Text extracted via OCR from the original document. May contain errors from the scanning process.
Case 19-1540, Document 215, 12/03/2019, 2719370, Page1 of 59
DEBRA ANN LIVINGSTON, Circuit Judge, concurring in part and dissenting in part:
Although not expressly provided for in the Constitution, Congress’s power
to conduct investigations for the purpose of legislating is substantial, “as
penetrating and far‐reaching as the potential power to enact and appropriate
under the Constitution.” Eastland v. U.S. Servicemen’s Fund, 421 U.S. 491, 504 n.15
(1975) (quoting Barenblatt v. United States, 360 U.S. 109, 111 (1959)). Yet this power
is not unlimited. When Congress conducts investigations in aid of legislation, its
authority derives from its responsibility to legislate—to consider the enactment of
new laws or the improvement of existing ones for the public good.1 Congress has
no power to expose personal information for the sake of exposure, see Watkins v.
United States, 354 U.S. 178, 200 (1957) (expressing “no doubt that there is no
congressional power to expose for the sake of exposure” (emphasis added)), nor
None of these subpoenas issued in connection with an impeachment proceeding,
in which Congress’s investigatory powers are at their peak, but rather, as stated, “in aid
of legislation.” See Kilbourn v. Thompson, 103 U.S. 168, 190 (1880) (noting that “[w]here
the question of . . . impeachment is before [the House or the Senate] acting in its
appropriate sphere on that subject, we see no reason to doubt the right to compel the
attendance of witnesses, and their answer to proper questions, in the same manner and
by the use of the same means that courts of justice can in like cases”); Senate Select Comm.
on Presidential Campaign Activities v. Nixon, 498 F.2d 725, 732 (D.C. Cir. 1974) (citing to
Article I, Section 2 of the Constitution when noting that impeachment investigations in
the House have “an express constitutional source” which differentiates them from
Congress’s general oversight or legislative power).
1
1
Case 19-1540, Document 215, 12/03/2019, 2719370, Page2 of 59
may it seek information to enforce laws or punish for their infraction—
responsibilities which belong to the executive and judicial branches respectively,
and not to it. Id. at 187 (noting that Congress is neither “a law enforcement [n]or
trial agency,” as “[t]hese are functions of the executive and judicial departments
of government”). As the Supreme Court has put it: “No inquiry is an end in
itself; it must be related to, and in furtherance of, a legitimate task of the Congress.
Investigations conducted solely for the personal aggrandizement of the
investigators or to ‘punish’ those investigated are indefensible.” Id.
The legislative subpoenas here are deeply troubling. Targeted at the
President of the United States but issued to third parties, they seek voluminous
financial information not only about the President personally, but his wife, his
children, his grandchildren, his business organizations, and his business
associates. 2 Collectively, the subpoenas seek personal and business banking
records stretching back nearly a decade (and with regard to several categories of
The Plaintiff entities here are defined to include not only parents, subsidiaries,
related joint ventures and the like, but any “current or former employee, officer, director,
shareholder, partner, member, consultant, senior manager, manager, senior associate,
staff employee, independent contractor, agent, attorney or other representative of any of
those entities,” so that the banking records of numerous individuals beyond the
President’s immediate family are potentially included in this dragnet. J.A. at 47, 58.
2
2
Case 19-1540, Document 215, 12/03/2019, 2719370, Page3 of 59
information, with no time limitation whatsoever) and they make no distinction
between business and personal affairs, nor consistently between large and small
receipts and expenditures. To be sure, breadth may be necessary in legislative
subpoenas so that Congress can learn about a proposed subject of legislation
sufficiently to enact new laws or improve the old ones: such learning is “an
indispensable ingredient of lawmaking.” 3 Eastland, 421 U.S. at 505. Still, the
district court was of the view that in a routine civil case, it would have sent the
parties into a room with the instruction that “you don’t come out until you come
back with a reasonable subpoena.” J.A. at 94. The majority doesn’t disagree.
It, too, characterizes the subpoenas as “surely broad in scope.” Maj. Op. at 45.
It acknowledges that compliance will “subject [the President’s] private business
affairs to the Committees’ scrutiny,” id. at 48, and impose irreparable harm, id. at
13. It could have added that personal banking records of the President and his
family are not excluded, and that neither House committee seeking this
That said, “legislative judgments normally depend more on the predicted
consequences of proposed legislative actions and their political acceptability, than on
precise reconstruction of past events,” which appears to be the focus of the present
subpoenas. Nixon, 498 F.2d at 732.
3
3
Case 19-1540, Document 215, 12/03/2019, 2719370, Page4 of 59
information will commit to treating any portion of it as confidential, irrespective
of any public interest in disclosure. J.A. at 122–23.
The majority and I are in agreement on several points. First, we agree that
the Right to Financial Privacy Act (“RFPA”), 12 U.S.C. §§ 3401–3423, does not
apply to Congress because, as the majority correctly concludes, Congress is not a
“Government authority” within the meaning of that statute. Maj. Op. at 24–33.
We likewise agree that 26 U.S.C. § 6103 of the Internal Revenue Code does not pose
an obstacle to Deutsche Bank AG’s disclosure of tax returns in its possession in
response to the Committees’ subpoenas. Id. at 34–44. Accordingly, I concur
that, as to the statutory arguments presented by the Plaintiffs, they have raised no
serious question suggesting that the House subpoenas may not be enforced.
The statutory arguments, however, are not the only arguments presented.
The majority and I agree that this appeal raises an important issue regarding the
investigative authority of two committees of the United States House of
Representatives—the House Committee on Financial Services and the House
Permanent Select Committee on Intelligence (collectively, “Committees”)—in the
context of their efforts to obtain voluminous personal and business banking
records of the President of the United States, members of his immediate family,
4
Case 19-1540, Document 215, 12/03/2019, 2719370, Page5 of 59
his primary business organization and affiliated entities, and his business
associates. Maj. Op. at 4. In fact, the question before us appears not only
important (as the majority acknowledges) but of first impression: the parties are
unaware of any Congress before this one in which a standing or permanent select
committee of the House has issued a third‐party subpoena for documents
targeting a President’s personal information solely on the rationale that this
information is “in aid of legislation.” Trump Br. at 14; Tr. of Oral Arg. at 34:24–
35:3–4. But this House has now authorized all such House committees to issue
legislative subpoenas of this sort, so long as directed at information involving this
President, his immediate family, business entities, or organizations. H.R. Res.
507, 116th Cong. (2019); see also H.R. Res. 509, 116th Cong. § 3 (2019) (“House
Resolution 507 is hereby adopted.”).
In such a context, “experience admonishes us to tread warily.” United
States v. Rumely, 345 U.S. 41, 46 (1953). I agree with the majority that our review
of the denial of a preliminary injunction is “appropriately more exacting where
the action sought to be enjoined concerns the President . . . in view of ‘[t]he high
respect that is owed to the office of the Chief Executive,’” Maj. Op. at 11 (quoting
Cheney v. U.S. Dist. Court, 542 U.S. 367, 385 (2004)). We disagree, however, as to
5
Case 19-1540, Document 215, 12/03/2019, 2719370, Page6 of 59
the preliminary injunction standard to be applied. In my view, a preliminary
injunction may issue in a case of this sort when a movant has demonstrated
sufficiently serious questions going to the merits to make them a fair ground for
litigation, plus a balance of hardships tipping decidedly in that party’s favor, that
the public interest favors an injunction, and that the movant, as here, will
otherwise suffer irreparable harm. See Citigroup Glob. Markets, Inc. v. VCG Special
Opportunities Master Fund Ltd., 598 F.3d 30, 35, 38 (2d Cir. 2010).
And as to the merits showing, I respectfully disagree with the majority’s
determination that the Plaintiffs’ constitutional arguments and those raised by the
United States as amicus curiae are insubstantial—not sufficiently serious for closer
review.4 Maj. Op. at 89–98. I cannot accept the majority’s conclusions that “this
case does not concern separation of powers,” id. at 89, and that there is “minimal
at best” risk of distraction to this and future Presidents from legislative subpoenas
of this sort, id. at 97. Instead, I conclude that the Plaintiffs have raised serious
Given my determination herein that the Plaintiffs have made a showing of
“serious questions” as to the merits and that this case must be remanded, I need not now
address whether the Plaintiffs have also satisfied the “likelihood of success” standard—
and I do not do so, given the obligation in this context to avoid unnecessary judicial
determinations on constitutional questions implicating Congress’s investigative powers.
See Rumely, 345 U.S. at 46. I note, however, that I do not concur in the majority’s
determination that as to the present reach of these subpoenas, the Plaintiffs have shown
no likelihood of success.
4
6
Case 19-1540, Document 215, 12/03/2019, 2719370, Page7 of 59
questions on the merits, implicating not only Congress’s lawmaking powers, but
also the ability of this and future Presidents to discharge the duties of the Office of
the President free of myriad inquiries instigated “more casually and less
responsibly” than contemplated in our constitutional framework. Rumely, 345
U.S. at 46.
Nor do I agree with the majority’s determination substantially to affirm the
judgment and order compliance with these subpoenas. The majority itself
recognizes that these broad subpoenas cannot be enforced precisely as drafted
because they call for the production of material that may either bear “an
attenuated relationship” to any legislative purpose or that “might [even] reveal
sensitive personal details having no relationship to the Committees’ legislative
purposes.” Maj. Op. at 84 (emphasis added). The majority remands for a
culling process pursuant to which information disclosing, for instance, the
payment of medical expenses would be exempt from disclosure. Id. The
majority’s limited culling, however, is tightly restricted to specified categories of
information, leaving out almost all “business‐related financial documents” from
any review by the district court, id., irrespective of any threatened harm from
disclosure, and potentially leaving out substantial personal information as well.
7
Case 19-1540, Document 215, 12/03/2019, 2719370, Page8 of 59
Indeed, given the tight limitations imposed by the majority on the district court’s
review, even sensitive records reflecting personal matters unrelated to any
conceivable legislative purpose could potentially be disclosed.
I agree with the majority that remand is necessary. But we disagree as to
the reasons why. I conclude that the present record is insufficient to support the
majority’s determination that the voluminous records of Plaintiffs sought from
Deutsche Bank AG (“Deutsche Bank”) and Capital One Financial Corporation
(“Capital One”) should at this time be produced. 5 The majority concludes in
advance—before these records have been assembled—that only a select “few
documents” will implicate privacy concerns or bear “such an attenuated
relationship” to any legislative purpose that “they need not be disclosed.” Maj.
Op. at 85 (emphasis added). I disagree that the present record is sufficient to
make that determination or to conclude, more fundamentally, where the balance
of hardships lies with regard to the preliminary relief that the Plaintiffs seek. In
this sensitive separation‐of‐powers context, serious questions have been raised as
The Plaintiffs challenge the subpoenas as they relate to the banking records of
President Donald J. Trump, his family, and his businesses—the Plaintiffs here. Trump
Br. at 1. To the extent the subpoenas seek other information related to parties who are
not Plaintiffs, the subpoenas have not been challenged and are not part of this appeal.
5
8
Case 19-1540, Document 215, 12/03/2019, 2719370, Page9 of 59
to the historical precedent for these subpoenas; whether Congress has employed
procedures sufficient to “prevent the separation of power from responsibility,”
Watkins, 354 U.S. at 215, in seeking this President’s personal information; and
whether the subpoenas are supported by valid legislative purposes and seek
information reasonably pertinent to those purposes, see Quinn v. United States, 349
U.S. 155, 161 (1955) (noting that Congress’s power to investigate “cannot be used
to inquire into private affairs unrelated to a valid legislative purpose”). These
questions, like the balance of hardship question, also require further review.
As set forth herein, I would remand, directing the district court promptly to
implement a procedure by which the Plaintiffs may lodge their objections to
disclosure with regard to specific portions of the assembled material and so that
the Committees can clearly articulate, also with regard to specific categories of
information, the legislative purpose that supports disclosure and the pertinence of
such information to that purpose. The objective of this remand is the creation of
a record that is sufficient more closely to examine the serious questions that the
Plaintiffs have raised and to determine where the balance of hardships lies with
regard to an injunction in this case, and concerning particular categories of
information. The district court acknowledged that in a routine civil case, it would
9
Case 19-1540, Document 215, 12/03/2019, 2719370, Page10 of 59
not have ordered the disclosures here. The majority errs in implicitly concluding
that a President has less protection from the unreasonable disclosure of his
personal and business affairs than would be afforded any litigant in a civil case.
Only on the basis of this fuller record would I determine the question
whether a preliminary injunction should have issued, and with regard to what
portions of the records sought. In reaching this conclusion, I am guided by the
Supreme Court’s counsel in Rumely that in the context of delicate constitutional
issues involving limits on the investigative power of Congress, our duty is to avoid
pronouncement “unless no choice is left.” Rumely, 345 U.S. at 46; cf. Cheney, 542
U.S. at 389–90 (suggesting that courts should “explore other avenues” to avoid
adjudicating “overly broad discovery requests” and “unnecessarily broad
subpoenas” that present “collision course” conflicts between coequal branches).
Indeed, Rumely affirms that the duty of constitutional avoidance is “even more
applicable” in the context of congressional investigations “than to formal
legislation.” Rumely, 345 U.S. at 46; see also Tobin v. United States, 306 F.2d 270, 275
(D.C. Cir. 1962) (recognizing duty of courts in appropriate circumstances to avoid
“passing on serious constitutional questions” presented by Congress’s exercise of
10
Case 19-1540, Document 215, 12/03/2019, 2719370, Page11 of 59
its investigative power). Decision here may be required, but is premature on the
present record.
Remand will also afford the parties an opportunity to negotiate. This is not
the essential point of the remand I propose, but efforts at negotiation in this context
are to be encouraged, since they may narrow the scope of these subpoenas, and
thus avoid judicial pronouncement on the “broad confrontation now tendered.”
United States v. Am. Tel. & Tel. Co (AT&T I), 551 F.2d 384, 395 (D.C. Cir. 1976). The
Plaintiffs have repeatedly sought the opportunity to negotiate. Reply Br. at 6–7;
Tr. of Oral Arg. at 17:18–19, 18:3–20, 66:7‐67:2. And the Committees, while
preferring the more immediate disposition that the majority affords them, have
expressed a willingness to attempt negotiation on an expedited basis if requested
by this Court.6 Tr. of Oral Arg. at 46:8–19.
Before this Court, counsel for the Committees stated that “[i]f this court thinks
there should be negotiation . . . [p]lease make it really, really fast, because we think that
Mr. Trump’s statements make clear this is absolutely insincere . . . [b]ut fine, give us a
day.” Tr. of Oral Arg. at 46:8‐15. Counsel for the Plaintiffs specifically affirmed in
response that “I don’t think there is any basis to determine that we are being insincere,
and I certainly welcome, I think that we have made clear, sending this case back down
for judicially refereed negotiations on whatever timeline the court thinks is appropriate
is absolutely something we are willing to participate in in good faith.” Tr. of Oral Arg.
at 66:21‐67:2.
6
Referencing an October 8, 2019 letter from Pat A. Cippolone, Counsel to the
President, to the Speaker of the House of Representatives and three House committee
chairs (a letter that is not part of the record before us), the majority concludes that a
11
Case 19-1540, Document 215, 12/03/2019, 2719370, Page12 of 59
To be clear, and as set forth herein, the Plaintiffs have raised serious
questions on the merits as to these subpoenas, which implicate profound
separation‐of‐powers concerns. 7 But pending the full remand that I outline
remand for negotiation is futile because the President has prohibited certain members of
the Administration from appearing in connection with the ongoing impeachment
inquiry. Maj. Op. at 71–72, 72 n.66. With respect, however, this letter references only
the “impeachment inquiry” and not the legislative investigations at issue here. This
letter thus provides no basis for this Court to disregard the express representations of the
Plaintiffs’ attorney that the Plaintiffs, including the President, seek to negotiate in good
faith.
The majority suggests that these subpoenas do not implicate separation of
powers because, inter alia, President Trump is not suing in his official capacity. Maj. Op.
at 70. I disagree. As in Rumely, “we would have to be that ‘blind’ Court . . . that does
not see what ‘(a)ll others can see and understand,’” not to recognize that these subpoenas
target the President in seeking personal and business financial records of not only the
President himself, but his three oldest children and members of their immediate family,
plus the records of the Trump Organization and a litany of organizations with which the
President is affiliated. Rumely, 345 U.S. at 44 (quoting Bailey v. Drexel Furniture Co., 259
U.S. 20, 37 (1922)); see also id. (acknowledging “wide concern, both in and out of Congress,
over some aspects of the exercise of the congressional power of investigation”); cf. Dep’t
of Commerce v. New York, 139 S. Ct. 2551, 2575 (2019) (noting that courts are “’not required
to exhibit a naiveté from which ordinary citizens are free’’’(quoting United States v.
Stanchich, 550 F.2d 1294, 1300 (2d Cir. 1977) (Friendly, J.))). Indeed, the Committees
themselves acknowledge that “President Trump and the Trump Organization” are the
focus of their investigations, see 165 Cong. Rec. H2698 (daily ed. Mar. 13, 2019) (statement
of Rep. Waters), and that “given the closely held nature of the Trump Organization,”
investigation must “include [the President’s] close family members,” District Court Doc.
No. 51 at 25–26. To be sure, Presidents are not immune from legislative subpoenas.
But as I explain below, this dragnet around the President implicates separation‐of‐powers
concerns for this and future Presidents, supporting a remand as to all the Plaintiffs here.
To the extent that certain of the requested records may ultimately be found not to
implicate separation‐of‐powers concerns, such a determination can only properly be
made following a remand for development of the record.
7
12
Case 19-1540, Document 215, 12/03/2019, 2719370, Page13 of 59
herein, I defer for now the question whether they have also shown a balance of
hardships tipping decidedly in their favor. The remainder of this opinion sets
out the reasons for my conclusions: (1) that the Plaintiffs have raised serious
constitutional questions as to these legislative subpoenas; and (2) that the serious
question formulation of the preliminary injunction standard is applicable,
contrary to the majority’s position.
I
A
To reiterate, the subpoenas here are very troubling. Congress “cannot
legislate wisely or effectively in the absence of information respecting the
conditions which the legislation is intended to affect or change.” Eastland, 421
U.S. at 504. At the same time, ill‐conceived inquiries by congressional
committees “can lead to ruthless exposure of private lives in order to gather data”
that is unrelated and unhelpful to the performance of legislative tasks. Watkins,
354 U.S. at 205. And the “arduous and delicate task” of courts seeking to
accommodate “the congressional need for particular information” with the
individual’s “personal interest in privacy,” id. at 198, does not grow easier when
Congress seeks a President’s personal information. Indeed, given the “unique
13
Case 19-1540, Document 215, 12/03/2019, 2719370, Page14 of 59
constitutional position of the President” in our scheme of government, see Franklin
v. Massachusetts, 505 U.S. 788, 800 (1992), and the grave importance of diligent and
fearless discharge of the President’s public duties, our task grows more difficult.
See Nixon v. Fitzgerald, 457 U.S. 731, 753 (1982) (recognizing that distraction from
public duties is “to the detriment of not only the President and his office but also
the Nation that the Presidency was designed to serve”).
The majority disagrees. It concludes that this case “does not concern
separation of powers” because the sought‐after records are personal, not official,
and because Congress “has not arrogated to itself any authority of the Executive
Branch,” nor “sought to limit any authority of the Executive Branch.” Maj. Op.
at 89. With respect, however, this conclusion gives too short shrift to the
Supreme Court’s analysis in Clinton v. Jones, 520 U.S. 681 (1997), on which the
majority principally relies. There, the Supreme Court concluded that permitting
a civil case to go forward “relat[ing] entirely to the unofficial conduct of the
individual who happens to be the President” did not represent a per se
impermissible intrusion by the federal judiciary on executive power and that the
doctrine of separation of powers did not impose a categorical rule that all such
private actions must be stayed against the President while in office. Id. at 701,
14
Case 19-1540, Document 215, 12/03/2019, 2719370, Page15 of 59
705–06. At the same time, however, the Court recognized that it is insufficient
that a branch “not arrogate power to itself”: “the separation‐of‐powers doctrine
[also] requires that a branch not impair another in the performance of its constitutional
duties.” Id. at 701 (emphasis added) (quoting Loving v. United States, 517 U.S. 748,
757 (1996)); see also Nixon v. Admin’r of Gen. Servs., 433 U.S. 425, 443–45 (1977).
And as for the judiciary in the context of private litigation against a sitting
President, “[t]he high respect that is owed to the office of the Chief Executive,” the
Court recognized, “though not justifying a rule of categorical immunity, is a
matter that should inform the conduct of the entire proceeding, including the
timing and scope of discovery.” Id. at 707; see also Fitzgerald, 457 U.S. at 751–56
(noting that the “special nature of the President’s constitutional office and
functions,” id. at 756, and the “singular importance,” id. at 751, of her duties
require particular “deference and restraint,” id. at 753, in the conduct of litigation
involving the President).
The majority concludes that legislative subpoenas to third parties targeting
a President’s personal or financial information, however broad and tangentially
connected to any legislative purpose, do not seriously implicate separation of
powers on the theory that “any concern arising from the risk of distraction in the
15
Case 19-1540, Document 215, 12/03/2019, 2719370, Page16 of 59
performance of the [President’s] official duties is minimal,” Maj. Op. at 90, perhaps
less than that, id. at 103–05, at least as compared to the potential burden of standing
trial in a civil case while President, which Jones held is not categorically prohibited
by separation‐of‐powers concerns. 8 But this analysis is flawed in two key
respects.
First, the Jones Court concluded that the burden in that case—namely, a civil
suit against the President while in office—did not categorically constitute a
“constitutionally forbidden impairment of the Executive’s ability to perform its
constitutionally mandated functions” in light of the long history of judicial review
of executive action and of presidential amenability to judicial process. 520 U.S.
at 702; see also id. at 701–06. In assessing the separation‐of‐powers issue, the Court
The majority also relies on the fact that President Trump seeks a preliminary
injunction in his individual capacity, not his official capacity, and that the United States
has filed an amicus curiae brief rather than a motion to intervene in asserting its view that
this case presents “thorny constitutional questions involving separation of powers” and
that the district court’s order should be reversed. Brief of United States as Amicus
Curiae at 27; see Maj. Op. at 91 n.76. In Jones itself, however, President Clinton
proceeded in his individual capacity and the United States filed an amicus brief
addressing its separation‐of‐powers concerns. The Court nonetheless noted that “[t]he
representations made on behalf of the Executive Branch as to the potential impact” of a
rule permitting private litigation to proceed against a sitting President “merit . . .
respectful and deliberate consideration,” 520 U.S. at 689–90, and concluded, as already
observed, that as to any civil action regarding personal conduct permitted to proceed,
“the conduct of the entire proceeding, including the timing and scope of discovery,”
should be informed by respect for the Office of Chief Executive, id. at 707.
8
16
Case 19-1540, Document 215, 12/03/2019, 2719370, Page17 of 59
heavily weighed the pragmatic accommodation between the judiciary and the
executive demonstrated by longstanding interbranch practice. See id. at 704–05
(discussing historical practice and the manner in which the judiciary has
permissibly burdened the Executive Branch). It directed inferior courts that even
as it rejected a rule of categorical immunity, the President’s unique role in the
constitutional framework should inform the entire conduct of any civil action, id.
at 707, and that “the availability of sanctions” would “provide[] a significant
deterrent to litigation directed at the President in his unofficial capacity for
purposes of political gain or harassment,” id. at 708–09. The Jones Court was thus
solicitous of separation‐of‐powers concerns in the context of litigation over a
President’s personal conduct; moreover, it continued a long tradition of placing
“great weight” on historical practice in addressing questions “concern[ing] the
allocation of power between two . . . branches of Government,” NLRB v. Noel
Canning, 573 U.S. 513, 524 (2014) (quoting The Pocket Veto Case, 279 U.S. 655, 689
(1929)).9
The high value placed on historical practice “is neither new nor controversial.”
Noel Canning, 573 U.S. at 525. James Madison observed that a “regular course of
practice” could “liquidate & settle” constitutional meaning in the face of “difficulties and
differences of opinion” involved in the practice of government under the Constitution.
James Madison, Letter to Spencer Roane (Sept. 2, 1819), in 8 Writings of James Madison 450
(Gaillard Hunt ed., 1908)); see also Noel Canning, 573 U.S. at 525 (collecting cases stating
9
17
Case 19-1540, Document 215, 12/03/2019, 2719370, Page18 of 59
Here, the parties have not identified, and my own search has failed to
unearth, any previous example, in any previous Congress, of a standing or
permanent select committee of the House of Representatives or the Senate using
compulsory process to obtain documents containing a President’s personal
information from a third party in aid of legislation. Trump Br. at 14; Tr. of Oral
Arg. at 34:24–35:4. Historical practice instead suggests that, on the few past
occasions on which a President’s personal documents have been subpoenaed from
third parties, such requests have emanated either from a special committee
established and authorized to pursue a specific, limited investigation or from a
committee proceeding under the impeachment power. 10 It is possible that a
the relevance of past practice to separation‐of‐powers issues).
President Andrew Johnson had his personal bank records examined as part of
his impeachment, but those records appear to have been relevant because of personal
loans made to him by the Treasury Department. See Stephen W. Stathis, Executive
Cooperation: Presidential Recognition of the Investigative Authority of Congress and the Courts,
3 J.L. & Pol. 183, 219 (1986); see also Michael Les Benedict, The Impeachment of President
Andrew Johnson, 1867–68, in 1 Congress Investigates: A Critical and Documentary History 254,
264–68 (Roger A. Bruns et al. eds., rev. ed. 2011). President Clinton may have had some
financial information, or at the very least some financial information of then–First Lady
Hillary Clinton, examined by the Whitewater Special Committee, though it appears to
have been turned over voluntarily. See S. Rep. No. 104‐280, at 155–61 (1996). The
House and Senate Banking Committees also appear to have subpoenaed witnesses to
testify regarding Whitewater and the death of Vince Foster; however, they do not appear
to have subpoenaed the President’s personal financial information. See Stephen
Labaton, The Whitewater Affair: The Hearing; House Committee Told of Contacts Over
Whitewater, N.Y. Times, July 27, 1994, at A1 (describing testimony); Raymond W. Smock,
10
18
Case 19-1540, Document 215, 12/03/2019, 2719370, Page19 of 59
contrary example exists. But the historical precedent for the congressional
subpoenas here, in contrast to the judicial processes assessed in Jones, is sparse at
best, and perhaps nonexistent.11 And this paucity of historical practice alone is
The Whitewater Investigation and Impeachment of President Bill Clinton, 1992–98, in 2 Congress
Investigates: A Critical and Documentary History, supra, at 1041, 1044–45. President Nixon
voluntarily disclosed several years of tax returns to a House Committee; that same
Committee used statutory authority not at issue here to procure additional information
from the IRS. See S. Rep. No. 93‐768, at 1–3 (1974); Memorandum from Richard E. Neal,
Chairman, to the Members of the H. Comm. on Ways and Means 3 (July 25, 2019),
https://perma.cc/UYZ2‐QTCU. Other investigations do not appear to have involved
either subpoenas of the President’s personal financial information or subpoenas to third
parties to obtain documents concerning the President in a personal capacity. See
generally Stathis, supra.
Notably, the dearth of historical practice here may be partially attributable to
the fact that “[t]he authority to issue a subpoena was once delegated from the full House
to its committees very sparingly because the power appears long to have been deemed
too serious a matter for general delegation.” Todd David Peterson, Contempt of Congress
v. Executive Privilege, 14 U. Pa. J. Const. L. 77, 106 (2011) (internal quotation marks and
citation omitted). It appears that the House did not authorize standing committees to
issue subpoenas until 1975. Id. at 107. Moreover (and more generally), it should also
be noted that disputes between the two elected branches over congressional subpoenas
have historically been resolved through a process of direct negotiation and
accommodation between these two branches, undertaken outside the supervision of the
federal courts. See, e.g., Comm. on Judiciary, U.S. House of Representatives v. Miers, 558 F.
Supp. 2d 53, 56–57 (D.D.C. 2008) (noting that “negotiation and accommodation . . . most
often leads to resolution of disputes between the political branches” and “strongly
encourag[ing] the political branches to resume their discourse and negotiations in an
effort to resolve their differences constructively”). The majority rejects this approach
due to its view that this case does not involve separation of powers, Maj. Op at 69–73;
however, given the expressed willingness of the parties to negotiate and my view that
separation‐of‐powers concerns are present here, the traditional practice of further
negotiation is a viable resolution.
11
19
Case 19-1540, Document 215, 12/03/2019, 2719370, Page20 of 59
reason for courts to pause in assessing this dispute between a President and two
House committees.12
The second flaw in the majority’s analysis lies in its assumption that third‐
party subpoenas of this sort pose, at best, “minimal” risk of distraction to this and
future Presidents. Maj. Op. at 90. Contrary to the majority’s suggestion, it is not
at all difficult to conceive how standing committees exercising the authority to
issue third‐party subpoenas in aid of legislation might significantly burden
presidents with myriad inquiries into their business, personal, and family affairs.
See Watkins, 354 U.S. at 205 (recognizing potential for “ruthless exposure of private
lives” by committees seeking information “neither desired by the Congress nor
useful to it”); cf. Jones, 520 U.S. at 701–02 (considering the likelihood that frivolous
civil litigation against the President could overly burden the Executive Branch).
Jones relied on the relative rarity of civil litigation against past presidents to
discount concerns of distraction, see 520 U.S. at 702, but the subjects on which
This Court’s recent decision in Trump v. Vance, 941 F.3d 631 (2d. Cir. 2019), is not
to the contrary. The Vance panel explicitly relied on the “long‐settled” amenability of
presidents to judicial process, and in particular to subpoenas issued as part of a criminal
prosecution, to inform its holding that the state grand jury subpoena to a third‐party
custodian of the President’s tax returns at issue in that case was lawful. See id. at 640
(discussing the historical practice of ordering presidents to comply with grand jury
subpoenas). Here, there is no such longstanding practice, and the subpoenas in
question were not issued by a grand jury as part of a criminal investigation.
12
20
Case 19-1540, Document 215, 12/03/2019, 2719370, Page21 of 59
legislation might be had are vast.13 And the risk of undue distraction from ill‐
conceived inquiries might be particularly acute today, in an era in which (as the
Supreme Court and individual Justices have repeatedly acknowledged) digital
technologies have lodged an increasingly large fraction of even our most intimate
information in third‐party hands. See, e.g., Riley v. California, 573 U.S. 373, 395
(2014) (discussing how “Internet search and browsing history” can “reveal an
individual’s private interests or concerns”); Carpenter v. United States, 138 S. Ct.
2206, 2261 (2018) (acknowledging “powerful private companies” collecting “vast
quantities of data about the lives of ordinary Americans”) (Alito, J., dissenting);
United States v. Jones, 565 U.S. 400, 417 (2012) (noting that in the digital age, “people
13
To be clear, while civil litigation against sitting presidents is unusual, presidents
are routinely the subjects of congressional investigation while in office—as they must be,
and for appropriate reasons. But there is no substantial historical precedent for the use
of subpoena power to obtain a President’s personal information from a third party in aid
of legislation. And as to such subpoenas, there is no analogue for the possibility of
sanctions in the civil litigation context, which the Jones Court relied on as “provid[ing] a
significant deterrent to litigation directed at the President in his unofficial capacity for
purposes of political gain or harassment.” 520 U.S. at 708–09. Nor do established rules
of procedure provide a mechanism for narrowing congressional subpoenas so as to avoid
“embarrassment, oppression, or undue burden.” Fed. R. Civ. P. 26(c)(1). Historically,
in those few instances in which investigators have sought a President’s personal
documents, Congress has instead typically proceeded pursuant to the political checks
inherent in the invocation of impeachment authority or the narrow authorization
afforded to a special committee.
21
Case 19-1540, Document 215, 12/03/2019, 2719370, Page22 of 59
reveal a great deal of information about themselves to third parties in the course
of carrying out mundane tasks”) (Sotomayor, J., concurring).
To be clear, this is not to suggest that a President is immune from legislative
subpoenas into personal matters—not at all. But as the D.C. Circuit recognized
in Trump v. Mazars (while concluding that the House Committee on Oversight and
Reform possessed authority to issue a legislative subpoena to President Trump’s
accounting firm), “separation‐of‐powers concerns still linger in the air” with
regard to such subpoenas. Trump v. Mazars USA, LLP, 940 F.3d 710, 726 (D.C. Cir.
2019). And in such a circumstance, there is reason to conclude that courts must
not only undertake the “arduous and delicate task” of “[a]ccommodat[ing] . . . the
congressional need for particular information with the individual and personal
interest in privacy,” Maj. Op. at 51 (quoting Watkins, 354 U.S. at 198). They must
also take on the equally sensitive task of ensuring that Congress, in seeking the
President’s personal information in aid of legislation, has employed “procedures
which prevent the separation of power from responsibility,” Watkins, 354 U.S. at
215 (discussing such procedures in the context of a threat to individual rights from
congressional investigations), and which ensure due consideration to the
separation‐of‐powers concerns that the Supreme Court identified and deemed
22
Case 19-1540, Document 215, 12/03/2019, 2719370, Page23 of 59
essential for judicial respect in Jones. See Jones, 520 U.S. at 707 (noting that “high
respect that is owed to the office of the Chief Executive,” while not mandating
categorical immunity from suit for private conduct while in office, should “inform
the conduct of the entire proceeding, including the timing and scope of
discovery”); Cheney, 542 U.S. at 385 (noting that President’s “constitutional
responsibilities and status [are] factors counseling judicial deference and restraint”
in conduct of litigation) (quoting Fitzgerald, 457 U.S. at 753 (alteration in Cheney)).
B
These subpoenas are deeply problematic when considered against the
backdrop of these separation‐of‐powers concerns. In fact, this much is evident
from even cursory consideration of the differences between the present case and
Mazars, the only other precedent directly addressing a legislative subpoena served
on a third party and seeking a President’s personal financial information.14 In
Mazars, the D.C. Circuit recently upheld a legislative subpoena directed at the
As noted at the outset, see supra page 5, the parties are unable to cite any
Congress before this one in which a standing committee of the House of Representatives
has issued a third‐party subpoena for documents targeting a President’s personal
information solely in aid of legislation. The practice appears to have begun with the
committees of this House of Representatives, which has issued such subpoenas
repeatedly, thus raising the separation‐of‐powers concerns discussed herein.
14
23
Case 19-1540, Document 215, 12/03/2019, 2719370, Page24 of 59
President’s accounting firm, concluding that it had properly issued in connection
with the consideration of changes to laws relating to financial disclosures required
of Presidents.15 Mazars, 940 F.3d at 748. At the same time, the Mazars Court
pointedly suggested that the articulation of just any rationale for concluding that
a sitting President’s personal information might inform a committee in
considering potential legislation is not enough to state a valid legislative purpose:
Just as a congressional committee could not subpoena
the President’s high school transcripts in service of an
investigation into K‐12 education, nor subpoena his
medical records as part of an investigation into public
health, it may not subpoena his financial information
except to facilitate an investigation into presidential
finances.
Judge Rao dissented, concluding that even assuming the Committee on
Oversight and Reform had a legislative purpose, it had also asserted an intent to
determine “whether the President broke the law,” an inquiry that “must be pursued
through impeachment,” and not via Congress’s authority to investigate for legislative
purposes. Mazars, 940 F.3d at 748 (Rao, J., dissenting). In the instant case, given the
need for remand here, I need not now determine whether the House Committees have
avowed such an intent, so I have no occasion to consider the arguments raised in Judge
Rao’s thorough analysis. However, it is worth noting that nowhere in the Mazars
majority or Judge Rao’s extensive discussion of historical practice, id. at 718–24 (majority
opinion), 757–67 (Rao, J., dissenting), is there any hint of a prior occasion on which a
standing or permanent select committee has used compulsory process to obtain
documents targeting a President’s personal information from a third party justified solely
on the basis of future legislation.
15
24
Case 19-1540, Document 215, 12/03/2019, 2719370, Page25 of 59
Id. at 733. Key to the result in Mazars, then (and assuming, arguendo, that it was
correctly decided) was the majority’s conclusion that there was “no inherent
constitutional flaw in laws requiring Presidents to publicly disclose financial
information” and that the subpoena on its face thus properly sought relevant
information “about a subject on which legislation may be had.” Id. at 737
(quoting Eastland, 421 U.S. at 508).
This case is significantly different, at least as to the subpoenas issued by the
Committee on Financial Services. This Committee seeks a universe of financial
records sufficient to reconstruct over a decade of the President’s business and
personal affairs, not in connection with the consideration of legislation involving
the Chief Executive, but because the President, his family, and his businesses
present a “useful case study,” according to the Committee, for an inquiry into the
lending practices of institutions such as Deutsche Bank and Capital One.16 District
Court Doc. No. 51 at 25. More specifically, the Committee is investigating
The Capital One subpoena, moreover, seeks the President’s personal and
business financial records starting from the exact date on which he became the
Republican nominee for President—an unusual date, to be sure, for specifying the precise
moment at which his banking records became a useful point of inquiry into the possibility
of tightening up the regulation of lending practices with potentially “broad effects on the
national economy.” District Court Doc. No. 51 at 25.
16
25
Case 19-1540, Document 215, 12/03/2019, 2719370, Page26 of 59
“whether existing policies and programs at financial institutions are adequate to
ensure the safety and soundness of lending practices and the prevention of loan
fraud,” id. at 12, as well as “industry‐wide compliance with banking statutes and
regulations, particularly anti‐money laundering policies,” id. at 13. The
Committee urges that “[b]ecause of his prominence, much is already known about
Mr. Trump, his family, and his business, and this public record establishes that
they serve as a useful case study for the broader problems” under its
consideration. 17 Id. at 25. The majority endorses this statement of legislative
purpose and intimates (albeit with no evidence in the record before us) that past
transactions between Deutsche Bank and the President in his pre‐presidential
business life may have violated banking regulations and that “no other bank
would extend credit” to President Trump. Maj. Op. at 73 n.67, 74.
To be sure, legislative subpoenas issue not when all is known, but on the
reasonable theory that “[a] legislative body cannot legislate wisely or effectively”
The House Financial Services Committee asserts that the subpoenas’ objective
can be derived in part from House Resolution 206, which affirms that the House
“supports efforts to close loopholes that allow corruption, terrorism, and money
laundering to infiltrate our country’s financial system.” H.R. Res. 206, 116th Cong.
(2019). House Resolution 206, however, does not materially aid in defining more clearly
the reasons for the Committee’s “case study” approach, as it does not call for a
congressional investigation, much less one by a designated committee, nor does it
reference the President and his family.
17
26
Case 19-1540, Document 215, 12/03/2019, 2719370, Page27 of 59
without obtaining “information respecting the conditions which the legislation is
intended to affect or change.” Eastland, 421 U.S. at 504 (quoting McGrain v.
Daugherty, 273 U.S. 135, 175 (1927) (alteration in Eastland)). But the rationale
proffered for these subpoenas of the House Financial Services Committee falls far
short of demonstrating a clear reason why a congressional investigation aimed
generally at closing regulatory loopholes in the banking system need focus on over
a decade of financial information regarding this President, his family, and his
business affairs. 18 Nor does the proffered rationale reveal how the broad
purposes pursued by the Committee are consistent with the granular detail that
these subpoenas seek. See Watkins, 354 U.S. at 204 (noting the troubling tendency
of some legislative investigations to “probe for a depth of detail . . . removed from
any basis of legislative action” and to “turn their attention to the past to collect
minutiae on remote topics”).
Thus, the majority references the fact that Deutsche Bank “has been fined in
connection with a $10 billion money laundering scheme.” Maj. Op. at 73 n.67. But the
record is devoid of any claim, much less any evidence, that this fine had anything at all to
do with the President, his children, his business organizations, or his business associates,
all of whom will be irreparably harmed by the majority’s endorsement of the “case study”
approach of the House Financial Services Committee.
18
27
Case 19-1540, Document 215, 12/03/2019, 2719370, Page28 of 59
This is a reason for pause. As suggested by Judge Katsas in his dissent
from the denial of rehearing in banc in Mazars, the “uncompromising extension of
McGrain v. Daugherty” to this new context raises the serious question whether
future Presidents will be routinely subject to the distraction of third‐party
subpoenas emanating from standing committees in aid of legislation—a practice
for which there is scant historical precedent, as already discussed. Trump v.
Mazars USA, LLP, No. 19‐5142, 2019 WL 5991603, at *1 (D.C. Cir. Nov. 13, 2019)
(Katsas, J., dissenting from the denial of rehearing en banc). Some case study
rationale (in this instance, to learn whether regulators were adequately equipped
to scrutinize Deutsche Bank’s and Capital One’s lending practices in relation to the
President before he obtained the Office of Chief Executive) will always be present.
But the regular issuance of third‐party legislative subpoenas by single committees
of one House of Congress targeting a President’s personal information would be
something new, potentially impairing public perceptions of the legislative branch
by fueling perceptions that standing committees are engaged, not in legislating,
but in opposition research.19 More relevant here, such investigative practices by
Such subpoenas, moreover, will inevitably result, as here, in recourse to the
courts, potentially embroiling them, as well, in political battles between committees of
Congress and the President.
19
28
Case 19-1540, Document 215, 12/03/2019, 2719370, Page29 of 59
Congress, undertaken “more casually and less responsibly” than is the
constitutional ideal, see Rumely, 345 U.S. at 46, pose a serious threat to “presidential
autonomy and independence,” Mazars, 2019 WL 5991603, at *1 (Katsas, J.,
dissenting from the denial of rehearing en banc). And this is a substantial concern
in our constitutional scheme, which relies on the proposition that the occupant of
the Office of Chief Executive is positioned to “‘deal fearlessly and impartially with’
[its] duties,” even as Presidents may be “easily identifiable target[s]” of legal
process, personally vulnerable by virtue of the “visibility of [the] office and the
effect of [their] actions on countless people.” Fitzgerald, 457 U.S. at 752–53
(quoting Ferri v. Ackerman, 444 U.S. 193, 203 (1979)).
To be sure, the third subpoena to Deutsche Bank, which is identical to the
Deutsche Bank subpoena issued by the Committee on Financial Services, emanates
from the Permanent Select Committee on Intelligence and is more closely linked
to the consideration of legislation related to the Office of the Chief Executive and
to this President’s affairs, as a recent candidate. 20 The majority is correct,
As the majority states, the Chair of the Intelligence Committee has publicly
affirmed that the Committee is investigating matters related to interference by the
Russian government in the U.S. political process and that the information sought from
Deutsche Bank will inform legislative proposals to protect this process from foreign
influence. Maj. Op at 62–64. The House Intelligence Committee, moreover, has an
oversight function to which its subpoena could conceivably relate. At the same time,
20
29
Case 19-1540, Document 215, 12/03/2019, 2719370, Page30 of 59
moreover, that once presented with adequate evidence of legislative authorization
and purposes, it is not the province of courts to inquire into legislators’ motives,
see Maj. Op. at 50–51, and that “motives alone would not vitiate an investigation
which had been instituted by a House of Congress if that assembly’s legislative
purpose is being served.” Watkins, 354 U.S. at 200.
At the same time, as the majority also affirms, the record must provide
“sufficient evidence of legislative authorization and purposes to enable
meaningful judicial review.” Maj. Op. at 55. And this is particularly the case
when a congressional investigation even potentially trenches upon constitutional
however, no House resolution appears specifically to reference this investigation, at least
as it relates to efforts to seek the President’s financial information, nor is such a legislative
purpose easy to square with the extraordinary breadth of the Deutsche Bank subpoenas.
The Chair, moreover, has also affirmed that the Committee’s investigation is in
furtherance of Congress’s duty to “ensure that U.S. officials—including the President—
are serving the national interest and, if not, are held accountable.” Press Release,
Permanent Select Comm. on Intelligence, Chairman Schiff Statement on House
Intelligence Committee Investigation (Feb. 6, 2019), bit.ly/2UMzwTE. The Plaintiffs
argue that the subpoena is thus not in furtherance of legislative purposes, but represents
an effort by the Committee to itself conduct intelligence and law enforcement activities.
Trump Br. at 35–36. Indeed, at oral argument, the Committees’ lawyer appeared
explicitly to equate these subpoenas to those issued in connection with federal criminal
investigations. Tr. of Oral Arg. at 59:14–60:2. While I do not decide whether the
Intelligence Committee has affirmatively avowed an improper purpose, the amorphous
nature of the Committee’s legislative purpose would be clarified by my proposed
remand, as would the connection between this purpose and the particular disclosures
that are sought.
30
Case 19-1540, Document 215, 12/03/2019, 2719370, Page31 of 59
limits on Congress’s investigative power. See Rumely, 345 U.S. at 46 (noting that
such limits should be identified by courts only after “Congress has . . .
unequivocally authoriz[ed] an inquiry of dubious limits”). Indeed, in such
circumstances, the Supreme Court has made clear that courts are to look to the
“instructions to an investigating committee,” as “embodied in the authorizing
resolution,” to ascertain whether the legislative assembly has “assay[ed] the
relative necessity of specific disclosures.” Watkins, 354 U.S. at 201, 206.
Considered in light of the separation‐of‐powers concerns that persist with regard
to these subpoenas, the Plaintiffs have raised a serious question on this front as
well.
As to both the House Financial Services and Intelligence Committee
subpoenas, there is an open question as to whether these subpoenas have been
authorized by the House of Representatives in a manner permitting this Court to
determine whether they are “in furtherance of . . . a legitimate task of the
Congress.” Watkins, 354 U.S. at 187. As the Watkins Court explained, “[t]he
theory of a committee inquiry is that the committee members are serving as the
representatives of the parent assembly in collecting information for a legislative
purpose” and that “the House or Senate shall have instructed the committee
31
Case 19-1540, Document 215, 12/03/2019, 2719370, Page32 of 59
members on what they are to do with the power delegated to them.” Id. at 200–
01. The majority acknowledges Watkins’s requirement that an authorizing
resolution “spell out [an investigating committee’s] jurisdiction and purpose with
sufficient particularity” as to ensure that “compulsory process is used only in
furtherance of a legislative purpose.” Id. at 201; see Maj. Op. at 51, 79–80.
Critically, moreover, the majority itself recognizes that “[i]t is not clear whether
this passage can be satisfied” with regard to these subpoenas by the principal
instruction in place here, at the time the subpoenas issued: namely, the
instruction “that the House gives to a committee pursuant to a House rule defining
a standing committee’s continuing jurisdiction.” Maj. Op. at 52–53.
The majority treats House Resolution 507 as the cure‐all solution to this key
uncertainty, rejecting the Plaintiffs’ argument that it is not properly considered on
the subject of legislative authorization and purposes because it issued after the
subpoenas themselves.21 But House Resolution 507 falls far short of a specific
The majority’s support for this conclusion derives solely from cases discussing,
in the contempt prosecution context, what evidence may be considered in evaluating
whether a question posed to a witness before a congressional committee was pertinent to
an investigation’s inquiry. See Watkins, 354 U.S. at 201–02; Rumely, 345 U.S. at 48; Shelton
v. United States, 327 F.2d 601, 607 (D.C. Cir. 1963); see also Maj. Op. at 54–58. This issue
is distinct from the threshold question of whether a committee is adequately authorized,
so that the majority must necessarily reason by analogy, and its conclusion is far from
inevitable, particularly in the context of third‐party subpoenas aimed at a President’s
21
32
Case 19-1540, Document 215, 12/03/2019, 2719370, Page33 of 59
“authorizing resolution” issued to make clear that a designated committee is to
undertake an investigation on a particular subject within its domain. To be sure,
McGrain found sufficient a resolution that did not “in terms avow that it [was]
intended to be in aid of legislation,” on the theory that “the subject‐matter was
such that [a] presumption should be indulged” that legislating “was the real
object.” 273 U.S. at 177–78. But in a context like this, presenting serious
constitutional concerns, courts “have adopted the policy of construing . . .
resolutions . . . narrowly, in order to obviate the necessity of passing on serious
constitutional questions.” Tobin, 306 F.2d at 274–75. And this resolution on its
face discusses none of the subpoenas here, nor even the work of the committees
from which they issued. Instead, House Resolution 507 authorizes any subpoena,
by any standing or permanent select committee, already issued or in the future to
be issued, so long as it concerns the President, his family, or his business entities
and organizations:
personal information, where the President must be able efficiently (and without undue
distraction) to determine what, if any, steps she should take, either to assist the inquiry
or, as here, to litigate. I need not address this question, however, because, even
assuming that Resolution 507 is properly considered, a serious question remains as to
whether it constitutes what the majority acknowledges is required: “sufficient evidence
of legislative authorization and purposes to enable meaningful judicial review.” Maj.
Op. at 55.
33
Case 19-1540, Document 215, 12/03/2019, 2719370, Page34 of 59
Resolved, That the House of Representatives ratifies and
affirms all current and future investigations, as well as all subpoenas
previously issued or to be issued in the future, by any standing or
permanent select committee of the House, pursuant to its jurisdiction
as established by the Constitution of the United States and rules X and
XI of the Rules of the House of Representatives, concerning or issued
directly or indirectly to —
(1) the President in his personal or official capacity;
(2) his immediate family, business entities, or
organizations;
. . .
(9) any third party seeking information involving,
referring, or related to any individual or entity described in
paragraphs (1) through (7).
H.R. Res. 507, 116th Cong. (2019); see also H.R. Res. 509, 116th Cong. § 3 (2019)
(“House Resolution 507 is hereby adopted”).
By purporting to authorize third‐party subpoenas for any and all past and
future investigations into the President’s personal and official business, Resolution
507 would appear to run directly into the primary concern in Watkins that
“[b]roadly drafted and loosely worded” resolutions can “leave tremendous
latitude to the discretion of investigators,” 354 U.S. at 201, and thus permit
committees “in essence, to define [their] own authority,” id. at 205. As Watkins
emphasized, “[a]n essential premise” underlying the investigatory powers of a
congressional committee to compel the production of documents or attendance by
an individual “is that the House or Senate shall have instructed the committee
34
Case 19-1540, Document 215, 12/03/2019, 2719370, Page35 of 59
members on what they are to do with the power delegated to them.” Id. at 201.
Absent that instruction, such subpoenas defy judicial review, the Watkins Court
understood, because “it is impossible . . . to declare that [a committee] has ranged
beyond the area committed to it by its parent assembly.” Id. at 205.
To be clear, Watkins addressed this problem in the context of a House
proceeding implicating a private citizen’s constitutional liberties, and not
separation of powers. But its caution is still relevant: that “excessively broad
charter[s]” to investigating committees make it difficult, if not impossible, for
courts “to ascertain whether any legislative purpose justifies the disclosures
sought and, if so, the importance of that information to the Congress in furtherance
of its legislative function.” Id. at 205–06. With respect, the majority thus errs in
dismissing the Department of Justice’s concern that the blank‐check approach
adopted here to authorizing third‐party subpoenas seeking personal information
about the President and his family represents “a failure of the House to exercise
‘preliminary control of the Committee[s],’” see Brief of United States as Amicus
Curiae at 19 (quoting Watkins, 354 U.S. at 203)—a failure which not only throws
into question the adequacy of authorization in this case, but which also raises
significant issues for the future regarding interbranch balance and the ability of
35
Case 19-1540, Document 215, 12/03/2019, 2719370, Page36 of 59
this and future Presidents to perform their duties without undue distraction, id. at
5–7; see Jones, 520 U.S. at 690 (noting that “representations made on behalf of the
Executive Branch as to the potential impact” of inquiries on the Office of the
President “merit our respectful and deliberate consideration”). 22 In short,
Resolution 507 itself, given its retrospective and prospective nature, and its
purported authorization of any and all third‐party committee subpoenas seeking
not only official, but personal information about the President, his family, and his
businesses, presents a serious question as to whether the House has discharged its
The Department of Justice argues that a clear statement rule should apply to the
authorization of legislative subpoenas seeking a President’s personal information. Brief
of United States as Amicus Curiae at 10. The majority dismisses this argument, noting
that neither Franklin v. Massachusetts, 505 U.S. 788, nor Armstrong v. Bush, 924 F.2d 282
(D.C. Cir. 1991), on which the Department relies, concern congressional subpoenas, but
statutes “claimed to limit presidential power.” Maj. Op. at 89. But Rumely makes clear
that the duty of constitutional avoidance (implemented, in part, through mechanisms
such as clear statement rules) “is even more applicable” in the context of congressional
investigations than in the interpretation of statutes. 345 U.S. at 46. It also affirms that
“[w]henever constitutional limits upon the investigative power of Congress have to be
drawn . . . , it ought only to be done after Congress has demonstrated its full awareness
of what is at stake by unequivocally authorizing an inquiry of dubious limits.” Id. In
short, while I need not at this time reach the question, the Department’s clear statement
argument merits serious consideration, as does its assertion that the House’s “blank‐
check” approach to use of compulsory process directed at the President, his family, and
his businesses runs afoul of Watkins’s caution that “[a] measure of added care on the part
of the House and the Senate in authorizing the use of compulsory process” would help
“prevent the separation of power from responsibility.” 354 U.S. at 215.
22
36
Case 19-1540, Document 215, 12/03/2019, 2719370, Page37 of 59
“responsibility . . . in the first instance, to insure that compulsory process is used
only in furtherance of a legislative purpose.” Watkins, 354 U.S. at 201.
II
These third‐party legislative subpoenas thus raise serious questions on the
merits, implicating substantial separation‐of‐powers concerns. In such a context,
Rumely’s caution kicks in, which “counsel[s] abstention from adjudication unless
no choice is left.” 345 U.S. at 46. The majority disagrees, asserting that even
assuming serious questions regarding the separation of powers have been raised,
affirmance here is still required because our “serious questions” approach to
whether a preliminary injunction should issue is unavailable in the context of these
third‐party legislative subpoenas. 23 I have already outlined my disagreement
The majority also argues that any serious questions presented here “are
properly rejected at this stage of the litigation” because they “involve solely issues of
law.” Maj. Op. at 101. I disagree. As an initial matter, our case law has recognized
that, in appropriate circumstances, purely legal issues can present sufficiently serious
questions to warrant a preliminary injunction. See, e.g., Haitian Centers Council, Inc. v.
McNary, 969 F.2d 1326, 1339–40 (2d Cir. 1993) (finding sufficiently serious questions
going to the merits based on the novel questions of law presented by plaintiffs’ claims),
judgment vacated as moot by Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 918 (1993); see also,
e.g., 11A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and
Procedure § 2948.3 (3d ed.) (Westlaw) (database updated August 2019) (referring to “the
existence of a factual conflict, or of difficult questions of law,” as components of the merits
showing in the preliminary injunction context (emphasis added)). Moreover, the
majority itself is remanding for some development of the factual record. As set forth
herein, I conclude that the majority’s limited remand is inadequate, and that the record
23
37
Case 19-1540, Document 215, 12/03/2019, 2719370, Page38 of 59
with the majority’s determination that “this case does not concern separation of
powers,” Maj. Op. at 89, and that the questions raised, even if “serious in at least
some sense, lack merit,” id. at 101. I also disagree as to the supposed
unavailability of our traditional preliminary injunction approach. Indeed, I
conclude, with respect, that the majority badly errs in deciding that this approach
is unavailable in the sensitive context of challenges to congressional subpoenas.
As the Supreme Court made clear in Winter v. Natural Resources Defense
Council, Inc., “[a] plaintiff seeking a preliminary injunction must establish that he
is likely to succeed on the merits, that he is likely to suffer irreparable harm in the
absence of preliminary relief, that the balance of equities tips in his favor, and that
an injunction is in the public interest.” 555 U.S. 7, 20 (2008). The majority
acknowledges that, as to the required merits showing, we have repeatedly said in
this Circuit that “district courts may grant a preliminary injunction where a
plaintiff . . . meets either of two standards: ‘(a) a likelihood of success on the
merits, or (b) sufficiently serious questions going to the merits to make them a fair
ground for litigation.’” Maj. Op. at 11–12 (quoting Kelly v. Honeywell Int’l, Inc.,
933 F.3d 173, 184 (2d Cir. 2019)). When a plaintiff has demonstrated only “serious
needs further factual development before the legal issues here can be adequately assessed.
38
Case 19-1540, Document 215, 12/03/2019, 2719370, Page39 of 59
questions” as to the merits, however, the plaintiff has a higher burden as to the
third element: he must show that the balance of hardships tips decidedly in his
favor. See Kelly, 933 F.3d at 184; Maj. Op. at 11–12. The majority also
acknowledges that we have reaffirmed our traditional approach in the wake of the
Supreme Court’s decision in Winter. See Citigroup, 598 F.3d at 38 (“hold[ing] that
our venerable standard for assessing a movant’s probability of success on the
merits remains valid”). 24 Irreparable harm is not in question in this case,
moreover, because, inter alia, the Plaintiffs have an interest in keeping their
banking records private from Congress and neither House committee will commit
to treating any portion of the voluminous personal and business records that they
seek as confidential. J.A. at 122–23. In such circumstances, the majority and I
are in agreement that compliance with these subpoenas will cause irreparable
Citigroup carefully assessed Winter’s import and concluded that our traditional
approach is wholly consistent with that precedent and is properly retained, given “[t]he
value of this circuit’s approach to assessing the merits of a claim at the preliminary
injunction stage,” which “lies in its flexibility in the face of varying factual scenarios and
the greater uncertainties inherent at the outset of particularly complex litigation.”
Citigroup, 598 F.3d at 35. Moreover, Citigroup made clear that, under either the “serious
questions” or the “likelihood of success” formulation, courts in this Circuit consider all
four elements articulated by the Supreme Court in Winter. See id. at 34, 38 (citing Winter,
555 U.S. at 20).
24
39
Case 19-1540, Document 215, 12/03/2019, 2719370, Page40 of 59
harm to the President, his family, his businesses, and his business associates.
Maj. Op. at 13–14.
The majority asserts that a preliminary injunction is nonetheless unavailable
based on our “serious questions” formulation of the merits inquiry because of the
so‐called “government action exception” to this formulation, as expressed by this
Court’s decision in Plaza Health Laboratories, Inc. v. Perales, 878 F.2d 577, 580 (2d
Cir. 1989). I disagree. To be sure, our case law has recognized three narrowly
defined situations in which a movant cannot obtain a preliminary injunction under
the “serious questions” formulation. See id.; Tom Doherty Assocs., Inc. v. Saban
Entm’t, Inc., 60 F.3d 27, 33–34 (2d Cir. 1995); Abdul Wali v. Coughlin, 754 F.2d 1015,
1025 (2d Cir. 1985). But Plaza Health, on which the majority relies, is not
applicable.
To explain my conclusion requires a step back from our traditional
formulation, to set forth why this Circuit was correct to reaffirm our serious
question approach—and, indeed, why we err today in expanding a formulaic
exception to it. While sometimes styled in our case law as its own “standard,”
see, e.g., Otoe‐Missouria Tribe of Indians v. N.Y. State Dep’t of Fin. Servs., 769 F.3d 105,
110 (2d Cir. 2014), the “sufficiently serious questions, plus a balance of hardships
40
Case 19-1540, Document 215, 12/03/2019, 2719370, Page41 of 59
tipping decidedly in favor of the moving party” approach is not actually a separate
test at all, but rather a way of articulating one point on a single sliding scale that
balances likelihood of success against hardship in determining whether a
preliminary injunction should issue. See 11A Charles Alan Wright, Arthur R.
Miller & Mary Kay Kane, Federal Practice and Procedure § 2948.3 (3d ed.) (Westlaw)
(database updated August 2019) (hereinafter “Wright & Miller”) (referring to the
Second Circuit’s “serious questions” formulation as “[p]robably the most often‐
quoted statement” of the sliding scale principle). Likelihood of success, while of
“particular importance” in this inquiry, is not determinative, but must be
considered and balanced with the relative hardship each side is likely to face from
the determination whether an injunction issues, with the so‐called “serious
questions” standard emerging as simply one point on the sliding scale at which an
injunction may be warranted.25 Id. This flexible approach is particularly well‐
suited to the preliminary injunction context, where courts act pursuant to
As Judge Frank articulated decades ago, when “the balance of hardships tips
decidedly toward plaintiff,” it should “ordinarily be enough that the plaintiff has raised
questions going to the merits so serious, substantial, difficult and doubtful, as to make
them a fair ground for litigation and thus for more deliberate investigation.” Hamilton
Watch Co. v. Benrus Watch Co., 206 F.2d 738, 740 (2d Cir. 1953).
25
41
Case 19-1540, Document 215, 12/03/2019, 2719370, Page42 of 59
equitable principles.26 See, e.g., Holland v. Florida, 560 U.S. 631, 649–50 (2010) (“In
emphasizing the need for flexibility . . . we have followed a tradition in which
courts of equity have sought to relieve hardships which, from time to time, arise
from a hard and fast adherence to more absolute legal rules, which, if strictly
applied, threaten the evils of archaic rigidity.” (internal quotation marks,
alterations, and citations omitted)).
Against this backdrop, our so‐called “exceptions” to the serious questions
formulation are best understood not in prescriptive terms, but as the articulation
of principles guiding the application of the sliding scale calculus in particular
scenarios. As relevant here, the Plaza Health “exception” thus reflects a
considered judgment, drawing on equitable ideas, that “[w]here the moving party
seeks to stay government action taken in the public interest pursuant to a statutory
or regulatory scheme,” the serious questions formulation should be generally
unavailable precisely because the balance of hardships is so unlikely to tip
Indeed, confining preliminary injunctions to circumstances in which a plaintiff
has shown there is no difficult question of law that could ultimately go against him would
“deprive the remedy of much of its utility.” Wright & Miller § 2948.3; see also Citigroup,
598 F.3d at 35 (noting that “[p]reliminary injunctions should not be mechanically
confined to cases that are simple or easy,” as happens when the likelihood‐of‐success
standard is formulaically employed).
26
42
Case 19-1540, Document 215, 12/03/2019, 2719370, Page43 of 59
decidedly in that party’s favor. Able v. United States, 44 F.3d 128, 131 (2d Cir. 1995)
(quoting Plaza Health, 878 F.2d at 580). In issuing a preliminary injunction based
on the conclusion that it does, a court impermissibly “substitute[s] its own
determination of the public interest” for the one reflected in the statutory or
regulatory scheme. Id. at 132.
Accordingly, where government action has been fairly characterized as
taken pursuant to a statutory or regulatory scheme, we have generally applied the
likelihood‐of‐success standard. See Citigroup, 598 F.3d at 35 n.4 (articulating the
exception as limited to situations in which “a moving party seeks to stay
government action taken in the public interest pursuant to a statutory or
regulatory scheme”). And where movants have sought preliminarily to enjoin
government action pursuant to a federal statutory or regulatory scheme, we have
explained that in the context of such action, “developed through presumptively
reasoned democratic processes” and resulting from “the full play of the
democratic process involving both the legislative and executive branches,” it is
difficult to envision any circumstance in which a movant could demonstrate that
the balance of hardships tips decidedly in his favor. Able, 44 F.3d at 131.
43
Case 19-1540, Document 215, 12/03/2019, 2719370, Page44 of 59
The majority argues that the Plaza Health exception sweeps more broadly,
relying for this proposition on cases involving action taken by state and local
governments. 27 See Maj. Op. at 15–16. While certain of these cases did not
analyze why the Plaza Health exception was applicable, and appear simply to have
assumed that the government action in question was taken pursuant to a statutory
or regulatory scheme, see, e.g., Cent. Rabbinical Cong., 763 F.3d at 192; Monserrate,
599 F.3d at 154, those that did engage with this analysis explicitly identified a
statutory or regulatory scheme and accordingly concluded that the presumptive
See, e.g., Cent. Rabbinical Cong. of U.S. & Canada v. N.Y.C. Dep’t of Health & Mental
Hygiene, 763 F.3d 183, 192 (2d Cir. 2014) (likelihood‐of‐success standard applied to
preliminary injunction sought by religious organizations against a city ordinance based
on the court’s conclusion, without further analysis, that the ordinance constituted
“government action taken in the public interest pursuant to a statutory or regulatory
scheme” (citation omitted)); Monserrate v. N.Y. State Senate, 599 F.3d 148, 154 (2d Cir. 2010)
(same, as to a preliminary injunction seeking to unwind the expulsion of a state senator);
NAACP v. Town of East Haven, 70 F.3d 219, 223 (2d Cir. 1995) (likelihood‐of‐success
standard applied to a preliminary injunction seeking to enjoin a town from hiring police
officers or firefighters, based on the court’s conclusion that the town acted “in the public
interest” and “pursuant to established municipal regulations and state civil service
laws”); N.Y. Urban League, Inc. v. State of New York, 71 F.3d 1031, 1036 n.7 (2d Cir. 1995)
(applying likelihood‐of‐success standard to a preliminary injunction seeking to bar
transit authority from implementing a proposed fare increase on the basis that the action
in question “was to be implemented in accordance with the special powers” of the transit
authority board as set forth in a state statute); see also Molloy v. Metro. Transp. Auth., 94
F.3d 808, 811 (2d Cir. 1996) (relying on New York Urban League in applying the likelihood‐
of‐success standard to a preliminary injunction sought against transit authority’s
implementation of a staff reduction plan).
27
44
Case 19-1540, Document 215, 12/03/2019, 2719370, Page45 of 59
public interest weighed against the movant, see, e.g., NAACP, 70 F.3d at 223; see
also, e.g., Otoe‐Missouria Tribe of Indians, 769 F.3d at 110 (determining that New
York’s ban on certain loans was “a paradigmatic example of governmental action
taken in the public interest, one that vindicated proven policies implemented
through legislation or regulations” and therefore applying the likelihood‐of‐
success standard (internal quotation marks and citations omitted)).28
Where, by contrast, government action has not been taken pursuant to a
specific statutory or regulatory scheme, the narrow Plaza Health exception has not
been applied, precisely because the public interest has not been presumed to rest
with a single party. This explains why this Court recently upheld the denial of a
preliminary injunction sought by President Trump to restrain the enforcement of
a grand jury subpoena issued by the New York County District Attorney without
applying the Plaza Health exception in determining the applicable preliminary
injunction standard. See Trump v. Vance, 941 F.3d 631, 639–40 (2d Cir. 2019). It
explains our decision in Haitian Centers Council, Inc. v. McNary, 969 F.2d 1326 (2d
Cir. 1993), judgment vacated as moot by Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 918
Such cases may also exhibit an especial hesitancy on the part of federal courts
to substitute their own view of the public interest for that reached by local and state
governments in light of principles of comity and federalism.
28
45
Case 19-1540, Document 215, 12/03/2019, 2719370, Page46 of 59
(1993), in which we applied the serious questions standard to an injunction sought
against the actions of the Immigration and Naturalization Service only after
rejecting the government’s argument that the action was taken “pursuant to
Congress’[s] broad grant of authority in the [Immigration and Nationality Act],”
and reasoning that “in litigation such as is presented herein, no party has an
exclusive claim on the public interest,” id; see also, e.g., Patton v. Dole, 806 F.2d 24,
29–30 (2d Cir. 1986); Hudson River Sloop Clearwater, Inc. v. Dep’t of Navy, 836 F.2d
760, 763 (2d Cir. 1988); Mitchell v. Cuomo, 748 F.2d 804, 806–07 (2d Cir. 1984); cf.
Carey v. Klutznick, 637 F.2d 834, 839 (2d Cir. 1980) (rejecting the Census Bureau’s
argument that “the public interest [rests] solely with it”).
The government action at issue in the instant case plainly falls outside the
current confines of the narrow Plaza Health exception. Here, far from a situation
in which a movant seeks to enjoin action that is the product of “the full play of the
democratic process,” Able, 44 F.3d at 131, these legislative subpoenas, with due
respect, do not constitute governmental action pursuant to a statutory or
regulatory scheme and do not reflect the presumptively public‐interested actions
of both the legislative and executive branches. Rather, each subpoena is the
46
Case 19-1540, Document 215, 12/03/2019, 2719370, Page47 of 59
product of a sub‐component of a single chamber of one branch of the federal
government and, critically, implicates the interests of another branch.29
The majority’s approach, which concludes that, because the Committees act
pursuant to powers under the Constitution, such action should “[s]urely . . . not”
be evaluated under a “less rigorous standard” than that “applied to plaintiffs
seeking to preliminary enjoin state and local units of government” in cases such as
Central Rabbinical Congress and Monserrate, Maj. Op. at 20–21, is misguided for two
reasons. First, by deeming the “serious questions” standard to be less rigorous,
the majority ignores the fact that the ultimate burden is equivalent under both
standards.30 More fundamentally, the majority errs by categorically extending
Indeed, precisely because subpoenas of this sort implicate separation of powers
so that neither Congress nor the Plaintiffs can be taken to represent the public interest
with regard to their enforcement, the D.C. Circuit in Mazars declined to determine, in an
analogous context, what deference it owed to the congressional subpoena reviewed in
that case. Mazars, 940 F.3d at 726.
29
As is the nature of a sliding scale, the variables move in tandem and the
Plaintiffs’ ultimate burden is equivalent either way. The majority perceives tension
between this Court’s observation in Citigroup that the “overall burden” of the serious
questions standard is “no lighter than the one it bears under the ‘likelihood of success’
standard,” Citigroup, 598 F.3d at 35, and language in our other opinions that refers to the
likelihood‐of‐success standard as “more rigorous,” see, e.g., Cent. Rabbinical Cong., 763
F.3d at 192. See Maj. Op. at 14 n.22. I disagree. Because one standard requires a more
demanding showing as to the merits and a correspondingly less demanding showing as
to hardship, while the other standard requires the reverse, the overall burdens are clearly
equivalent. Deeming the likelihood‐of‐success standard to be “more rigorous” refers
only to its increased rigor as to the required merits showing. It was for this reason,
30
47
Case 19-1540, Document 215, 12/03/2019, 2719370, Page48 of 59
the Plaza Health exception to a situation in which “no party has an exclusive claim
on the public interest,” Time Warner Cable of N.Y.C. v. Bloomberg L.P., 118 F.3d 917,
923 (2d Cir. 1997) (quoting Haitian Centers, 969 F.2d at 1339), when the so‐called
“government action exception” is premised entirely on the assumption that the
public interest weighs decidedly against the movant.
To be clear, preliminary injunctions constitute an extraordinary form of
relief and should not issue lightly. See, e.g., Mazurek v. Armstrong, 520 U.S. 968,
972 (1997) (quoting Wright & Miller § 2948). The majority’s expansion of our so‐
called “government action exception” into the delicate arena of congressional
investigations, however, is unwise, precisely because this is a context in which
flexible application of equitable principles is vital. Historically, federal courts
have undertaken some of their most difficult assignments in the context of
reviewing the actions of congressional committees. The Supreme Court has thus
been required to take on the ”arduous and delicate task” of
“[a]ccommodat[ing] . . . the congressional need for particular information with the
among others, that we concluded in Citigroup that the Supreme Court’s decision in Winter
revealed “no command . . . that would foreclose the application of our established ‘serious
questions’ standard as a means of assessing a movant’s likelihood of success on the
merits” against the other components required to obtain preliminary relief. 598 F.3d at
38.
48
Case 19-1540, Document 215, 12/03/2019, 2719370, Page49 of 59
individual and personal interest in privacy.” Watkins, 354 U.S. at 198. It has
been called upon to address the “[g]rave constitutional questions” presented when
“the power of Congress to investigate” appears to encroach on the limits on that
power imposed by the Bill of Rights and, in particular, the First Amendment.
Rumely, 345 U.S. at 44, 48. Disputes between congressional committees and
Presidents arising from subpoenas, as here, also not uncommonly require courts
to “search for accommodation between the two branches”—a task for which this
Circuit’s flexible approach to making the difficult judgment whether a preliminary
injunction should issue is particularly well‐suited. United States v. Am. Tel. & Tel.
Co (“AT&T II”), 567 F.2d 121, 131 (D.C. Cir. 1977).
In short, we should not deprive ourselves of our traditional approach in
such a sensitive context. As we affirmed in Citigroup, “[r]equiring in every case
a showing that ultimate success on the merits is more likely than not is
‘unacceptable as a general rule,’” and also “deprive[s] the remedy of much of its
utility.” 598 F.3d at 35–36 (quoting Wright & Miller § 2948.3). Because this case
is not squarely covered by Plaza Health or any other previously‐articulated
“exception,” I conclude we are bound to (and should) undertake our usual
approach: namely, to consider the Plaintiffs’ showing as to the merits, balance of
49
Case 19-1540, Document 215, 12/03/2019, 2719370, Page50 of 59
hardships (merged here with the public interest inquiry, see Nken v. Holder, 556
U.S. 418, 435 (2009)), and irreparable harm and determine whether an injunction
is warranted under either the likelihood of success or serious questions standard.
As set forth already, moreover, these subpoenas do, in fact, present serious
questions implicating not only the investigative authority of these two House
committees, but the separation of powers between Congress and the Presidency.
*
*
*
Having determined that Plaintiffs have raised serious questions as to the
merits, in the usual case, the next step would be to assess the balance of hardships.
But this leads to my final point of departure from the majority. The majority
orders immediate compliance with these subpoenas save for a “few documents
that should be excluded” pursuant to its call for a restricted culling of certain
records assembled under specific subpoena categories. Maj. Op. at 86. In
contrast, I would not remand for the limited culling ordered by the majority, but
would instead remand in full, directing that the district court assist in the
development of the record regarding the legislative purposes, pertinence, privacy,
and separation‐of‐powers issues at stake in this case.
50
Case 19-1540, Document 215, 12/03/2019, 2719370, Page51 of 59
I would request the district court on remand promptly to implement a
procedure by which the Plaintiffs identify on privacy or pertinency grounds
specific portions of the material assembled in response to these subpoenas for
nondisclosure. Like the majority, I would then provide counsel for the
Committees with an opportunity to object, but I would also require counsel,
provided with a general description of such material, to articulate clearly the
legislative purpose that disclosure serves and to specify how the material sought
is pertinent to that purpose. Even assuming, arguendo, that the Committees act
pursuant to adequate authorization from the House as a whole, serious questions
persist as to the ends the Committees are pursuing and whether these ends are
adequate to justify the sought‐after disclosures.31 A fuller record would permit a
more informed calculus regarding balance of hardships and would further clarify
the stakes as to the serious questions that the Plaintiffs have already raised. This
full remand is superior to the majority’s approach for at least three reasons.
As to the “case study” rationale proffered by the House Financial Services
Committee, for instance, if that Committee is unable more clearly to articulate the
pertinence of its subpoenas to the legislative purposes it pursues, see Watkins, 354 U.S. at
214–15, the balance of hardships may well lie with the Plaintiffs, who will suffer
irreparable harm from the disclosure of their private and business affairs.
31
51
Case 19-1540, Document 215, 12/03/2019, 2719370, Page52 of 59
First and most fundamentally, remand is necessary here because the present
record does not permit a full assessment of either the serious questions raised by
these novel subpoenas or the balance of hardships with regard to specific
disclosures. The present record is wholly insufficient to support the conclusion
that the voluminous material sought pursuant to these subpoenas should at this
time be produced. Serious questions arising from the lack of historical precedent
for these subpoenas, their questionable authorization, their legislative purposes,
and the pertinence of particular disclosures remain. The record as to hardship,
moreover, is sparse, and does not reflect either parties’ concerns as to the disclosure
or nondisclosure of particular categories of information sought by these
extraordinarily broad subpoenas. The majority disagrees on both counts,
concluding that while the questions here may be “serious,” they are without merit,
Maj. Op. at 100–01, and that even if the balance of hardships tips in Plaintiffs’ favor,
it does not do so “decidedly,” Maj. Op. at 102. For the reasons already expressed,
however, I cannot join in this assessment.
Next (and notably), a broader remand is necessary here, even taking the
majority on its own terms—even assuming (incorrectly) that the district court’s
judgment could be substantially affirmed on the present record. This is because
52
Case 19-1540, Document 215, 12/03/2019, 2719370, Page53 of 59
the majority’s remand is inadequate to address the privacy and pertinency
concerns that the majority itself identifies and deems important. As to sensitive
personal information and an unspecified category of “nonpertinent” material, the
majority concludes that the Plaintiffs should be afforded an opportunity to object
to disclosure on privacy and pertinency grounds. It notes that “[t]he Committees
have advanced no reason why the legislative purposes they are pursuing require
disclosure” of “payment for anyone’s medical expenses,” for instance, and the
majority thus forbids it. Maj. Op. at 84. But by providing the Plaintiffs with an
opportunity to object only as to limited, specific categories of information sought
pursuant to these subpoenas, the majority creates the very potential for
unwarranted disclosure of sensitive information that it purports to disallow. The
majority thus orders compliance with, for instance, the Deutsche Bank subpoena’s
demand for “any document related to any domestic or international transfer of
funds in the amount of $10,000 or more,” including any “check,” J.A. at 38,
providing no opportunity for Plaintiffs to object that the sought‐after material is
sensitive and related to no legislative purpose at all.
Perhaps there is no material responsive to this category that would trigger
Rule 26(c)(1)’s protections against “embarrassment, oppression, or undue burden”
53
Case 19-1540, Document 215, 12/03/2019, 2719370, Page54 of 59
in a routine civil case. Fed. R. Civ. P. 26(c)(1). Perhaps such material does exist.
We cannot know until the documents are assembled and objections are made.
The privacy and pertinency concerns that the majority purports to address simply
cannot be addressed in the abstract. And by declining a full remand to permit a
record to be made, the majority affords less protection against the unwarranted
disclosure of personal information regarding a sitting President and his family
than would be afforded to any litigant in a civil case.
Finally, I also disagree with the majority’s implicit assessment that the
Plaintiffs have demonstrated no stake in the privacy of their business‐related
information that merits further review. Indeed, to the extent that the majority
does show a reasonable concern for the needless disclosure of Plaintiffs’ private
and nonpertinent information, this concern does not generally extend to private
business information at all, even though such information may implicate the same
issues of privacy and (non)pertinence. To be sure, the majority is correct that
Congress must have the ability to investigate businesses (even closely‐held ones)
in aid of legislation. And such investigations, serving a public good, will
sometimes cause competitive harm.32 But particularly in light of the very broad
32
Federal Rule of Civil Procedure 26(c)(1)(G) permits a district court to issue
54
Case 19-1540, Document 215, 12/03/2019, 2719370, Page55 of 59
disclosure sought by these subpoenas (which, with regard to many transactions,
could require the production of information from both this year and from decades
ago), the majority has proffered no clear reason for denying the Plaintiffs an
opportunity to object more generally to the disclosure of such material.
The majority argues that any hardship from business disclosures is offset in
this case by the fact that Presidents already “expose for public scrutiny a
considerable amount of personal financial information pursuant to the financial
disclosure requirement of the Ethics in Government Act, 5 U.S.C. app. §§ 101‐111.”
Maj. Op. at 102. But this is beside the point—or perhaps makes the point that the
majority’s approach is problematic.
Public disclosures made pursuant to the Ethics in Government Act are
required by law, pursuant to a statute that has run the gantlet of bicameralism and
presentment. In making disclosures pursuant to this Act, a President complies
with a statute that presumptively reflects a democratically enacted consensus
protective orders to prevent public disclosure of “confidential . . . commercial
information,” a protection not afforded or offered to the Plaintiffs by the Committees
here. The majority does not include these competitive harms as “irreparable injuries”
in its analysis, restricting its focus only to “loss of privacy.” See Maj. Op. at 101–02. The
irreversible nature of the competitive harm risked by immediate and unconditional
disclosure, and the lack of safeguards common to typical discovery procedures in civil
litigation, further buttress my view that these subpoenas, as drafted, raise serious
questions which a remand would aid in resolving.
55
Case 19-1540, Document 215, 12/03/2019, 2719370, Page56 of 59
regarding the financial disclosures that a Chief Executive should be required to
make. These House subpoenas, by contrast, require “considerably more financial
information,” as the majority concedes, but themselves raise substantial questions
as to whether they are supported by “sufficient evidence of legislative
authorization and purposes to enable meaningful judicial review.” Maj. Op. at
55, 102. And as Judge Katsas suggested in dissent from the denial of rehearing
in banc in Mazars, the scope of required disclosure “is determined . . . by the whim
of Congress—the President’s constitutional rival for political power—or even, as
in this case, by one committee of one House of Congress.” Mazars, 2019 WL
5991603, at *1 (Katsas, J., dissenting from the denial of rehearing en banc). In such
circumstances, and taking the majority’s analysis on its own terms, it is not clear
why the majority limits its remand to the particular categories of information that
it has selected, as opposed to permitting a more general opportunity to object
regarding nonpertinent business information and the irreparable injury that will
attend its disclosure.
For all the reasons that I have laid out here, this matter should be returned
to the district court. The remand that I have outlined would clarify the issues at
stake so that a reasoned determination could be made as to whether serious
56
Case 19-1540, Document 215, 12/03/2019, 2719370, Page57 of 59
questions persist, and where the balance of hardships lies. Indeed, given the lack
of historical precedent for these subpoenas; their extraordinary breadth; and the
persistent questions here regarding authorization, legislative purposes, and
pertinence, a remand for development of the record with regard to specific
categories of information is far preferable to the majority’s approach.
Such a procedure would also encourage negotiation between the parties and
potentially narrow the scope of this dispute. Because I conclude, contrary to the
majority, that this case implicates the Supreme Court’s caution to “tread warily”
in matters pitting the power of Congress to investigate against other substantial
constitutional concerns, Rumely, 345 U.S. at 46, and because the “serious
questions” delineated above sound in separation of powers, see Pub. Citizen v. Dep’t
of Justice, 491 U.S. 440, 466 (1989) (noting that the Supreme Court’s “reluctance to
decide constitutional issues is especially great where . . . they concern the relative
powers of coordinate branches of government”), this matter falls within a range of
cases in which we should attempt, if possible, to “avoid a resolution that might
disturb the balance of power between the two branches,” AT&T II, 567 F.2d at 123.
Perhaps that is not possible here. But as the D.C. Circuit has recognized in the
past, congressional committees and the Chief Executive “have a long history of
57
Case 19-1540, Document 215, 12/03/2019, 2719370, Page58 of 59
settlement of disputes that seemed irreconcilable” and such resolutions, where
possible, are to be preferred, since “[a] court decision selects a victor, and tends
thereafter to tilt the scales.” AT&T I, 551 F.2d at 394; see also id. at 391 (noting
possibility of “better balance . . . in the constitutional sense” from “political
struggle and compromise,” rather than court decision); Rumely, 345 U.S. at 45–46
(noting that a “[c]ourt’s duty to avoid a constitutional issue, if possible, applies not
merely to legislation . . . but also to congressional action by way of resolution”—
indeed, most especially in this context).
Accordingly, I would withhold decision as to balance of hardships and
remand to permit the district court and the parties the opportunity to provide this
Court with an adequate record regarding the legislative purpose, pertinence,
privacy and separation of powers issues in this case. Such a procedure, as in
AT&T I, 551 F.2d at 394–95, and AT&T II, 567 F.2d at 128–32, could narrow the
scope of the present dispute. But it is required in any event, because the record
simply does not support the majority’s decision to order immediate compliance
with these subpoenas, but for a “few documents,” Maj. Op. at 85, falling within its
preselected categories. To be clear, I reach this resolution guided by the Supreme
Court’s admonition in Rumely that the outer reaches of Congress’s investigative
58
Case 19-1540, Document 215, 12/03/2019, 2719370, Page59 of 59
power are to be identified reluctantly, and only after Congress “has demonstrated
its full awareness of what is at stake by unequivocally authorizing an inquiry of
dubious limits.” 345 U.S. at 46. Serious questions persist with regard to these
subpoenas—questions demanding close review lest such novel subpoenas prove
a threat to presidential autonomy not only now but in the future, and “to the
detriment of not only the President and his office but also the Nation that the
Presidency was designed to serve.” Fitzgerald, 457 U.S. at 753. Once the parties
have provided this Court with the information that I would seek on remand, we
would at that point have a sufficient record on which to make a prompt and
reasoned determination as to where the balance of hardships lies and whether the
Plaintiffs, having raised serious questions on the merits, are entitled to preliminary
relief.
59