Can Congress Do Anything about Trump’s Abuse of the Pardon Power?

Andrew Kent
Friday, July 24, 2020, 11:36 AM

The case against the constitutionality of self-pardons is strong. Beyond barring the president from pardoning himself, what else could Congress do?

President Trump delivers remarks on July 3, 2020. (Official White House Photo by Sheelah Craighead)

Published by The Lawfare Institute
in Cooperation With

President Trump continues to misuse the constitutional power to pardon. Most recently, he commuted the sentence of Roger Stone, seemingly as a reward for Stone’s refusal to tell prosecutors, Congress or the public what he knows about Trump’s involvement with Wikileaks and materials hacked by Russia during the 2016 campaign. And there is no reason to think that the country has seen the last of this behavior. Paul Manafort’s representatives are likely redoubling their efforts to get Trump’s former campaign chairman the same kind of deal as Stone. Trump might fear legal jeopardy, should he lose the election, related to his taxes and business dealings as well as obstruction of justice or other potential crimes committed during the presidency. Trump has made clear that he believes he can pardon himself and has thought about the possibility of doing so: in 2018, he tweeted, “I have the absolute right to PARDON myself . . . .”

Congress, in turn, is considering whether there is anything it could do to prevent or nullify Trump’s apparently corrupt pardons. House Speaker Nancy Pelosi called the recent Stone commutation an “act of staggering corruption," and promised that "Congress will take action to prevent this type of brazen wrongdoing.”

This week, House Democrats on the Judiciary Committee introduced two measures. One bill, from Rep. Adam Schiff, would require Justice Department and White House documents to be turned over to Congress for certain pardons of offenses against the United States arising “from an investigation in which the President, or a relative of the President, is a target, subject, or witness.” It also clarifies or supplements existing law by criminalizing bribery in relation to presidential pardons and commutations, expressly including the president as subject to the prohibition. Jack Goldsmith and Bob Bauer analyzed the bill on Lawfare. The second, an amendment introduced by Rep. Jamie Raskin, would ban presidential self-pardons: “The President’s grant of a pardon to himself or herself is void and of no effect, and shall not deprive the courts of jurisdiction, or operate to confer on the President any legal immunity from investigation or prosecution."

These actions follow a spring 2019 hearing of a House Judiciary subcommittee examining the constitutionality of self-pardons and other corrupt pardons, and whether legislation could be drafted to prevent them. I testified at the hearing and have communicated a bit with the committee’s majority staff about these issues.

This post will analyze, first, whether it is constitutional for Congress—by legislation—to ban self-pardons. Next, I ask whether Congress could ban pardons of the president’s family members or confederates, if the president’s motive is to cover up wrongdoing in which the president is involved. And finally, instead of—or in addition to—flat bans on certain kinds of pardons (or commutations), is there anything—besides the Schiff bill—that Congress can do to make it more legally, practically or reputationally costly for a president to corruptly use the pardon power?

Can Congress Ban Self-Pardons?

A statutory ban would likely be constitutional, for a simple reason: the president likely lacks constitutional power to pardon himself. An unauthorized—and hence void—action by the president can be banned by Congress. Even the Justice Department’s Office of Legal Counsel (OLC), which historically takes very broad views of presidential power and immunity from regulation, seems to agree that Congress may ban, including with criminal penalties, actions by a president that are not constitutionally authorized. For instance, OLC opines that there are no constitutional separation-of-powers problems with applying the federal bribery statute to the president because the president has no constitutional authority to commit acts covered by the statute. The same logic should mean that a ban on self-pardons is fine as long as self-pardons are unconstitutional.

My congressional testimony discussed the constitutionality of self-pardons, and Lawfare has previously covered the issue. Earlier this year, Goldsmith and Bauer wrote that the constitutionality of self-pardons is “legally unsettled,” and recommended that Congress ban them by statute because “Congress’s constitutional judgment can matter a lot, both in informing subsequent judicial review of a self-pardon and in giving a president pause about issuing one in the first place.”

My hearing testimony recommended a congressional ban, so I fully agree with Goldsmith and Bauer that such a statute would be salutary. I think the case against the constitutionality of self-pardons is quite strong, stronger than they seem to think.

No president has ever purported to pardon himself—though at least two have contemplated it, including the current president. In August 1974, at the nadir of the Nixon presidency, OLC issued an opinion stating, with only the briefest explanation, that the president lacked power under the Constitution to pardon himself. The full reasoning was that “[u]nder the fundamental rule that no one may be a judge in his own case, it would seem that the question [of whether the president may pardon himself] should be answered in the negative.” This OLC opinion came during a time of speculation in the press on the topic of a Nixon self-pardon. (In the Washington Post, for example, Timothy H. Ingram pondered, “Could Nixon Pardon Nixon?”). The historical record now shows that there were White House discussions on the topic. And a secret White House legal opinion concluded that a presidential self-pardon was constitutional. In addition, in a brief filed nine months earlier in the criminal investigation concerning Vice President Spiro Agnew, Solicitor General Robert Bork had implied that a presidential self-pardon would be lawful.

Thus, during the Watergate crisis the executive branch was divided within itself on the constitutionality of the self-pardon. No court has ruled on the issue. Despite the novelty and difficulty of the question, I believe that the best understanding of the Constitution is that the president lacks authority to self-pardon and hence that a self-pardon would be ultra vires and void.

Article II, section 2 of the Constitution provides that “[t]he President . . . shall have Power to grant Reprieves and Pardons for Offenses against the United States, except in Cases of Impeachment.” A reprieve is a temporary suspension or stay of a criminal sentence. A pardon is, according to the Supreme Court, “[a]n executive action that mitigates or sets aside punishment for a crime.”

Pardons have deep historical roots, but the framers of the U.S. Constitution were mostly focused on English practice during the 17th and 18th centuries. Pardons were issued by the king and colonial governors, and were understood to be acts of grace and mercy, necessary to soften the severity of a criminal justice system in which most serious crimes were capital and which gave few means of defense to the accused. English monarchs sometimes abused their prerogative to pardon, and therefore the American framers were well aware of both the benefits and risks of a broad pardon power. They chose a broad but not unlimited one. Of course, impeachment and removal is available (in theory, at least) as a check on a president’s misuse of the pardon power; here I will address whether the Constitution places other limits on self-pardons.

Because the Constitution does not, on its face, expressly rule in or rule out a self-pardon and courts have not settled the issue, answering this question requires turning to the traditional methods of constitutional interpretation, including the meaning the text would have had to the adopting generation; the purposes motivating the adoption of the Constitution and the provisions at issue; the fit of the particular clause within the larger Constitution, its structure and its principles; and, where available, both practical and formal interpretations of the Constitution by Congress and the executive branch.

Article II of the U.S. Constitution twice imposes a duty of “faithful execution” on the President, who must “take Care that the Laws be faithfully executed” and must take an oath or affirmation before assuming his or her duties to “faithfully execute the Office of President.” Thus, the Constitution, statutes and the duties of the office as a whole must be faithfully carried out by the President.

With my Fordham colleagues Ethan Leib and Jed Shugerman as co-authors, I published a lengthy research paper in the Harvard Law Review on the origins and historical meaning of the Constitution’s faithful execution clauses. We found that the language of “faithful execution” was, for centuries before 1787, very commonly associated with the performance of executive offices. Governors of American colonies pre-independence, post-independence state governors, executive officers under the Articles of Confederation government and other executives—such as mayors and governors of corporations—were required, before entering office, to take an oath for the due or faithful execution of their office. The “faithful execution” duties imposed on public officials directed them to follow the standing law; avoid taking unauthorized profits or engaging in other financial corruption; carry out the law and their duties honestly, impartially and diligently, focusing on the public good rather than private benefit; stay within their limited authority as they executed their offices; and to account for official funds and their official acts at the direction of the legislature. Leib and I have deepened our findings in a new paper. People experienced in law or government in 1787—like the drafters and ratifiers of the U.S. Constitution—would have been aware of this general framework of faithful execution duties.

We argue that this understanding was embodied in the faithful execution clauses of Article II. These duties of faithful execution look a lot like fiduciary duties in modern private law. This “fiduciary” reading of the original meaning of the faithful execution clauses supports reading Article II to limit presidents to exercise their power honestly, impartially, diligently, in good faith, for the public interest and not for reasons of self-dealing, self-protection or other bad faith or personal reasons.

A self-pardon would seem to be utterly inconsistent with this historical meaning, as my co-authors and I have argued. Richard Epstein rejected this view in a Wall Street Journal op-ed, writing that self-pardons are constitutional because the Article II pardon power is not a “law” that must be faithfully executed by the president under Article II’s Take Care Clause. Epstein ignores, however, that the full office of the presidency must be faithfully executed, as must the Constitution. The pardon power is clearly covered by the faithful execution requirement. (At the same time, any critics of our historical work could of course contest our take on the substantive meaning of the faithful execution clauses).

But before turning to other arguments for and against self-pardons, it is worth noting that constraining a president to prevent self-dealing and related abuses of office is consistent with other features of Article II and the Constitution as a whole. The worst features of monarchy were rejected by the founders in drafting the Constitution. Unlike the king or queen, the U.S. chief magistrate did not have total and perpetual immunity from legal accountability. By banning titles of nobility, and providing that the president would be elected to a term of years, not chosen on hereditary principles and not ruling for life, the Constitution addresses the fear that a chief executive’s primary interest would be perpetuation of dynastic successors and retainers rather than the good of the country. Many English monarchs had been foreign-born and still held lands and titles abroad, giving them personal interests that might differ from those of the public. In response, the Constitution requires that the president be a citizen. The president is given a salary, which according to the Constitution may not be raised or lowered by Congress during the president’s time in office. The Constitution also prohibits the president from imposing taxes or otherwise raising funds on his own authority, and from accepting bribes, gifts or other emoluments of office from foreign governments or state governments. By so doing, the Americans framers intended to check typically monarchical kinds of financial self-dealing. Other scholars such as Daniel Hemel, Eric Posner and Brian Kalt note that the Constitution contains additional principles barring self-dealing and related kinds of corruption.

As noted at the outset, a few days before Nixon resigned, OLC concluded that a self-pardon was likely unconstitutional because of “the fundamental rule that no one may be a judge in his own case.” This rationale has been seconded by some commentators, such as Akhil Amar and Peter Shane.

Some proponents of the constitutionality of the self-pardon (such as Jonathan Turley) have responded that pardoning is not an act of judging and, therefore, OLC’s reasoning misses the mark. It does seem more correct to view the pardon as an executive rather than a judicial act. The Supreme Court, per Chief Justice William Rehnquist, described a pardon as “[a]n executive action that mitigates or sets aside punishment for a crime” and per Chief Justice John Marshall, wrote that the pardon power “proceed[s] from the power entrusted with the execution of the laws.” But even if the pardon power is fully executive in nature, that does not undermine OLC’s conclusion. The maxim that no person may be a judge in his or her own case states a centuries-old principle foundational to the rule of law, which has long been applicable to people other than judges—for instance, to prosecutors, judges and jurors—as Philip Bobbitt and others have noted. Whether the decision to grant a pardon is best viewed as a prosecutorial-executive decision or a quasi-judicial one, it violates a deep-seated principle of the rule of law, which has something like constitutional status in the Anglo-American legal tradition.

Relatedly, ensuring that the president is not above the law is also a core rule-of-law value that would be violated by a self-pardon. As the Supreme Court has stated, “No man in this country is so high that he is above the law. All the officers of the government, from the highest to the lowest, are creatures of the law and are bound to obey it.” In the famous Nixon tapes case, the Supreme Court reiterated this, rejecting Nixon’s contention that the president “is above the law.” The Supreme Court expressly weighs rule-of-law values and seeks to preserve means of presidential accountability when deciding novel separation of powers questions about presidential power. This is seen in the Court’s recent ruling about subpoenas to Trump’s accounting firm and bank from various House committees, as well as other decisions.

The framers of the Constitution had divergent opinions about whether a sitting president could be criminally prosecuted. But the Constitution itself makes clear, in a provision dealing with impeachment, that prosecution may follow a president’s removal from office by impeachment and conviction. In addition, a former president may be criminally charged if absence from office is due to resignation, electoral defeat or a 22nd Amendment term limit. Especially given current Department of Justice policy that a sitting president may not be prosecuted, ensuring accountability for criminal misdeeds and preserving the rule of law would seem to require that a president may not grant himself permanent impunity with a self-pardon. A president with unfettered discretion to make himself permanently unaccountable at law savors too much of the legally untouchable English monarchy that the American founders rejected. It is also hard to see how using a pardon to accomplish self-impunity is consistent with duties of faithful execution. Bobbitt appears to agree with this conclusion.

Bobbitt also points out that the Constitution speaks of the president “grant[ing]” a pardon, thus employing a legal term that meant conveying a chattel or status to a third party. It makes no sense, and is contrary to traditional legal usage, Bobbitt contends, to think that the president could be both the grantor and grantee of a pardon. Brian Kalt has made a similar argument. While perhaps not dispositive by itself, this argument supports the conclusions I have reached on other grounds.

Also supportive of my conclusion—though again, not dispositive in my view—is an argument by Laurence Tribe, Richard Painter and Norm Eisen: “The Constitution specifically bars the president from using the pardon power to prevent his own impeachment and removal. It adds that any official removed through impeachment remains fully subject to criminal prosecution. That provision would make no sense if the president could pardon himself.” I don’t think a constitutional provision applying to many federal officers “makes no sense” because in one very specific and unusual circumstance it produces an anomalous result for the president. But I support the point that, as to the president, there are many historical, textual, structural and principled (rule of law) reasons pointing toward the unconstitutionality of a self-pardon.

The most common argument in favor of self-pardons is that the pardon power is phrased in very broad language, with only two express limitations— (1) that pardons reach federal offense, not state offenses and not private civil suits, and (2) that pardons cannot negate impeachments—and therefore the pardon must be absolute and subject to no other limitations. This argument has been made by former Attorney General Alberto Gonzales; judge and law professor Richard Posner; law professors Richard Epstein, Saikrishna Prakash, Michael Stokes Paulsen and John Yoo; and lawyer and commentator Andrew McCarthy. I am sure there are more.

Though often repeated, the argument from the breadth of the text is not particularly persuasive. For one thing, the broad language of the pardon clause is understood to have an important, unwritten limitation: it can only be used to pardon crimes already committed, not future crimes (see this point made by the Supreme Court and Edward Corwin, the great scholar of the presidency). If a president could pardon future crimes, this would be equivalent to having the dangerous power to preemptively dispense with the laws—a power that England’s Parliament wrestled away from the crown, and that no one can plausibly argue would have been given back to the new American president by the generation that revolted against George III. If the pardon power concededly has this one atextual limitation—a rule-of-law, faithful execution-type limitation—why not another?

More generally, it is the norm in U.S. constitutional law that seemingly broad text in the Constitution is constrained both by other express parts of the document (for instance, the Bill of Rights) and by implicit constitutional principles. Examples of the latter are that congressional power is limited by state sovereign immunity that reaches beyond the text of the 11th Amendment and by implicit anti-commandeering rules. Presidential power granted by the Constitution is similarly subject to all kinds of express constitutional limitations (for instance, the Bill of Rights; the Habeas Suspension Clause; congressional power grants that cover similar areas as presidential grants, such as war-making; Congress’s Necessary and Proper power). And there are implied limitations on presidential power too (for instance, Congress’s exclusive power over funding and near exclusive power over and office creation necessarily limits the president’s express Article II powers, for instance with regard to sending ambassadors and other diplomats, or being commander-in-chief).

Indeed, the Supreme Court has found that there are limitations on the pardon power besides the ones in the Constitution’s plain text. In addition to the rule against preemptive pardons, in another decision the Court held that the pardon power “cannot touch moneys in the Treasury of the United States . . .” or property vested in private third-parties when it blots out the punishment from a federal crime. Although the pardon power is exceptionally broad, it is still “part of the Constitutional scheme” and thus should be subject to constitutional limitations.

The argument that self-pardons are permissible because of the breadth of the constitutional text is buttressed by dicta in some Supreme Court decisions, which have stated—addressing very different contexts than a purported self-pardon—that the pardon power is “plenary,” “granted without limit,” and “unlimited,” but for the express restriction about impeachments. Yet broad, general statements in Supreme Court opinions that were unnecessary to the decision are not automatically applicable to later cases, especially those addressing different questions.

Michael McConnell argued in the Washington Post that the Constitution's drafters at the Philadelphia Convention in 1787 specifically contemplated and approved presidential self-pardons. This is very unlikely to be correct, as my coauthors and I have written. But even if notes from the Philadelphia convention reflect that one or two framers said things that might suggest contemplation of a self-pardon, McConnell’s argument should be discounted. The notes from the Philadelphia Convention are well known to be incomplete and imprecise. The proceedings were held in secret, and for several decades little information about what had transpired was public. The secret intentions of the drafters are not what made the Constitution the supreme law of the land.

As Chief Justice Marshall wrote in one of the most important Supreme Court cases of all time, when the proposed Constitution drafted at Philadelphia was released to the public, it "was a mere proposal, without obligation or pretensions to it." The people of the states met in conventions to debate whether to adopt the Constitution. "From these conventions," Marshall wrote, "the Constitution derives its whole authority." That is why the most plausible and widely-accepted version of originalist constitutional interpretation looks not to the intentions of the drafters at Philadelphia, but at the objective meaning that the Constitution's words would have conveyed to the American public at the time of ratification.

Solicitor General Bork did not so much argue for the constitutionality of self-pardons as assume it without explanation. His 1973 brief addressed why the vice president, unlike the president, is not immune from criminal prosecution while in office. As part of this discussion, Bork wrote that the Constitution’s drafters “gave [the president] ‘Power to grant Reprieves and Pardons for Offenses against the United States, except in Case of Impeachment’ . . . a power that is consistent only with the conclusion that the President must be removed by impeachment, and so deprived of the power to pardon, before criminal process can be instituted against him.” It is impossible to address Bork’s reasoning because he did not explain it.

The arguments favoring self-pardons run the gamut from weak to extraordinarily weak. Weighed against strong and semi-strong arguments on the other side, the best reading of the Constitution prohibits self-pardons. As a result, Congress has the power to prohibit an action—self-pardons—that the president lacks any constitutional authority to engage in.

Could Congress Ban Other Self-Interested Pardons?

If self-pardons could be banned, could Congress also ban pardons of the president’s family members or confederates who may be linked with the president in criminal activity?

Unlike a presidential self-pardon, this may actually have occurred prior to Trump. There are still unresolved questions about whether George H.W. Bush or Bill Clinton may have used the pardon power this way. As discussed above, the Roger Stone commutation seems likely to have been a corrupt pay-off to a co-wrongdoer.

Debates at the Philadelphia Convention and in the Federalist Papers make it clear that the Founding generation contemplated the possibility that a president would use the pardon power to shield treasonous or corrupt associates from criminal responsibility. Alexander Hamilton suggested that the remedy for such an abuse would be impeachment. In discussing the same issue, James Wilson argued that the Philadelphia Convention should not bar pardons for treason because sufficient safety was ensured by the fact that a president guilty of pardoning co-conspirators could be impeached and criminally prosecuted.

These hints of original understanding, coupled with capacious constitutional text and the Supreme Court’s broad pronouncements about the “plenary” pardon power, suggest that it may be constitutional—though of course dishonorable, corrupt and subject to review through the process of impeachment—for a president to pardon family members or other associates who have engaged with him in misconduct. But I do not believe the pardon power should be viewed as entirely unlimited in these circumstances. A pardon of a third-party motivated principally by the president’s desire to protect himself would seem to violate the faithful execution principles sketched above. Likewise, I think it is plausible to argue, though admittedly a harder case, that a pardon of a close family member or friend, linked with the president in personal or official misconduct, motivated by a desire to shield them from legal accountability for personal reasons unrelated to the public good, could also violate the faithful execution principles.

These questions are tricky, however. For one thing, the president’s subjective motivations would seem to be a paramount consideration, and it would be problematic for Congress or the courts to gather the information needed to make judgments about motivations in individual cases. (Of course, this might be easier in the case of a president like Trump, who often says the quiet part out loud.) On an unsettled issue like this, it is worth bearing in mind Goldsmith and Bauer’s reminder that Congress’s constitutional judgment can have persuasive power.

What Else Can Congress Do?

Beyond outlawing certain kinds of pardons, Congress has constitutional power to take other actions that would make it more costly—legally, practically or reputationally—for a president to corruptly use the pardon power. Here are some possibilities:

  • Congress could bar the use of any appropriated funds on the preparation or issuance of a self-pardon, or for pardons, commutations, or reprieves of family members of the president and/or persons linked with the president in wrongdoing. Supreme Court cases suggest (albeit in very different circumstances) that “Congress [cannot] [] limit the effect of [the president’s] pardon.” But this funding limitation does not purport to affect the validity of the pardons; it simply reflects a judgment by Congress that the U.S. taxpayers should not pay for the president’s staff to spend time on narrow classes of pardons that Congress deems corrupt. Congress could choose to couple a funding ban with a criminal penalty for violations, like that found in the Anti-Deficiency Act. I suppose that someone might suggest that Supreme Court case law dealing with “unconstitutional conditions” restricts Congress here. I don’t see it that way. That body of law, widely criticized as muddled and incoherent, primarily seeks to protect individual constitutional rights and federalism-related constitutional prerogatives of the states from forced waiver via conditions attached by Congress in order to access federal spending or other federal benefits like licenses. The leading case finding a funding condition relating to federal executive branch operations unconstitutional is actually best viewed as turning on congressional abuse of individual rights. (Congress enacted a prohibited bill of attainder in the guise of a limitation on federal spending with respect to individually-named executive officials.)
  • Congress could provide that, after leaving or being removed from office, a former president who had purported to pardon himself or herself would be ineligible to receive any of the benefits provided by sections (a), (b) or (c) of the Former President’s Act, codified at 3 U.S.C. § 102 Note. (I would recommend leaving in place Secret Service protection and a pension benefit for the surviving spouse upon a former president’s death.) Even if a court were to take the view that self-pardons are or may be constitutional, I do not think the proposal tries to undo the effect of the pardon. It instead exempts from gratuitous public benefits any president who has engaged in action that Congress deems corrupt and unconstitutional. Perhaps one could argue that a congressional prohibition along these lines would be an unconstitutional condition, because the president in his personal capacity has a right to receive a pardon granted by himself in his official capacity—but this just reveals the gross self-dealing and twisting of constitutional language that a self-pardon would involve.
  • Congress could provide by statute that, because a self-pardon is an admission of guilt and/or a corrupt and unconstitutional action, a president who purports to pardon himself or herself shall be forever barred from voting in an election for an office under the United States and, after the end of the presidential term during which the purported pardon occurred, from holding any office under the United States. This would take away very important rights granted by the Constitution and statute—that is, the rights of the president as a citizen to vote and run for office. It is thus more likely than my other proposals to be found unconstitutional.
  • Congress could broaden the Schiff bill to apply to all self-pardons. Congress might also directly state its constitutional judgment that no executive privilege should attach to executive branch documents covered by the Schiff bill on the ground that they may reflect corrupt, criminal and/or unconstitutional actions by the president and those working for him or her.

I’m sure there are many other possible legislative responses as well. A number of groups, among them Protect Democracy, have done important work suggesting legislative reforms aimed at Trump abuses, including abuses of the pardon power.

Controversy surrounding the pardon power is nothing new. Gerald Ford’s blanket pardon of Nixon, suspected by some to be part of a corrupt bargain through which Nixon resigned so Ford could secure the presidency, is one of many examples. Clinton and Bush, the elder grappled with their own controversies, as discussed above. For some of these past incidents, Congress bestirred itself to hold hearings and ask questions. Perhaps the relentless and unvarnished nature of Trump’s abuses of the pardon power will finally lead to some statutory regulation being adopted. Though the task is difficult, it will be far from impossible for Congress to design meaningful restraints while staying within the bounds of the Constitution.

Andrew Kent is a professor at Fordham University School of Law and holds the John D. Feerick Research Chair.

Subscribe to Lawfare