Ruling in Trump v. Mazars on July 9, the Supreme Court held that courts must take into account separation of powers concerns in resolving disputes over congressional subpoenas seeking personal information of the president. The court found that the split panels at the U.S. Courts of Appeals for the D.C. Circuit and the Second Circuit had failed to adequately account for “weighty” separation of powers considerations when rejecting challenges to House committee subpoenas seeking financial records relating to President Trump, his affiliated business entities and his family members. In a 7-2 decision authored by Chief Justice John Roberts, the Supreme Court vacated the judgments below and remanded to the district courts for further proceedings. (The court also ruled in a related but distinct case concerning Trump’s efforts to block subpoenas for his financial records from New York state law enforcement; a summary of that decision, Trump v. Vance, is available here.)
The court defined the question presented as a matter of first impression, stating it had “never addressed a congressional subpoena for the President’s information.” Such subpoenas, the court found, “unavoidably pit the political branches against one another.” This is so even when, as in these cases, subpoenas implicate the president in his personal, not official, capacity, or are issued to third parties. While the court affirmed that it is the “‘duty of all citizens to cooperate’” with congressional information requests, the majority found that courts presiding over subpoenas seeking information from the president must account for the “special” separation of powers issues these disputes raise.
Roberts was joined in the majority by Justices Ruth Bader Ginsburg, Stephen Breyer, Sonia Sotomayor, Elena Kagan, Neil Gorsuch and Brett Kavanaugh. Justices Clarence Thomas and Samuel Alito both penned dissents.
In April 2019, the U.S. House of Representatives issued four subpoenas for Trump’s financial records. The House Committee on Financial Services sent a subpoena to Deutsche Bank concerning “foreign transactions, business statements, debt schedules, statements of net worth, tax returns, and suspicious activity identified by Deutsche Bank” and also subpoenaed Capital One for similar matters. The House Permanent Select Committee on Intelligence also subpoenaed Deutsche Bank. Finally, the House Committee on Oversight and Reform subpoenaed Trump’s personal accounting firm, Mazars USA, LLP, for matters related to the president and his businesses. All three committees provided individual justifications for their subpoenas, arguing that the records were needed for the House to weigh potential legislation in areas such as terrorism, money laundering and foreign interference in elections.
Trump, his children and the Trump businesses sought to block the subpoenas, contesting the Oversight Committee subpoena in the U.S. District Court for the District of Columbia (in what became Trump v. Mazars) and the subpoenas from the Financial Services and Intelligence committees in the Southern District of New York (in what became Trump v. Deutsche Bank). The president argued that Congress lacked a “legitimate legislative purpose” and that the action was a violation of the separation of powers. In Mazars, the district court ruled for the House on the basis that the tax returns “served a valid legislative purpose” because the returns related to legislation addressing financial disclosure requirements for presidential candidates and presidents. The D.C. Circuit affirmed this ruling. In Deutsche Bank, the district court and Second Circuit also upheld the subpoenas, finding that they were sufficiently related to legislation being reviewed by the committees concerning national security, terrorism, money laundering and “the global movement of illicit funds through the real estate market.” The Supreme Court consolidated Mazars and Deutsche Bank.
The Majority Opinion
Roberts distinguishes the dispute before the Supreme Court as a “significant departure from historical practice.” From George Washington’s administration to the present, he writes, the political branches have resolved information disputes out of court, through the “‘hurly-burly, the give-and-take of the political process.’” These more than two centuries of established practice “impose on [the court] a duty of care” not to upset the conflict resolution mechanisms the two branches have developed and employed over time.
The chief justice situates his analysis within a body of case law regarding Congress’s subpoena powers. To begin, each chamber of Congress has implied constitutional authority to conduct investigations and issue subpoenas. Congress’s power to obtain information is “‘broad’” and “‘indispensable,’” Roberts writes but not without limit. The guiding test: congressional subpoenas must address a “‘valid legislative purpose.’” They must be “‘related to, and in furtherance of, a legitimate task of the Congress.’”
Roberts outlines further limits established by precedent. First, Congress may not issue subpoenas for a “law enforcement” purpose—which would involve exercising powers belonging to the executive and judicial branches. Investigations may not seek “‘exposure for the sake of exposure’” or to simply penalize those investigated. Finally, Congress’s oversight and investigative powers are limited by the Constitution, as well as common law and constitutional privileges. These include attorney-client and executive privileges.
With this in mind, the chief justice rejects both the president’s and the House’s proposals for how to evaluate congressional subpoena enforcement disputes implicating the executive branch. Either proposal, he finds, would “transfor[m]” how the two branches have historically resolved such disputes.
First, Roberts rebuffs the president’s and the solicitor general’s call for the House to show a “‘demonstrated, specific need’” for information and explain that the information sought is “demonstrably critical” to a legislative purpose. The Supreme Court and the D.C. Circuit had outlined these “demanding standards” in United States v. Nixon and Senate Select Committee on Presidential Campaign Activities v. Nixon—cases concerning the Watergate tapes, over which President Nixon had asserted executive privilege. Though information protected by executive privilege is given the “‘greatest protection,’” Roberts declines to categorically “transplant that protection root and branch” to subpoena enforcement suits that do not implicate privileged information. To do so, he writes, “would risk seriously impeding Congress in carrying out its responsibilities.”
But Roberts also rejects the House’s position that the validity of the subpoenas should follow the same legal standard guiding the Supreme Court’s evaluation of congressional subpoenas not implicating the president’s information. The House’s proposal, the chief justice finds, fails to account for profound separation of powers concerns, for it identifies no limits on Congress’s subpoena authority. According to the House proposal, the majority cautions, Congress could “walk away from the bargaining table” and enforce the disputed subpoena against the executive branch in court. This unlimited power would “transform” how the two branches have historically resolved information access disputes, altering the balance of power at the executive branch’s expense.
The majority finds these separation of powers concerns to be no less—and indeed perhaps more—prominent in the context of subpoenas seeking personal papers of the president or those served on third parties. Roberts states that the “President is the only person who alone composes a branch of government.” Because only an ambiguous line separates “the Office of the President and its occupant,” requests for personal information from the president may still trigger an interbranch dispute. Further, the sensitive nature of the requested documents and their possibly tenuous connection to a legislative purpose may heighten the risk of congressional abuse.
Finally, the chief justice states that these concerns over separation of powers are animated “no matter where the information is held—it is, after all, the President’s information.” If this were not so, he reasons, Congress could circumvent constitutional requirements and “declare open season on the President’s information” held by third parties.
Finding that neither the parties nor the courts below had adequately accounted for these separation of powers concerns, Roberts sets forth a four-factor balancing test for evaluating the validity of congressional subpoenas seeking personal information of the president. Courts weighing these factors are to account for the “significant legislative interests of Congress,” along with the “‘unique position’ of the President.”
While acknowledging that other factors may be relevant, Roberts sets out four in particular. First, courts should consider whether “other sources could reasonably provide Congress the information it needs in light of its particular legislative objective.” Second, subpoenas may be “no broader than reasonably necessary to support Congress’s legislative objective.” Third, “Congress must adequately identif[y] its aims and explai[n] why the President’s information will advance its consideration of the possible legislation.” Finally, courts “should be careful to assess the burdens imposed on the President by a subpoena.”
Thomas opens by arguing that Congress has no authority to subpoena personal documents—regardless of who they belong to. In his view, the only possibility of Congress obtaining these documents would be through a formal investigation of the president carried out under Congress’s impeachment powers.
The justice takes issue with the argument that Congress’s power to subpoena is an implied power under its power to legislate. He points to Marbury v. Madison, quoting: “The powers of the legislature are defined, and limited; and that those limits may not be mistaken, or forgotten, the constitution is written.” Moving from Marbury to McCulloch v. Maryland and Thomas’s own dissent in United States v. Comstock, he writes that there must be strong underlying evidence that the power was meant by the Founders to be implied. As a result, implied powers are “very limited.”
There is no evidence from the founding, Thomas argues, that the power of subpoenaing personal documents was implied to Congress. In his view, key precedents dealing with legislative subpoenas—Kilbourn v. Thompson and McGrain v. Daugherty—do not demonstrate any support for “private, nonofficial documents.”
Thomas rebuffs the House’s argument that Congress enjoys the same investigation powers as did the British Parliament when the United States was founded. “Parliament,” he writes, “was supreme. Congress is not.” He points out that Article I forbids “bills of attainder,” a power held by Parliament at the time of the Founding. Likewise, the “Bill of Rights” focuses on protecting the individual from the government—including Congress. He also notes the Supreme Court’s finding in Kilbourn that Congress has fewer investigative powers than Parliament.
What’s more, Thomas writes, 18th century American legislatures did not have these powers. Such legislatures sometimes sought testimony, but their requests concerned official government information, not personal information. It’s true that some legislatures in this period did conduct nonlegislative investigations—but in Mazars, the House was not claiming that authority but instead claimed an authority to investigate derived from the power to legislate. Finally, Thomas notes, the other areas investigated by colonial and state legislatures were radically different from the subjects investigated by Congress today. Those legislatures investigated issues such as libel and insults—something the current Congress would not touch.
Thomas then examines the evolution of congressional subpoenas beginning at the Founding, arguing that early investigations into generals and members of Congress did not look at personal documents. In fact, Thomas writes, the only instances in which Congress sought personal documents involved material from a federally chartered bank.
And the practice of issuing legislative subpoenas for nonofficial information was controversial throughout the 19th century. Thomas writes that when the Supreme Court first considered legislative subpoenas in Kilbourn, in 1880, it “cast … doubt on legislative subpoenas generally” and “held that the subpoena at issue was unlawful.” In McGrain, the Supreme Court later upheld Congress’s power to issue subpoenas as derived from its power to legislate—but, Thomas argues, McGrain “lacks any foundation in text or history.” Though the Supreme Court has rolled back Congress’s authority to compel testimony or documents after McGrain, Thomas would instead decline to apply McGrain altogether. For “it is readily apparent,” he writes, “that the Committees have no constitutional authority to subpoena private, nonofficial documents.”
Finally, Thomas writes that the only concrete example of Congress being able to subpoena personal documents has been through the impeachment process. If Congress seeks to obtain Trump’s financial records, he argues, “it should proceed through the impeachment power.”
Alito expresses admiration for Thomas’s argument but writes separately to argue against these particular subpoenas—assuming arguendo that they are not “categorically barred.” Where Congress is not using its impeachment power to subpoenas these documents, Alito argues, its reasoning must be held to a high standard. While Alito agrees with the majority that the cases must be remanded, he writes that the House should be required to be much more descriptive as to the legislation under consideration and how the records are relevant to that legislation. If the House is not ordered to do this, he concludes, then the subpoenas are inappropriate.