Pages

Tuesday, November 25, 2014

The Realities of Administrative Discretion

Christopher H. Schroeder

For the Symposium on Administrative Reform of Immigration Law



Under our constitutional system of separation of powers, does the President have the authority to defer the deportation of the undocumented parents of children who are lawfully present in the United States, to permit these persons to apply for work authorization and also to expand the Deferred Action for Childhood Arrivals first announced in 2012?  Or is the President violating the Constitution by refusing to execute the immigration laws?

Here are three reactions to these questions. 

First, these two questions are not opposite sides of the same coin.  Whatever answer you give to the first question, the answer to the second one is a resounding NO.  The allegation that the President is violating the Constitution rests on the claim that he is violating his duty to take care that the laws be faithfully executed.  That depends crucially on what it means to “faithfully execute” a law.  In 1823, Attorney General Wirt wrote that the duty requires that the law be executed “not with perfect correctness of judgment, but honestly.”   That description makes the best sense of the language of the Constitution; it reflects the wisdom of the Founders who were practical people fully aware that all humans, including Presidents, are fallible; and it appreciates the background understandings of the time. 

The 33 page legal opinion issued by the Department of Justice’s Office of Legal Counsel analyzes options available to the President in executing the existing immigration laws.  It is a careful and thorough analysis.  I find it a convincing defense of the President’s actions, but even if you don’t agree, this only establishes that people can have honest disagreements over how to interpret a statute.  (For the views of one very distinguished administrative law professor who disagrees with my assessment of the OLC opinion, see Peter Schuck’s op-ed in the NYT.)  That is enough to secure a negative answer to the second question posed above.

Nonetheless, the President’s critics – and especially the House Republican majority – want to wage a major battle with the President over the extent of his executive authorities, his ability to “run around Congress,” to behave as an emperor or king, or to “shred the Constitution.”  My second response is that the OLC opinion demonstrates that the President’s immigration actions are going to be a very poor field on which to wage that battle.  The question of the legality of granting deferred action and making application for work authorization available to 4.4 million undocumented is, as a legal matter, a garden variety administrative law question of the extent of the discretion that has been delegated, originally to the Department of Justice and later by transfer of authority and additional legislation to the Department of Homeland Security.   A court might find that DHS’s deferred action directive is outside the scope of its statutory authority.  I doubt it, but let’s assume that is possible.  If it does, then the action will be consigned to the pile of agency actions that have been overturned by courts over the years as exceeding their authorities under the relevant statutes.  To my knowledge, in none of these prior decisions has a court ever even contemplated the question of a constitutional violation by the President.  A court would not go there this time around, either.


The OLC opinion is devoid of any invocations of presidential authority to ignore the Congress.  To the contrary, one of the principles it identifies as limiting executive discretion is that decisions ought to be “consonant with, rather than contrary to, the congressional policy underlying the statues the agency is charged with administering.”  It then makes the case that the actions announced by the President satisfy this principle.  Again, you may disagree, but it is extremely difficult to see how large questions of constitutionality could be teed up in any litigation over these decisions, because as I have said, the President seems to be making a conventional case about the scope of authority granted to an administrative agency by Congress, pursuant to statute.
 
To be sure, the move from the status quo ante to the deferred action/work authorization regime outlined in these actions is a dramatic, profound and extremely consequential change, most especially for the 4.4 million affected undocumented workers and their families and loved ones.  Can it really be the case that the Congress has delegated the ability to make such a change to the executive without requiring the President to return to the Congress for new legislation?
 
I suspect this question – basically, “under our system of government, how can this be?” – would continue to fuel debate and anxiety over the President’s actions even if the Constitutional questions were taken entirely off the table.  (That, of course, is not going to happen.  We have become all too accustomed to believing that whenever our government goes seriously off the rails, however the President’s critics may define those rails, they must be violating the Constitution.)  Surely ours cannot be a system of government that permits the President to take such dramatic action, affecting so many people, angering so many others, when making similar changes in the immigrations laws has been the subject of active legislative debate, when the Congress has been unable to find a satisfactory bill to enact, and when the leadership after the recent election  -- but before his unilateral actions -- has said they want to work with the President to find a way forward on immigration?

This brings me to my third response:  For almost all discussions of presidential power with respect to domestic policy, whether it be immigration enforcement, clean air regulation, implementation of the Affordable Care Act, or other policies, it is essential to start with the basic legal building blocks of the modern administrative state.  These blocks are not etched in stone, to butcher a metaphor, but they are currently foundational, and until one or more of them is changed they will make a great many questions of expansive executive authority garden variety issues of administrative law and not constitutional crises.

Here is the way a leading undergraduate political science text on the Congress begins its discussion of “congressional control of the bureaucracy:”

“’Congressional power … is never lost, rarely taken by force, and almost always given away.’ Because no law can be sufficiently detailed to cover every conceivable circumstance, Congress allows executive officials wide discretion in implementing the laws it passes.  This delegation of authority occurs because legislators lack the time, knowledge, or expertise to address the complexities of contemporary administration.” [Davidson, Oleszek, Lee and Schickler, Congress and Its Members 330 (14th ed., 2014)

 Over the years, Congress has enacted many laws that only reinforce the truth that they “allow[] executive officials wide discretion,” discretion that includes interpreting ambiguous statutory content, filling in gaps, applying old statutes to situations not even conceived of at the time of enactment, and determining priorities in both the civil and criminal enforcement of the laws.

On the interpretive front, the canonical Supreme Court rendering of the executive branch’s interpretive discretion is found in Chevron v. NRDC.  In this 1984 case, the Court held that whenever the Congress had spoken in a statute to “the precise question at issue,” the agency must follow the decision of the Congress.  Otherwise, the agency was authorized to choose any “permissible construction” of the statute.  In the years since Chevron, whenever a reviewing court has found that the statute has not spoken to the precise question at issue, the court’s review of the agency’s construction of the statute has tended to be very permissive, indeed.  Rob Glicksman and I have done a review of federal decisions reviewing EPA decisions in the 1990s in which we found that in these so-called Chevron Step Two cases, EPA won 93% of them.  We have compiled but not completely analyzed raw data for the 2000s, but the Chevron Step Two data look similar.  Studies by others, not limited to review of the EPA, are to the same effect.

Under the Chevron framework, Congress has multiple means for controlling executive discretion.  It can, for example, enact a law that prohibits deferred action for the persons for whom the President wishes to establish deferred action.   A bill passed the House this summer that would have done this with respect to his initial deferred action decision in 2012, but it never received a vote in the Senate.  Alternatively, it could establish that such persons should be placed in the highest priority for deportation, which would not directly bar deferred action but would defeat the administration’s efforts to assign a low priority to such persons’ deportation, effectively doing the same thing.   Or it could bar by statute considering keeping families intact as any kind of positive consideration for any of the mechanisms that currently exist for granting leniency or temporary exemptions from deportation.   The OLC opinion places significant weight on that factor in justifying the President’s decision; eliminating that facotr would require a rethinking of its legal basis and might well prevent it.

So far as the existing immigration laws are concerned, however, neither this Congress nor any prior Congress has placed any of these limitations into the statute.  In providing very little by way of limitations on enforcement decisions, the Immigration and Naturalization Act is consistent with the vast majority of laws that the Congress passes.  In such cases, the Supreme Court has been, if anything, even more permissible than its Chevron formulation.  In Heckler v. Chaney, written by Justice Rehnquist one year after Chevron,  the Court ruled that an agency’s decision whether to enforce or decline to enforce a statute was a question over which it had so much latitude that a Court ought not even review that decision to see whether it met the “permissible construction” standard.  Heckler left open the possibility that a litigant might be able to convince a court that an agency had “consciously and expressly adopted a general policy that is so extreme as to amount to an abdication of its statutory responsibilities.”  Here – and unlike Heckler – there certainly is an express and consciously adopted general policy to challenge, but so long as enforcement resources at DHS permit going after at most 400,000 deportations a year, it will be very hard to make the case that DHS is abdicating its responsibilities under the statute.
 
Whatever one may think of the desirability of the President taking these dramatic immigration steps, this is the basic legal framework within which executive branch decisions regarding how to execute the laws are made today.  As a matter of legal permissibility, such actions need to be analyzed within this framework, under which decisions of great moment can and are made without reverting to new congressional action, precisely because prior congressional action has put in place statutes giving discretion to the executive branch.  That discretion remains in place until changed by future congressional action.  The Supreme Court is quite unlikely to save Congress from its propensity to write statutes that give the executive branch these authorities.   The modern Court firmly believes that the amount of discretion granted to an agency by statute is almost entirely for Congress to decide.  The last time it heard a challenge to a statute based on Congress giving too much power away was American Trucking Association v. Whitman, and there, in an opinion by Justice Scalia, the Court reaffirmed a strong disinclination to police statutory delegations of decision making authority.
 
As President, Barack Obama is as entitled as any other President to act within this legal framework.  Of course, any particular action must fit not only within the general framework just sketched out above; it must also be consistent with existing statutory law in the all the ways the framework requires.  In the case of the immigration laws, the particular actions he announced last Thursday night satisfy those requirements.
 
Christopher H. Schroeder is Professor of Law and Public Policy Studies at Duke University.  He can be reached at schroeder@law.duke.edu.