Note: since my last post the world has lost another lion of congressional oversight, Mort Rosenberg, who for 35 years was every House and Senate committee’s “go to” expert at the Congressional Research Service for advice and assistance on congressional investigations, executive privilege, vacancies, and many other issues. Mort is mentioned in this post as he has been in many others over the years. He will be greatly missed.
Today concludes the series of posts I began on April 8 on “Immunity, Impeachment and Juristocracy.” The prior posts are as follows:
Part I: Unoriginalist Sin (Apr. 8, 2025)
Part II: Nixon v. Fitzgerald (Apr. 9, 2025)
Part III: “Core” Official Conduct (Apr. 13, 2025)
Part IV: Noncore Official Conduct and Evidentiary Use (Apr. 17, 2025)
Part V: Above the Law (Apr. 27, 2025)
Part VI: Chilling Doom (Apr. 30, 2025)
The upshot of these posts is that Congress cannot rely on other branches or institutions to police, punish or deter executive branch crime or wrongdoing. This is most obviously true of the president, but it also applies to subordinate executive branch officers. Therefore, Congress must assume this responsibility itself.
Historically, congressional investigation of executive branch wrongdoing has generally been an outgrowth of ordinary congressional oversight. Only rarely do such investigations turn into (or begin as) formal impeachment inquiries and even more rarely does the House actually impeach an officer of the executive branch. And, of course, the Senate has never convicted a president or other executive branch officer. The primary consequences for executive branch wrongdoing have been reputational (amplified through the media), forced resignation, firing, and criminal sanctions. But for the reasons discussed in the prior posts (and illustrated by daily headlines), these consequences are much less likely than they once were and their deterrent effect has dramatically diminished if not evaporated.
Professor Jonathan Shaub has posted this article which grapples with the difficulty of holding the executive accountable in the current legal and political environment. See Jonathan David Shaub, White House Inspection, 103 Wash. U. L. Rev. __ (forthcoming 2026). He argues that Congress is ill-equipped to fill the gap:
Congress as investigator—whether pursuant to legislative oversight authority or impeachment authority—has a number of shortcomings, many, but not all, of which are well-known and noncontroversial. First, and perhaps most obviously, Congress is partisan, and, in our current polarized environment, this partisan character combined with the dominant two-party system hampers both the reality and perception of congressional investigation. Second, Congress lacks the ability to enforce its investigative demands on any realistic timeframe. Third, perhaps less obviously, Congress is a poor choice because it lacks expertise and key institutional incentives—in part because of its politicization—related to the confidentiality of investigative files.
White House Inspection, supra, at *42.
None of these concerns are unfounded, but, as discussed below, I think there are reasonably persuasive responses to each. Whether Congress will prove capable of meeting the challenge of the present moment only time will tell. But in a very real sense there is no alternative. Congress is the only institution with the inarguable constitutional authority and duty to hold the executive accountable. Even Shaub’s proposal (which we will discuss shortly) depends in large part on Congress fulfilling the role the framers entrusted to it.
Shaub’s Objections to Congressional Investigations
Partisanship and Polarization. Shaub acknowledges that because Congress is separately elected and independent of the executive branch, “there is an argument that Congress is the ideal of an independent, otherwise accountable investigator.” White House Inspection, supra, at *42. He contends, however, that in our hyper-partisan age it is impossible for Congress to be or to be perceived as being a fair or neutral investigator, which he contrasts with the Department of Justice’s aspiration that “its actions have ‘the appearance and reality of fairness and impartiality.’” Id. at *43.
It is undeniable that congressional investigations have the reputation of being more about politics and publicity and less about procedural regularity and professional rigor. One is more likely to associate congressional investigations, as opposed to ordinary law enforcement or administrative investigations, with terms like “political theater,” “show trial,” or “witch hunt.” See generally Josh Chafetz, Congressional Overspeech, 89 Fordham L. Rev. 529, 530-35 (2020).
We need not debate here whether this reputation is fully deserved or, for that matter, whether the performative aspects of congressional investigations are always a bad thing (consult Professor Chafetz’s article for a discussion of the latter). Suffice to say that there are two distinctive aspects of congressional investigations that are unlikely to change: (1) they are typically more open and public than their executive branch counterparts; and (2) they almost always consist of a majority and minority with different (though not always diametrically opposed) perspectives and interests.
It is another matter whether these distinctions are best summed up by saying that congressional investigations are (relatively) lacking in fairness and impartiality. There are no doubt senses and situations in which this broad generalization is accurate and others in which it is not. Congressional investigations may lack some of the procedural structure of executive branch investigations, but it is still better to be the target of a congressional investigation than a criminal or civil investigation by a federal agency. If the U.S. marshals are knocking on my door, I would rather they be serving a congressional subpoena than a grand jury subpoena. To say nothing of whatever paperwork ICE agents carry when they have decided you qualify for an all expenses paid trip to El Salvador.
To be sure, reality and reputation are two different things, and there is little question that congressional investigations have long been perceived as more political and partisan and less professional and objective than federal law enforcement. Thus, it may well be true, as Shaub says, that “even if a congressional investigation were objective and entirely fact-based, it would likely not be perceived as such.” White House Inspection, supra, at *43. The hyper-polarization of public opinion, however, means this is probably true of any type of investigation. Since Congress has a historical disadvantage in reputation relative to other institutions, it may be that the current informational environment places Congress on a more even playing field than it otherwise would be.
More importantly, though, over the last few months the institutional advantage of the Department of Justice (and the executive branch generally) has rapidly eroded if not entirely evaporated. No one thinks that Attorney General Bondi or FBI Director Patel are going to place impartiality in law enforcement above personal loyalty to the president. There is not really a pretense otherwise. The “presumption of regularity” that executive branch decisions formally enjoy in the courts, and informally enjoy in the media, academia and elite opinion, is unsustainable under the current circumstances. Perhaps that could change somewhat if the Supreme Court and the administration reach an equilibrium on a host of issues that have shocked the conscience of the lower courts (deportations, birthright citizenship, attacks on law firms, impoundments, among other things). But when the president routinely says things like “I don’t know” in response to the question whether he has the duty to uphold the Constitution, it is hard to imagine things will return to anything close to their prior state.
In the land of the blind, the one-eyed man is king. Whatever its investigative deficiencies, Congress has the opportunity to establish itself as the more responsible and credible institution. It can do that by following best practices and, wherever possible, aiming for bipartisan cooperation in its investigations. See Carl Levin & Elise J. Bean, Defining Congressional Oversight and Measuring its Effectiveness, 14 Wayne L. Rev. 1 (2018). This is obviously difficult when the target of the investigation is the president of the same political party as either the committee majority or minority (most likely the latter). But it is at least possible when the investigation focuses on subordinate officials, which may often be a more prudent way of seeking to expose and deter executive branch wrongdoing.
Furthermore, Congress can take steps to bolster its institutional capacity and make its investigations more procedurally regular and professionally rigorous. I will not attempt to detail them here, but one idea is to establish an internal check that regulates when committees can escalate from ordinary legislative oversight to investigation of specific executive crime or wrongdoing. Congress could also enhance its investigative capacity by establishing an office of congressional legal counsel or bolstering the capabilities of its existing legal offices (see this GAO report for details). Congress has a historic opportunity to redress the imbalance in legal talent between the legislative and executive branches by recruiting some of the top lawyers that the executive has fired, spurned or repulsed.
By building its own capacity and internal discipline, Congress can enhance its “soft power,” which, as Chafetz explained in a slightly different context, “can build or restore public trust in the institution, thus enhancing its ability to engage successfully in the public sphere.” Josh Chafetz, Congress’s Constitution 232 (2017).
Inability to Enforce Information Demands. Shaub contends that “Congress lacks the power to compel the production of information necessary to conduct a comprehensive investigation.” White House Inspection, supra, at *44. In particular, he notes, Congress has difficulty getting information out of the White House because of the longstanding executive branch position that top White House aides need not appear in response to congressional subpoenas. Id. at *45.
Readers of this blog know that I have written at length (some would say at nauseating length) about this subject. For the most part I have focused on the legal merits and procedural mechanics of congressional-executive information disputes. Here, however, I want to step back and look at the subject through a somewhat broader lens.
There can be no question Congress needs information to be able to accomplish, its legislative, oversight and impeachment functions. Moreover, as the Supreme Court recognized nearly a century ago, “[e]xperience has taught that mere requests for such information often are unavailing, and also that information which is volunteered is not always accurate and complete, so some means of compulsion are essential to obtain what is needed.” McGrain v. Daugherty, 273 U.S. 135, 175 (1927). This observation is as true for information sought from the White House or the executive branch as it is for information sought from anyone else.
The importance of compulsory process, however, should not be overstated. A great deal of information, including information on the activities of the U.S. government, can be obtained without compulsory process or the threat of compulsory process. (Just ask Jeffrey Goldberg or the intelligence services of any foreign government, if you don’t believe me). This is far more true today than it was in 1927.
An argument could be made that, strictly from the point of view of maximizing its ability to scrutinize various executive programs and actions, Congress would be better off focusing its efforts on obtaining information that is available without using compulsory process. For example, consider how much information the PSI minority was able to gather about Elon Musk’s conflicts of interest for this report. I can think of many blockbuster oversight hearings Congress could hold right now using just publicly available information, whistleblowers, disgruntled former federal officials and voluntary witnesses from outside the government. (I will discuss one example involving pardon abuse in some detail in a future post).
To be sure, there is a difference between conducting general oversight of executive branch programs and policies, on the one hand, and investigating particular factual allegations and controversies, on the other. Compulsory process is more important when the latter is involved, as it was in McGrain (where the Senate was investigating specific charges of misfeasance by Attorney General Daugherty).
For example, Congress could hold hearings today on Signalgate, but if it wanted to answer the specific question of how Jeffrey Goldberg was added to the Signal chat, it would need compulsory process to get records from Signal and the White House and testimony from White House aides and other key witnesses. The administration would almost certainly refuse to provide evidence (based on executive privilege, state secrets, absolute immunity or just the fact that it doesn’t want to) and it might try to stop Signal from providing information as well. Under current circumstances, as Shaub would correctly observe, there is little chance Congress could overcome determined resistance to get this information within a reasonable timeframe.
While this is lamentable, one should not leap to the conclusion that Congress is incapable of successfully investigating executive crime or wrongdoing. Even if Congress cannot directly compel the executive branch to comply with subpoenas, it has a variety of tools to minimize, mitigate or circumvent compliance.
For example, while I find absurd the OLC position that White House aides are absolutely immune from testifying before Congress, even if they are fact witnesses to bribery or some other impeachable act, this does not mean that Congress is helpless to investigate or adjudicate the offense. After all, crimes are prosecuted every day even though prosecutors are prohibited from obtaining the single most important piece of evidence in most criminal cases, namely the testimony of the accused. Congress can likewise find other sources of evidence to make its case. It can also put pressure on the witnesses by holding them in criminal contempt (a crime that could theoretically be prosecuted by a future administration). It can put pressure on the White House itself to cooperate (e.g., by withholding funds). It can employ evidentiary presumptions that testimony withheld by the White House would be unfavorable to the president or other officer who is the subject of impeachment proceedings. It can even treat the refusal to cooperate as itself impeachable obstruction of Congress, as the House did (rather unsuccessfully) during the 2019-20 Trump impeachment.
Here it should be noted that these mechanisms are more likely to be successful if the investigation is not targeted at the president personally. If the president is the direct target, the administration will have little incentive to cooperate with the investigation. But if Congress focuses on crime or wrongdoing by subordinates, whether or not they are acting in accordance with the president’s wishes, Congress’s relative interests in enforcing its information demands become stronger and the executive’s in resisting become weaker. (In the Signalgate situation, for example, it is plausible that a relatively moderate amount of political pressure might be sufficient to convince the administration to cooperate). Think of it as a modern version of the principle that, while the king could do no legal wrong, “there was always a servant of the Crown, subject to parliamentary jurisdiction, who could be held responsible for any wrong done in the king’s name.” Chafetz, Congress’s Constitution 80 (2017).
Congress’s ability to enforce compulsory process, however, is not simply a matter of its practical ability to obtain the information it needs. Perhaps even more important are the implications for its institutional legitimacy. Here a comparison with the courts is instructive. No court has ever “compelled” the executive branch to provide information if by “compelled” one means forced by fines or imprisonment or the imminent threat thereof. It is doubtful that the courts have a means of compulsion if the president himself refuses to provide information. Yet the ultimate authority of courts to determine if the executive branch must provide information in judicial proceedings has rarely been questioned, and President Nixon famously complied with the Supreme Court’s decision that he must turn over the White House tapes.
By the same token Congress’s institutional legitimacy requires that it have the authority to determine what information is relevant and necessary in its own proceedings. This does not necessarily mean that it has the unilateral right to compel the production of information subject to a claim of privilege or other legal right to withhold, but it does mean that congressional demands for information have the same prima facie validity as judicial demands for information.
Unfortunately, Congress itself bears some responsibility for devaluing its own investigatory authority. The House has made it very easy for committees to issue subpoenas by delegating to the chairman the authority to do so unilaterally. This has led to many subpoenas being issued without regard to whether there is a plan to enforce them or a willingness to bear the costs of doing so. When subpoenas are defied or ignored by the executive branch, the typical response is for committee members to issue critical press releases or go on television or social media to complain, and then to move on to the next issue. Over time, as Professor Shaub has observed, “[s]ubpoenas have thus become just another part of the political theater that is the oversight process—performance rhetoric that does not have any legal effect in practice.” Jonathan David Shaub, The Executive’s Privilege, 70 Duke L. J. 1, 42 (2020).
Congress (the House in particular) should address this problem with reforms that control the use of compulsory process for purely political or symbolic purposes. It is probably too late to take away from committee chairs the power to issue subpoenas, but the House can require procedures both before subpoenas are issued (such as mandating an attempt to reach accommodation with the executive branch) and before enforcement (to ensure that both subpoena recipients and other institutions know that enforcement will occur only when the House is serious about its intent to obtain the information at issue). In 2019, I sketched out one such proposal here.
Times have changed, however, since 2019. At the time my focus was on an efficient way for Congress to obtain executive branch information with the assistance of the courts. Now I believe Congress also needs a mechanism to compel the production of information without the assistance of the courts. This is true for two reasons. First, as the prior posts in this series show, the Supreme Court in recent years has leaned ludicrously in the direction of executive power. Second, building Congress’s “soft power” requires that it have an independent means of vindicating its institutional interests without relying on the courts. This is so even though there is reason for cautious optimism that the courts will be more favorable to congressional interests in the current environment of blatant executive overreach.
This means once again dusting off the idea of inherent or direct contempt, which (as readers of this blog well know) is the process by which Congress may arrest, try and imprison a recalcitrant witness until he purges his contempt by providing the information demanded. A variant of this process, proposed by the late Mort Rosenberg, would use civil fines, rather than imprisonment, as the punishment to compel compliance. The advantage of Rosenberg’s proposal would be that contempt could be employed against an executive official without the unseemly and dangerous prospect of physically arresting her. Rosenberg also proposed the idea of the House appointing its own lawyer to serve as a private prosecutor of criminal contempt if the executive branch refuses to prosecute. (This latter proposal, which is similar to what Judge Boasberg is considering in the Alien Enemies Act case, would face some serious constitutional objections).
All of these ideas are worth considering, but the most important thing is for Congress to demonstrate the willingness and ability to devote resources and political capital toward making its investigations effective. It certainly cannot rely on the executive branch to prosecute contemnors, even if they are not asserting a governmental privilege. For example, there apparently has been no effort by either the Biden or Trump administrations to prosecute Ralph de la Torre, a healthcare CEO who failed to appear in response to a Senate subpoena, and who was referred for prosecution by a unanimous Senate vote in September 2024.
If Congress is to re-establish its authority to use inherent contempt, it will not want to begin with a witness who is a current executive branch official. Nor will it want to proceed against a witness unless he has important information, and it can be shown that this information is critical to a significant inquiry. The witness should be of demonstrable significance at least equal to that of Mal Daugherty, the brother of then Attorney General Harry Daugherty, whose contempt was at issue in McGrain. As in McGrain, though, the investigation would ideally involve oversight of the executive branch and potential executive crime or wrongdoing.
To be sure, inherent contempt does not preclude judicial involvement. The contemnor has the right to seek relief from the courts through a habeas petition. But Congress will not be in court as a litigant pleading for judicial assistance, but as a co-equal branch of government whose actions should be entitled to a presumption of regularity. Moreover, so long as the inherent contempt proceeding is handled correctly, Congress will be on firm legal ground under the authority of McGrain and its progeny. There are undoubtedly some justices (well, at least one ) on the Supreme Court who would overrule McGrain or read it so narrowly as to make it of little value. Nonetheless, if there were ever a time that the Court can appreciate the need for vigorous congressional oversight of the executive, that time is now.
Confidentiality Concerns. Shaub’s final objection relates to the difficulty of keeping investigative material confidential in a congressional investigation. As an institution, Congress is less oriented toward secrecy than the executive. Congress lacks incentives to keep information confidential and often has “overwhelming partisan incentives to disclose information damaging to the other party.” White House Inspection, supra, at *46. Congressional rules establish a presumption that hearings will be open and committee files are generally accessible to all members of the committee and even non-committee members. Shaub notes that “[t]his diffusion of authority and information inevitably leads to more opportunities for and less sanction of information leaks and politically motivated selective disclosures, even of highly sensitive information.” Id.
Concerns about Congress’s inability to maintain confidentiality are neither new nor unfounded. As James Hamilton, who served as counsel to the Senate Watergate Committee, wrote a half-century ago, “Congress has long been plagued by improper or injudicious release of information received in confidence.” James Hamilton, The Power to Probe 273 (1976). Hamilton devotes a chapter of his book to complaints about leaks from the Watergate investigation and to criticism of the cavalier attitude of many in official Washington toward such leaks. Id. at 273-300. Senator Ervin himself comments in the introduction that “I join Mr. Hamilton in deploring the inexcusable leaks of confidential information by persons who Sam Dash, the chief counsel for the Senate Select Committee, and I were unable to identify. Id. at XV.
On the other hand, the Watergate investigation and many others have been successful despite the problem of leaks. Congress, moreover, could adopt rules, practices and structures that limit leaks if it were so incline, as examples such as the intelligence committees, the ethics committees and the Joint Committee on Taxation illustrate.
Potential reforms would include establishing a professional nonpartisan staff to handle sensitive investigations where confidentiality is at a premium. Such a staff could be located outside the existing committee structure and could be headed by an appointed congressional official, who would report either to leadership (as the parliamentarians, legal counsel and other congressional officials do) or to a special committee on executive crime (perhaps with equal party representation along the lines of the ethics committees). More modest reforms, such as limiting access to sensitive information to the chair and ranking members of the investigating committee, might well be sufficient for ordinary investigations of executive branch wrongdoing.
In the most high profile investigations, particularly those involving allegations of wrongdoing by the president himself, maintaining confidentiality will be challenging for any institution, not just Congress. This was true even during the Ken Starr investigation of Clinton and other independent counsel investigations of yore (i.e., before everyone carried a personal communication device capable of instant transmission to the entire world). That is another reason why Congress may want to focus its investigations on subordinate executive officials, instead of trying to make every investigation into the next Watergate.
In short, confidentiality is a challenge for Congress, but its relative disadvantage in this area (a) should not be exaggerated, (b) can be mitigated, and (c) is not reason enough to outsource its investigative responsibilities to others.
Shaub’s Proposal
To compensate for the deficiencies he sees in both congressional and special counsel investigations, Shaub calls for “the creation of an independent, bipartisan, continuing, multimember body charged, at a minimum, with inspection of the president and the White House.” White House Inspection, supra, at *60. This commission would be located in the executive branch, but it would be given statutory independence (including for cause removal protection for its board) from the White House and DOJ. It would be given authority to issue subpoenas which could be enforced judicially, but the statute would require that the judicial proceedings be sealed to protect confidentiality. It would have no authority to impose sanctions, but it would be required to report the results of its investigations (or “inspections,” as Shaub terms them) to Congress, either at their conclusion or on a rolling basis. However, the commission would also be subject to robust confidentiality protections and would have the right to withhold certain details and its underlying files from Congress. See id. at 60-69.
My concerns with Shaub’s proposal begin with the question of when and how a statute of this nature could be enacted into law. The one thing we can say for sure is that it will not be during the current administration. Perhaps a future administration can be convinced to support or accept it, but it is likely that the price of doing so will include statutory changes designed to protect the president from investigations or investigatory steps he finds unacceptable. No administration is going to want to create its own Jack Smith, or Robert Hur, or Bob Mueller, or Ken Starr, or Lawrence Walsh, or . . . well, you get the picture.
Even if the statute is enacted more or less as Shaub has proposed it, there are reasons to question how it would operate. The first one to come to mind is what would trigger an inspection by the commission? Presumably inspections would be authorized by a vote of the multimember board but what type of information would be adequate to trigger an inspection? Could it be simply an allegation in the media? A complaint from Congress (of which there would be many)? And what threshold, if any, would need to be crossed for the allegations to warrant an inspection? Shaub suggests that inquiries would involve “misconduct,” but there is a broad spectrum of activity that could fall within that term. Would the commission have rules excluding misconduct which is too minor or personal or which is adequately policed by other processes (e.g., existing executive branch agencies that deal with conditions of employment, ethics, etc.)? What about abuse of presidential powers such as the pardon power? Or a range of other actions like signing executive orders targeting law firms or universities, or withholding foreign aid that Congress has mandated, or raising tariffs, or forgiving student loans, or giving amnesty to illegal aliens, or sending illegal aliens to foreign prisons, or . . . well, you get the picture.
Assume, though, that this problem can be worked out. Leave aside also the question whether the statutory guarantees of independence, particularly the “for cause” removal protections for the members of the commission, will survive constitutional scrutiny, which Shaub acknowledges cannot be taken for granted given the proclivities of the current Court. See White House Inspection, supra, at **73-76. The proposal still depends on critical assumptions that are, at best, yet unproven.
First, Shaub assumes that the White House will treat the commission more favorably than it treats Congress due to the former’s placement in the executive branch and its robust confidentiality protections. This assumption seems to rest first and foremost on generalizations about how past administrations have treated investigations by special counsels, inspectors general, and other entities in the executive branch. However, the current administration demonstrates that such generalizations are historically contingent and certainly not guarantees. (In his first term President Trump had no hesitation about treating the Justice Department and other departments or agencies run by his own appointees as “deep state” enemies). Nor is it obvious that even “normal” administrations will regard a new entity which is required to report regularly to Congress more favorably than it regards Congress itself.
Second, Shaub assumes that if the White House refuses to provide the commission with critical information, it will be able to obtain timely and effective judicial relief. Here again I have the impression Shaub is thinking of traditional disagreements that have arisen in the course of independent/special counsel investigations since Watergate, such as the Clinton administration’s assertion of a governmental attorney-client privilege. He is not thinking of a White House determined to keep an investigation from getting any information or conducting a scorched earth defense based on any conceivable legal or political argument.
As noted earlier, the ability of courts themselves to “compel” the executive branch to produce information itself rests largely on norms which can be threatened or disregarded by an administration contemptuous of such norms (or any administration if the political stakes are high enough). Moreover, the courts maintain these norms in part by attempting to avoid or delay conflict with the executive branch, particularly at the highest level, wherever possible. Thus, the same factors which make it difficult or impossible for congressional investigators to rely on the availability of judicial relief may likewise hamper the inspections of the proposed commission.
None of this is to say that Shaub’s proposal is hopeless or unworthy of consideration. An outside entity devoted to gather information on executive branch misconduct (however defined) could doubtless be helpful to Congress and to repairing the damage done by Trump and other decisions. I particularly like the idea of delegating to the commission the power of conducting presidential competency inspections pursuant to the 25th amendment, which would give it a direct constitutional authority and make it less likely to be ignored by the White House. See White House Inspection, supra, at **77-78. (On the other hand, try getting the president to sign that into law).
It is to say, however, that Congress should neither wait for nor count on the establishment of such a commission. Even if such a commission existed, Congress cannot and should not rely on it to fulfill its own constitutional role and responsibilities. Since the commission would have no power to impose sanctions or take any remedial actions, it would be largely ineffective unless Congress has the institutional capacity and “soft power” to act, whether in the form of impeachment proceedings or legislation, based on the information provided by the commission. Indeed, I would argue that the very fact we think an executive branch commission would have more legitimacy to conduct oversight of the executive branch than Congress is a symptom of the problem we must address.