" because that is one's personal opinion about the best way to structure political processes is far less "restrained" than a belief that leaves the formation of political processes up to the text of the Constitution and of statutes enacted by Congress.
10.25.2007 1:52am
Ilya Somin:
If that were so, then every school of constitutional interpretation would be just another ideology--that is, a preference for government processes that follow the constraints defined by the interpreter's construction of the Constitution. Thus every decision by a Supreme Court justice would be inherently ideological, and every Supreme Court justice an ideologue by definition.


Yes, as I noted in my last comment, ideology is a component of every theory of interpretation at some level. However, an ideology such as Breyer's commitment to deference to "expert" agencies is qualitatively different from one that is based on a more general theory of interpretation that applies the same rules to all political actors.



Alternatively, we can choose a non-vacuous definition of "ideology", one that excludes preferences for specific governmental processes irrespective of those processes' outcomes.

There is a big difference between an ideologically based commitment to following the text of statutes and of the Constitution and an ideological commitment to one's own vision of what the best governmental processes are - independent of what the text of the Constitution or statute requires.
10.25.2007 1:55am
Ilya Somin:
One other point on Breyer: his approach does not reinforce democracy, but actually undermines it to the extent that it privileges the decisions of unelected agency bureaucrats over those of elected members of Congress.

If an agency's interpretation of a statute deviates from the meaning enacted by Congress, democracy is (usually) best served by overruling the agency and forcing it to bow to the text of the statute enacted by the legislature.
10.25.2007 2:21am
frankcross (mail):
The trouble with the latter argument is that Congress can overrule the agency, if it wants to. If the judiciary overrules the agency, purportedly at the behest of Congress, but incorrectly, that would thwart the will of Congress. But of course the incorrect judicial decision could be overruled by Congress too. So I see no congressoinal power difference here.
10.25.2007 10:56am
Adam J:
"One other point on Breyer: his approach does not reinforce democracy, but actually undermines it to the extent that it privileges the decisions of unelected agency bureaucrats over those of elected members of Congress."

His method does not privilege decisions of agencies over elected decisions- he's never said that agencies can make decisions that aren't supported by Congress, only that he's not going to put his view of what Congress said ahead of the agencies view of what Congress said.

"If an agency's interpretation of a statute deviates from the meaning enacted by Congress, democracy is (usually) best served by overruling the agency and forcing it to bow to the text of the statute enacted by the legislature."

You're right about this of course... but I don't think you can argue that the court is always doing this when it overrules, it's quite possible the court is the one wrongly interpreting the statute. If the court was already right, then activism would probably be viewed as more of a virtue.
10.25.2007 12:10pm
Adam J:
If the court was ALWAYS right... that's what I meant
10.25.2007 12:11pm
Dan Simon (mail) (www):
However, an ideology such as Breyer's commitment to deference to "expert" agencies is qualitatively different from one that is based on a more general theory of interpretation that applies the same rules to all political actors.

Qualitatively different in what sense? That it fails to conform with your ideological preference for treating all political actors the same?

There is a big difference between an ideologically based commitment to following the text of statutes and of the Constitution and an ideological commitment to one's own vision of what the best governmental processes are - independent of what the text of the Constitution or statute requires.

Yes--you approve of the former, but not the latter. Otherwise, it's all just ideology--right?

One other point on Breyer: his approach does not reinforce democracy, but actually undermines it to the extent that it privileges the decisions of unelected agency bureaucrats over those of elected members of Congress.

I agree completely--but that's my beef with Breyer, judicial minimalist and democracy-lover that I am. Why on earth would you object to his views? Not only is he no more ideological than you are (since all theories of Constitutional interpretation are, to you, inherently ideological)--he's actually just as suspicious as you are of democratic decision-making driven by the preferences of rationally ignorant voters. Based on everything you've said, you should love the guy. Or are you just too much of a results-oriented ideologue to recognize a meta-level kindred spirit when you see one?
10.25.2007 1:30pm