The constitutional limits of prosecutorial discretion

An interesting analysis by Prof. Josh Blackman, which I thought I'd pass along; this isn't my core area, so I don't have a well-developed view on the subject, but Prof. Blackman's analysis [UPDATE: written before Ilya's post on the same subject was posted] struck me as quite interesting:

The Department of Justice Office of Legal Counsel's memorandum justifying President Obama's new executive action recognizes that prosecutorial discretion is not "unlimited." As a result, the memo attempts to draw a line between lawful prosecutorial discretion, and unconstitutional executive lawmaking.

Citing the Supreme Court's 1985 precedent of Heckler v. Chaney, the memo identifies four "principles governing the permissible scope of enforcement discretion." First, as the Court noted in Chaney, enforcement decisions should reflect "factors which are peculiarly within [the enforcing agency's] expertise." This factor would almost always seem to be satisfied.

Second, the President "cannot, under the guise of exercising enforcement discretion, attempt to effectively rewrite the laws to match its policy preferences." Specifically, the memo adds, "an agency's enforcement decisions should be consonant with, rather than contrary to, the congressional policy underlying the statutes the agency is charged with administering." This argument reflects the backdrop of Congressional acquiescence, and the failure of the legislative branch to set any broad policies concerning immigration laws.

Third, the President cannot "'consciously and expressly adopt[] a general policy' that is so extreme as to amount to an abdication of its statutory responsibilities." The memo continues, "Abdication of the duties assigned to the agency by statute is ordinarily incompatible with the constitutional obligation to faithfully execute the laws." The "Take Care" clause is violated by such an abdication.

The fourth principle, which is not from Chaney, but derived from its progeny is the most important: "non-enforcement decisions are most comfortably characterized as judicially unreviewable exercises of enforcement discretion when they are made on a case-by-case basis." The memo proceeds to explain that this final factor is key: "case-by-case enforcement decisions" renders non-enforcement lawful.

The most important hurdle DOJ must vault to justify the President's new executive action policy is to explain how the Administration's blanket policy of non-enforcement for those who meet certain criteria fits within Heckler's requirement for an individualized assessment.

The memo repeatedly cites Crowley Caribbean Transp., Inc. v. Peña (D.C. Cir. 1994), which warned that "a broad policy against enforcement poses special risks that [the government] 'has consciously and expressly adopted a general policy that is so extreme as to amount to an abdication of its statutory responsibilities.'" In other words, if the government adopts a "broad" blanket policy of non-enforcement, there are "special risks" that it amounts to what Heckler called an "abdication of its statutory responsibilities," and a possible violation of the Constitution.

The OLC memo parries this admonition, by explaining "That does not mean that all "general policies" respecting non-enforcement are categorically forbidden. There is a difference between a "general policy" (permissible) and "broad policy" ("risky" business). This is the essential inquiry—are President Obama's policies with respect to DACA and IAEA "general," or too "broad" so as to pose "special risks."

OLC must explain how these "general policies" still entail "individualized assessments." To do this, the memo turns to Justice Scalia's 1993 opinion in Reno v. Flores. This case considered an INS policy whereby alien juveniles who were arrested without their parents would not be released into the custody of "responsible adults." The petitioners contended the INS policy amounted to a "blanket' presumption of the unsuitability of custodians other than parents, close relatives, and guardians." In other words, there was no individualized assessment, and the INS was declining to enforce a statute would allow the release of the juveniles.

Agreeing that there may be risks to such a policy, Justice Scalia stressed that the Attorney's General "exercise of discretion … requires 'some level of individualized determination.'" But, Scalia countered, the government need not "forswear use of reasonable presumptions and generic rules." In other words, applying certain types of presumptions or rules, while they may seem to be a "blanket policy," are in fact forms of individualized assessment. Such approaches are not too "broad" to pose constitutional problems.

Justice Scalia continued, and noted that by asking a series of questions—such as whether the alien is "under 18 years of age" or if he has an "adult relative or legal guardian"—the INS "makes those determinations that are specific to the individual and necessary to accurate application of the regulation." With these questions, Scalia concludes, "the particularization and individuation need go no further than this."

Dissenting from this position was Justice Stevens, joined by Justice Blackmun. Justice Stevens would reject the notion that "mere presumptions" satisfy the requisite individualized determination requirement: "a blanket rule that simply presumes that detention is more appropriate than release to responsible adults is not narrowly focused." Stevens would require a meaningful assessment, on a case-by-case of whether an alien juvenile could be released into the custody of a "responsible adult."

Based on Reno, the OLC memo explains that "some 'general policies' may, for example, merely provide a framework for making individualized, discretionary assessments about whether to initiate enforcement actions in particular cases." This is how OLC justifies the President's new immigration policy, as well as DACA. The memo explains what approaches are permissible: "General policies" that "provide a framework" that allows the government to make "individualized, discretionary assessments" about whether or not to enforce the deportation laws. The rule, if it can be so simply stated, is that applying "reasonable presumptions and generic rules," pursuant to a "general policy" is permissible, so long as it involves "some level of individualized determination" with "case-by-case discretion."

At every step of the analysis, the memo returns to this framework. With respect to prioritization of criminals (felons over families, as the talking point goes), OLC explains that "although the proposed policy is not a 'single-shot non-enforcement decision,' neither does it amount to an abdication of DHS's statutory responsibilities, or constitute a legislative rule overriding the commands of the substantive statute." Why is this seemingly blanket policy not an abdication? Because it "provides a general framework for exercising enforcement discretion in individual cases, rather than establishing an absolute, inflexible policy of not enforcing the immigration laws in certain categories of cases."

In its discussion of "expansion of deferred action to new classes of aliens," OLC stresses that "Immigration officials cannot abdicate their statutory responsibilities under the guise of exercising enforcement discretion." To avoid this abdication, any program should "leave room for individualized evaluation of whether a particular case warrants the expenditure of resources for enforcement."

Finally to rebut claims that the "breadth" of the program raises "particular concerns," OLC falls back on the same line of reasoning. "The guarantee of individualized, case-by-case review helps avoid potential concerns that, in establishing such eligibility criteria, the Executive is attempting to rewrite the law by defining new categories of aliens who are automatically entitled to particular immigration relief." The last, best hope of a blanket non-enforcement policy is the appearance of an "individualized assessment."

I emphasize "appearance," because it is not clear the policy President Obama announced recently employs an actual "individualized assessment." While the OLC's theory seems consistent with precedent—and OLC went out of its way to do so—it remains to be seen whether the policy operates in accordance with this theory.

To use a historical example, consider President Obama's 2012 Deferred Action for Childhood Arrivals (DACA). As OLC noted in their memo, they "orally advised" the Administration that "class-wide basis would raise distinct questions not implicated by ad hoc grants of deferred action." Specifically, the memo warned that "granting deferred action automatically to all applicants who satisfied the threshold eligibility criteria" would be problematic. Secretary Janet Napolitano's June 2012 memo, announcing DACA, provided that "USCIS should establish a clear and efficient process for exercising prosecutorial discretion, on an individual basis." This is very similar language to Secretary Johnson's memo.

Despite paying lip service to discretion, according to a Brookings report, only 1% of applicants were denied deferrals. I could not find any explanation for why, under the capacious standards set by DHS, the denial rate was even this high. A 1% denial rate seems awfully close to "automatic" relief.

Justice Scalia's opinion in Reno stressed that "exercise of discretion … requires 'some level of individualized determination.'" The relevant question here, is "individualized determination" of what. In Reno, the discretion was based on an "individualized determination" of a statute passed by Congress—whether the alien juveniles could be released to certain adults. Here, the "individualized determination" is based on policies the Obama Administration itself created out of whole cloth. (As an aside about DACA, because the Dreamers did not have an independent path to citizenship, unlike the parents of U.S. Citizens under IAEA, it is unclear how the OLC's memo justifies deferring their deportation. I'll address this in a future post). Whatever deference may be owed to the statute passed by Congress, and implemented by INS in Reno, does not exist for President Obama's unilateral executive action.

OLC lists the factors to consider for IAEA , as explained in a Draft Memo by U.S. Citizenship and Immigration Services, cited in the OLC Memo (available here):

Specifically, DHS has proposed to implement a program under which an alien could apply for, and would be eligible to receive, deferred action if he or she is not a DHS removal priority under the policy described above; has continuously resided in the United States since before January 1, 2010; has a child who is either a U.S. citizen or a lawful permanent resident; is physically present in the United States both when DHS announces its program and at the time of application for deferred action; and presents "no other factors that, in the exercise of discretion, make[] the grant of deferred action inappropriate." Draft Memorandum for Leon Rodriguez, Director, U.S. Citizenship and Immigration Services, et al., from Jeh Charles Johnson, Secretary of Homeland Security, Re: Exercising Prosecutorial Discretion with Respect to Individuals Who Came to the United States as Children and Others at 4 (Nov. 17, 2014)

These factors are equally capacious as those under DACA, and are likely to yield a similar denial rate. Secretary Jeh Johnson provides absolutely no guidance of what this "exercise of discretion" should be, and what the grounds are for rejecting an application. This must be deliberate, as OLC felt compelled to address this absence. The memo explains, "The proposed policy does not specify what would count as such a factor; it thus leaves the relevant USCIS official with substantial discretion to determine whether a grant of deferred action is warranted." Yet, OLC still expects that officials "evaluating the alien's deferred action application must still make a judgment, in the exercise of her discretion, about whether that alien presents any other factor that would make a grant of deferred action inappropriate." While this absence of guidance should create a cause for concern, OLC is satisfied. "This feature of the proposed program ensures that it does not create a categorical entitlement to deferred action that could raise concerns that DHS is either impermissibly attempting to rewrite or categorically declining to enforce the law with respect to a particular group of undocumented aliens." This is likely discretion in name only.

OLC explains that "Some 'general policies' may, for example, merely provide a framework for making individualized, discretionary assessments about whether to initiate enforcement actions in particular cases." It is unclear how this policy does anything more than pay lip service to "individualized, discretionary assessments about whether to initiate enforcement actions in particular cases." By its own terms, the President's policy seems to flunk OLC's approach. The President has in mind a group he wants to exempt from the law, and created a policy to do just that. Further, from a practical perspective, if DACA is any guide, this "exercise of discretion" is not particularly meaningful.

With respect to the future applications of up to 4 million immigrants, only 24 Field Offices—whose workers we are told are extremely overworked—are expected to manage all of those applications. It's not hard to imagine that this individualized assessment quickly turns into a rubber stamp. It cannot be the rule of law that the President can create criteria that automatically apply to millions, then instruct his agents to check off a few boxes that will always be checked, and call it an individualized assessment. The policy is designed to exempt everyone who correctly signs up. This is not an instance of executive discretion, but of clerical approval. Such a ministerial task seems inconsistent with the type of analysis required by Heckler.

Finally, I think it is essential to separate the two types of executive action the President is taking: prioritization and deferred action. They are not the same.

The first—prioritization of deporting those who are dangerous—falls closer to traditional roles of prosecutorial discretion. But, in the usual case, once a person is not prioritized for enforcement, there are no further costs to the government (I will put aside for the moment the fact that a person who is not deported may inflict other social costs). Resources are saved, and they can be spent better elsewhere.

But here we have a one-two punch. First, the President prioritizes. Second, deferred action is offered for a large portion of those who fall outside of the priority category. Instead of simply rearranging priorities for the purpose of conserving resources with regard to one class, the policy creates a framework where brand new costs are incurred due to setting the priorities. And it is this second group—a materially larger group—that creates constitutional difficulties. The weakness of the latter aspect makes the justification for the former less plausible. In total, the President's total actions are legally lesser than the sum of its parts.

Beyond immigration, the same prosecutorial discretion was used to justify non-enforcement of Obamacare. This includes the waiver for the Obamacare individual mandate (the so-called administrative fix), where anyone who "believes" that Obamacare was "unaffordable" would be exempt from the mandate penalty. It seems this relief was virtually automatic for anyone who asked, without any individualized assessment. Same for the employer mandate, where the Administration exempted all businesses with between 50 and 99 employees till 2016. Again, automatic. It's unclear that any businesses that met these criteria were denied.

It cannot be the rule of law that the President can create arbitrary criteria of where the law will not apply, and then exempt anyone who meets those criteria. To quote the court in Pena, this is the very type of "a broad policy against enforcement [that] poses special risks [because] it 'has consciously and expressly adopted a general policy that is so extreme as to amount to an abdication of its statutory responsibilities.'"

The administrative fix is subject to a suit by the State of West Virginia, which is currently pending before D.D.C. on a motion for summary judgment. The employer mandate delay is the subject of the House of Representative's lawsuit. Both plaintiffs should amply cite the OLC memo to explain why the President lacks this discretion, with an important caveat—whatever vast discretion exists in the immigration context is not nearly as strong in health care law,which Congress has certainly not acquiesced to.

As I explain in Unprecedented: The Constitutional Challenge to Obamacare, the Solicitor General adopted the strategy of not offering a clear limiting principle about the scope of the commerce clause. The fear was that by offering a limiting principle, it would bind the government in ways they would not want to be bound in the future. The SG determined that if he was going to lose, he would rather let the Court craft a limiting principle for him. With this OLC memo, we see the wisdom of the SG's decision. By putting forth a standard that the administration may not have complied with, it has opened itself up to myriad challenges, some under way, and others to come. My sequel, tentatively titled Unraveled: Obamacare, Religious Liberty, and Executive Power, continues to get more interesting.