Concept
Enforcement and Impoundment
As chief executive, including the “take care” clause
Enforcement and Impoundment
The Take Care Clause directs that the President “shall take Care that the Laws be faithfully executed.”1 Two questions test how far that duty reaches when the President would rather do less than a statute appears to require. Must the President prosecute every violator of a federal law? May the President decline to spend money that Congress has appropriated?
The answers differ because the statutes differ in what they decide. A criminal statute defines offenses and leaves the choices about whom to charge, and when, to the officers who enforce it, and those officers exercise the executive power that Article II vests in the President.2 An appropriation that commands payment makes the choice in advance, so a refusal to pay displaces a decision Congress has already made. Discretion exists where Congress left room for it and ends where Congress has spoken, so in either setting the useful question is what the statute left to the executive.
Prosecutorial Discretion
No President can personally decide which of thousands of possible federal cases to bring, so the Take Care duty in the criminal law passes to the Attorney General and the United States Attorneys.3 United States v. Armstrong explains why courts defer to them: these officers retain “broad discretion” to enforce the Nation’s criminal laws because statute designates them as the President’s delegates to help him discharge his constitutional responsibility to take care that the laws be faithfully executed.4 Prosecution is the plainest example of executive power. Dissenting in Morrison v. Olson, Justice Scalia called governmental investigation and prosecution of crimes a “quintessentially executive function.”5 He framed the case around two questions: whether conducting a prosecution is the exercise of purely executive power, and whether the statute deprives the President of exclusive control over it. The Court, which upheld the independent counsel statute, appeared to him to concede an affirmative answer to both.6
A delegate who exercises the President’s own enforcement power makes an executive judgment, and the law gives that judgment the latitude it gives other executive judgments. Armstrong restates the rule from Bordenkircher v. Hayes: so long as the prosecutor has probable cause to believe the accused committed an offense defined by statute, the decision whether to prosecute, and what charge to file or bring before a grand jury, generally rests entirely in the prosecutor’s discretion.7 Probable cause is the floor, and above it the executive chooses. That fits the clause, because a duty of faithful execution is met by honest and competent enforcement across the statute book, and no one reads it to demand a prosecution for every violation.
Courts put the discretion into practice through a presumption of regularity, a default assumption that officials performed their duties lawfully, which a challenger must overcome with evidence. In the absence of clear evidence to the contrary, courts presume that prosecutors have properly discharged their official duties.8 Two reasons support the presumption, and the first is competence. The strength of the case, its deterrence value, the Government’s enforcement priorities, and the case’s place in the overall enforcement plan are “not readily susceptible to the kind of analysis the courts are competent to undertake.”9 A prosecutor sees the whole docket, the available staff, and the competing priorities, while a judge ruling on a motion sees one file and has no standard for deciding whether that case deserved a place ahead of others.
The second reason is that the inquiry itself does damage. Examining the basis for a prosecution delays the criminal proceeding, chills enforcement by exposing the prosecutor’s motives and decisionmaking to outside inquiry, and can undermine effectiveness by revealing the Government’s enforcement policy.10 A defendant allowed to probe why he was charged turns the prosecutor’s reasons into a side dispute, prosecutors who expect that dispute shape later decisions with litigation in mind, and any policy disclosed along the way becomes known to the people it targets. The first reason shows why courts lack the tools to second-guess a charging choice, and the second shows why the attempt would impair a core executive function.
The discretion still has constitutional limits. The equal protection component of the Fifth Amendment’s Due Process Clause forbids basing the decision to prosecute on “an unjustifiable standard such as race, religion, or other arbitrary classification.”11 A selective-prosecution claim is an independent assertion that the prosecutor charged for reasons the Constitution forbids, and it leaves the merits of the charge untouched; the standard for proving it is a demanding one.12 The claim borrows ordinary equal protection standards, so the claimant must show that the prosecutorial policy had a discriminatory effect and was motivated by a discriminatory purpose. In a race case, discriminatory effect means showing that similarly situated individuals of a different race were not prosecuted.13 “Similarly situated” means in a comparable position, such as suspected of the same conduct on comparable evidence. The comparison is required because the presumption of regularity yields only to clear evidence, and a list of who was prosecuted says nothing about who could have been. At the far end of the spectrum, a defendant may show that a criminal law is administered so exclusively against a particular class, and with so unequal and oppressive a mind, that the system of prosecution amounts to a practical denial of equal protection, an idea drawn from Yick Wo v. Hopkins.14
The Armstrong respondents, defendants in a federal crack cocaine case, moved for discovery on the claim that the prosecutor singled them out because they were black.15 Their support was an affidavit from a paralegal in the Federal Public Defender’s office, who reported that in every one of the 24 drug cases the office closed in 1991 the defendant was black, together with a study listing those defendants, their race, whether they were prosecuted for dealing cocaine as well as crack, and the status of each case.16 The Court held this was not enough even to obtain discovery, because the respondents had not shown that the Government declined to prosecute similarly situated suspects of other races.17 The 24-of-24 figure describes the defendants in one office’s closed cases. It identifies no suspect of another race who committed the same offense and went unprosecuted. The Court required “a credible showing of different treatment of similarly situated persons.”18 The case materials also included a lawyer’s statement that he could not recall a federal crack case against a non-black defendant and that many crack cases involving other racial groups were prosecuted in state court.19 The dispute over the standard is about how much of that kind of evidence suffices.
The threshold for discovery matches the standard for the claim for a practical reason. If discovery is ordered, the Government must assemble from its own files documents that might corroborate or refute the claim, which diverts prosecutors’ resources and may disclose strategy. A rigorous standard for the claim therefore calls for a correspondingly rigorous one for discovery.20
The first district supplies what the Armstrong respondents lacked, a comparison group that was treated differently and evidence of purpose. The second shows ordinary discretion at work.
The comparison step decided Armstrong, and it is where an exam answer should spend its effort, since a statistic about who was prosecuted does not by itself supply a comparison group.
The separation-of-powers reasoning behind prosecutorial deference also appears in standing doctrine, which lets a federal court hear only a plaintiff who has suffered a concrete injury. In Lujan v. Defenders of Wildlife, plaintiffs relied on the Endangered Species Act’s citizen-suit provision, which authorizes “any person” to sue to enjoin anyone, including federal agencies and their officials, from violating the Act. The Court held that this broad language could not authorize suit by plaintiffs who otherwise lacked standing under the injury and redressability requirements.21 It tied that result to the Take Care Clause. To let Congress convert the undifferentiated public interest in executive officers’ compliance with the law into an individual right enforceable in court would transfer from the President to the courts “the Chief Executive’s most important constitutional duty.”22 The later cases describe the resulting danger as turning federal courts into “virtually continuing monitors of the wisdom and soundness of Executive action.”23
The link to prosecution is direct. If any citizen could sue to compel compliance, a judge and no longer the executive would decide how vigorously, and against whom, a statute is enforced, which is the choice Armstrong leaves with prosecutors. A secondary account of Lujan describes the lead opinion as locating that prosecutorial discretion in the Take Care Clause itself, as a constitutional prerogative of the executive branch to determine how vigorously to enforce federal laws.24 Allen v. Wright reasoned the same way about a suit that sought to restructure the apparatus the Executive Branch had established to fulfill its legal duties: the Constitution assigns the Take Care duty to the Executive Branch and not to the Judicial Branch, which counsels against recognizing standing.25 The Court has since read Lujan to reach suits against private defendants as well, on the theory that suits unconstrained by the injury requirement would give plaintiffs, and through them judges, too much power to interfere with the President’s choices about how to prioritize and how aggressively to pursue legal actions against violators.26
The principle leaves the courthouse open to plaintiffs who can show a concrete injury. The Lujan opinion described the courts’ role as adjudicating claims that unauthorized administrative power infringes individual rights, and it said “individual rights” do not include a public right that a statute declares to belong to everyone.27 A neighbor whose property is damaged by an unauthorized agency action can sue, while a citizen whose only complaint is that an agency is not following the law cannot. Two opinions press the other side. A Lujan dissent argued that Congress does not violate separation of powers by structuring the procedures the Executive must follow, and that federal courts surely do not violate it when they enforce those procedures at Congress’s instruction and command.28 In Steel Co. v. Citizens for a Better Environment, Justice Stevens, concurring, found it hard to see how a citizen-suit provision impinges on the Executive’s Take Care power, especially where the plaintiff has more than the undifferentiated public interest in seeing the statute enforced.29 Both are minority positions, so an exam answer should give the majority’s Take Care reasoning first and the objections second.
Spending and Impoundment
The Take Care Clause directs that the President “shall take Care that the Laws be faithfully executed,” and an appropriation that directs money to be spent is one of the laws that command covers.31 Enforcement leaves the executive room to choose, because a criminal statute defines offenses and does not order the prosecution of every violator. Mandated spending removes that room. When Congress has said that a sum shall be paid for a stated purpose, the choice is already made, and a refusal to pay overrides it. Withholding appropriated money from the purpose Congress named is usually called impoundment, and the traditional authority against it is Kendall v. United States ex rel. Stokes.
The claimants in Kendall said the Postmaster General had refused to pay them for services under a contract to carry the mail. Congress had passed a law authorizing the Solicitor of the Treasury Department to arbitrate the claim and directing the Postmaster General to pay any award, and when he would not honor the award in the claimants’ favor, they sued.32 He answered that only the President could direct him to pay, so a judicial order enforcing the award would intrude on the executive department. The Court rejected the idea that every executive officer stands under the President’s exclusive direction. Congress may impose on an executive officer any duty it thinks proper, so long as the statute otherwise complies with the Constitution, and the President cannot direct the officer to violate that duty. The officer’s “duty and responsibility [of that officer] grow out of and are subject to the control of the law, and not to the direction of the President.”33
Kendall involved a payment due under an arbitration award and not a program appropriation, but its rule reaches spending by the same reasoning. A statute that directs payment of money places a duty on the officer who holds the funds. A presidential instruction to withhold the payment asks the officer to violate that duty, and the instruction rests on nothing but the President’s preference about the program. The Take Care Clause supplies no better source, because it obligates the President to use executive power consistently with all valid acts of Congress, whatever the President thinks of the program.34 Agencies can exercise only the authority that statutes delegate to them, and those statutes bind the agencies and the President alike.35 An agency told to keep funds that a statute commands it to pay would therefore act with no statutory authority, on the President’s word alone.
Courts can enforce such a duty without intruding on the President. Federal courts have authority to review executive action for compliance with the law, as Marbury v. Madison established.36 The removal cases follow the same logic: when Congress restricts the President’s power to remove an officer, it can enforce the restriction through judicial review of the removal decision, and because that review only ensures compliance with the statute, it does not “put any additional burden on the President’s exercise of executive authority.”37
Two structural reasons support the rule. The first concerns when the Constitution gives the President a say over spending: the President’s check comes when the bill is presented for signature, before it becomes law. Impoundment would give the President a second veto after the law is in force, at a point where the legislature that wrote the appropriation cannot respond, and it would amend or repeal a statute without any of the steps that enacting it required. The second reason is the one developed in the discussion of the clause above, that the duty to execute reaches the laws and no further. That limit runs in both directions. It gives the President no claim to do more than a statute allows, and it gives none to do less than a statute commands, since the command is part of what the law says. In Jackson’s terms from the Youngstown discussion, a refusal to spend mandated funds is incompatible with the expressed will of Congress, so the President’s power is at its lowest ebb and consists of the President’s own constitutional powers minus Congress’s constitutional powers over the subject. Neither the vesting sentence nor the Take Care Clause says anything about holding back money that Congress has directed be spent, so once Congress’s authority over appropriations is subtracted, nothing in Article II remains to carry the claim.38
An exam stem that reports the President “declined to spend” appropriated money invites the quick answer that the refusal is unlawful impoundment, and the quick answer is sometimes wrong. The deciding question is what the statute commands.
The first scene is Kendall in a new setting. The second depends on how much choice the statute’s words leave, and a directive that pushes past those words falls back into the first scene. Where a statute does leave the amount or timing to an agency, the agency’s authority comes from the statute, and presidential supervision operates within it, though the extent of that supervision is disputed.39
The line-item veto decision applies the same reasoning to a statute that tried to give the President a power over spending. In Clinton v. City of New York, the Line Item Veto Act authorized the President to prevent particular items of budget authority from having legal force or effect, a power the Act labeled cancellation.40 The Court held the Act unconstitutional. As the dissent describes the majority’s approach, it was a literal one, treating cancellation as a repeal or amendment of a law that skipped the Constitution’s exclusive procedures for enacting or repealing legislation.41 On that view the President can approve or reject a bill in full but cannot cancel parts of it, because only Congress, through those procedures, can amend or repeal a statute. The decision bears on impoundment through an argument from the greater to the lesser. If a statute passed by both Houses and signed by the President could not confer a power to cancel spending items after enactment, a President who withholds mandated funds with no statute at all stands on weaker ground, because the effect on the appropriation is the same and the claim of authority is smaller.
The dissent shows where the argument for presidential control over spending is strongest and why it did not prevail. It treated the Act as the earlier law that governed the later 1997 New York health care provision, so the President, in exercising the power the Act explicitly delegates, “followed that law to the letter” and executed the law without repealing it.42 The Act itself was enacted through the constitutional procedures, the dissent said, and its failure to require the President to satisfy those procedures did not make it unconstitutional.43 The majority rejected that reading, so the dissent supplies a counterargument and states no rule. It still marks a point an exam answer should keep. Where Congress writes the room for choice into the statute, as the discretionary grant statute in the example does, the agency’s authority to decline or redirect funds exists because the statute supplies it.
Step two does the work, since most stems that mention a refusal to spend are decided by whether the text commands payment or leaves a choice. Step three places the claim in Jackson’s third category, and step four keeps the directive in the supervisory role the previous subsection described.