Emily's List v. Federal Election Commission

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued May 4, 2009 Decided September 18, 2009

No. 08-5422

EMILY’S LIST,
APPELLANT

v.

FEDERAL ELECTION COMMISSION,
APPELLEE

Appeal from the United States District Court
for the District of Columbia
(No. 1:05-cv-00049-CKK)

Robert F. Bauer argued the cause for appellant. With
him on the briefs were Ezra W. Reese and Brian G. Svoboda.

David Kolker, Associate General Counsel, Federal
Election Commission, argued the cause for appellee. With
him on the brief were Thomasenia P. Duncan, General
Counsel, Harry J. Summers, Assistant General Counsel, and
Vivien Clair, Attorney. Gregory J. Mueller, Attorney, entered
an appearance.

Donald J. Simon, Fred Wertheimer, J. Gerald Hebert,
Trevor Potter, and Paul S. Ryan were on the brief for amici
 2
curiae Campaign Legal Center and Democracy 21 in support
of appellee.

Before: HENDERSON, BROWN, and KAVANAUGH, Circuit
Judges.

Opinion for the Court filed by Circuit Judge
KAVANAUGH, with whom Circuit Judge HENDERSON joins,
and with whom Circuit Judge BROWN joins as to Part IV
except footnotes 17, 18, and 20.

Opinion concurring in part filed by Circuit Judge
BROWN.

KAVANAUGH, Circuit Judge: A non-profit group known
as EMILY’s List promotes abortion rights and supports pro-
choice Democratic women candidates. It challenges several
new Federal Election Commission regulations that restrict
how non-profits may spend and raise money to advance their
preferred policy positions and candidates. EMILY’s List
argues that the regulations violate the First Amendment.

The First Amendment, as interpreted by the Supreme
Court, protects the right of individual citizens to spend
unlimited amounts to express their views about policy issues
and candidates for public office. Similarly, the First
Amendment, as the Court has construed it, safeguards the
right of citizens to band together and pool their resources as
an unincorporated group or non-profit organization in order to
express their views about policy issues and candidates for
public office. We agree with EMILY’s List that the new FEC
regulations contravene those principles and violate the First
Amendment. We reverse the judgment of the District Court
and direct it to enter judgment for EMILY’s List and to vacate
the challenged regulations.
 3
I

In the wake of the 2002 Bipartisan Campaign Reform Act
and the Supreme Court’s 2003 decision in McConnell v. FEC,
the election season of 2004 erupted with bitter accusations
about the activities of certain non-profit entities. The
controversy was popularly known by a single term – “527s” –
that refers to the section of the tax code applicable to non-
profits engaged in political activities. The debate arose after
wealthy individuals contributed huge sums of money to non-
profits ranging from America Coming Together to
MoveOn.org to Swift Boat Veterans for Truth in order to
support advertisements, get-out-the-vote efforts, and voter
registration drives. In total during the 2004 campaign, these
groups reportedly spent several hundred million dollars.

As the campaign unfolded, many in both major parties –
including President Bush and Senator Kerry – questioned the
activities of certain non-profits. Some encouraged the FEC to
ban large donations to non-profit entities in the same way that
Congress in BCRA had banned large contributions to political
parties. Proponents of additional regulation reasoned that
non-profits had replaced political parties as the soft-money
“loophole” in the campaign finance system. See Edward B.
Foley & Donald Tobin, The New Loophole?: 527s, Political
Committees, and McCain-Feingold, BNA MONEY & POL.
REP., Jan. 7, 2004.

In response, the FEC did not ban non-profits from
receiving and spending large donations, as some had urged.
But the FEC did limit how much non-profits such as
EMILY’s List could raise and spend. The FEC achieved this
objective by dictating that covered non-profits pay for a large
percentage of election-related activities out of their hard-
 4
money accounts. See 11 C.F.R. §§ 106.6(c), (f).1 Because
donations to those hard-money accounts are capped at $5000
annually for individual contributors, the FEC’s allocation
regulations substantially restrict the ability of non-profits to
spend money for election-related activities such as
advertisements, get-out-the-vote efforts, and voter registration
drives. The regulations separately require that donations to
non-profits be considered hard money subject to the $5000
cap if the corresponding solicitation indicated that donations
would be used to support the election or defeat of a federal
candidate. See id. § 100.57.

In early 2005, EMILY’s List filed suit, arguing that the
new regulations violated the First Amendment and the Federal
Election Campaign Act. In 2008, the District Court upheld
the regulations in their entirety.

II

To assess the constitutionality of the new FEC
regulations, we initially must address at some length the
relevant First Amendment principles set forth by the Supreme
Court.

A

Ratified in 1791, the First Amendment provides that
“Congress shall make no law . . . abridging the freedom of
speech.” U.S. CONST. amend. I. This guarantee “has its
fullest and most urgent application precisely to the conduct of
1
A hard-money account is subject to source and amount
limitations. For example, under the statute, EMILY’s List cannot
accept donations of more than $5000 annually into its hard-money
account from any single contributor. 2 U.S.C. § 441a(a)(1)(C). A
soft-money account may receive unlimited donations.
 5
campaigns for political office.” Buckley v. Valeo, 424 U.S. 1,
15 (1976) (internal quotation marks omitted). The
Amendment “protects political association as well as political
expression.” Id.; see also NAACP v. Alabama ex rel.
Patterson, 357 U.S. 449, 460 (1958).

In analyzing the interaction of the First Amendment and
campaign finance laws, the Court has articulated several
overarching principles of relevance here.

First, the Court has held that campaign contributions and
expenditures constitute “speech” within the protection of the
First Amendment. In Buckley, the foundational case, the
Court definitively ruled that “contribution and expenditure
limitations operate in an area of the most fundamental First
Amendment activities.” 424 U.S. at 14. The Court has never
strayed from that cardinal tenet, notwithstanding some
passionate objections. See, e.g., Nixon v. Shrink Mo. Gov’t
PAC, 528 U.S. 377, 398 (2000) (Stevens, J., concurring)
(“Money is property; it is not speech.”); J. Skelly Wright,
Politics and the Constitution: Is Money Speech?, 85 YALE
L.J. 1001 (1976).

Second, the Court has ruled that the Government cannot
limit campaign contributions and expenditures to achieve
“equalization” – that is, it cannot restrict the speech of some
so that others might have equal voice or influence in the
electoral process. In perhaps the most important sentence in
the Court’s entire campaign finance jurisprudence, Buckley
stated: “[T]he concept that government may restrict the
speech of some elements of our society in order to enhance
the relative voice of others is wholly foreign to the First
Amendment.” 424 U.S. at 48-49. The Court added that the
Government’s interest in “equalizing the relative ability of
 6
individuals and groups to influence the outcome of elections”
does not justify regulation. Id. at 48.

In Davis v. FEC, the Court strongly reiterated that
“equalization” is not a “legitimate government objective.”
128 S. Ct. 2759, 2773 (2008). The Davis Court approvingly
quoted Justice Kennedy’s observation in Austin v. Michigan
State Chamber of Commerce that “the notion that the
government has a legitimate interest in restricting the quantity
of speech to equalize the relative influence of speakers on
elections” is “antithetical to the First Amendment.” Id.
(citation and internal quotation marks omitted); see also
Austin v. Mich. State Chamber of Commerce, 494 U.S. 652,
684 (1990) (Scalia, J., dissenting) (“This illiberal free-speech
principle of ‘one man, one minute’ was proposed and soundly
rejected in Buckley”).2

Third, the Court has recognized a strong governmental
interest in combating corruption and the appearance thereof.
See Buckley, 424 U.S. at 26-27, 45-48; see also McConnell v.
FEC, 540 U.S. 93, 154 (2003). This, indeed, is the only
interest the Court thus far has recognized as justifying
campaign finance regulation. Davis, 128 S. Ct. at 2773
(“Preventing corruption or the appearance of corruption are

2
The Court’s rejection of the equalization argument is
consistent with its broader First Amendment jurisprudence: “As a
general matter, the American First Amendment tradition requires
that the financial, political, or rhetorical imbalance between the
proponents of competing arguments is insufficient to justify
government intervention to correct that imbalance.” Frederick
Schauer & Richard H. Pildes, Electoral Exceptionalism and the
First Amendment, 77 TEX. L. REV. 1803, 1825 (1999); see
generally Lillian R. BeVier, Money and Politics: A Perspective on
the First Amendment and Campaign Finance Reform, 73 CAL. L.
REV. 1045 (1985).
 7
the only legitimate and compelling government interests thus
far identified for restricting campaign finances.”) (citation and
internal quotation marks omitted). Importantly, the Court has
emphasized that the anti-corruption rationale is not boundless.
The core corruption that Government may permissibly target
with campaign finance regulation “is the financial quid pro
quo: dollars for political favors.” FEC v. Nat’l Conservative
PAC (NCPAC), 470 U.S. 480, 497 (1985). This anti-
corruption interest is implicated by contributions to
candidates: “To the extent that large contributions are given
to secure a political quid pro quo from current and potential
office holders, the integrity of our system of representative
democracy is undermined.” Buckley, 424 U.S. at 26-27; see
also Citizens Against Rent Control v. City of Berkeley, 454
U.S. 290, 296-97 (1981) (“Buckley identified a single narrow
exception to the rule that limits on political activity were
contrary to the First Amendment”; the exception relates “to
the perception of undue influence of large contributors to a
candidate”). Based on the close relationship between
candidates and parties and record evidence demonstrating that
political parties sold access to candidates in exchange for
contributions, the Court has held that the anti-corruption
interest also justifies limits on contributions to parties. See
McConnell, 540 U.S. at 154; see also Buckley, 424 U.S. at
38.3

Fourth, in applying the anti-corruption rationale, the
Court has afforded stronger protection to expenditures by
citizens and groups (for example, for advertisements, get-out-
the-vote efforts, and voter registration activities) than it has
3
Contributions include coordinated expenditures – that is,
expenditures coordinated with a candidate or party. See
McConnell, 540 U.S. at 121; FEC v. Colo. Republican Fed.
Campaign Comm., 533 U.S. 431, 443 (2001); see also 2 U.S.C. §
441a(a)(7)(B).
 8
provided to their contributions to candidates or parties. The
Court has explained that contributions to a candidate or party
pose a greater risk of quid pro quo corruption than do
expenditures. See Buckley, 424 U.S. at 46-47. At the same
time, the Court has stated that limits on contributions to
candidates or parties pose only a “marginal restriction upon
the contributor’s ability to engage in free communication.”
Id. at 20-21. By contrast, expenditure restrictions limit
“political expression at the core of our electoral process and
of the First Amendment freedoms.” Id. at 39 (internal
quotation marks omitted). A “restriction on the amount of
money a person or group can spend on political
communication during a campaign necessarily reduces the
quantity of expression by restricting the number of issues
discussed, the depth of their exploration, and the size of the
audience reached.” Id. at 19. The Court’s jurisprudence, in
short, reflects a “fundamental constitutional difference
between money spent to advertise one’s views independently
of the candidate’s campaign and money contributed to the
candidate to be spent on his campaign.” NCPAC, 470 U.S. at
497 (emphases added); see also Randall v. Sorrell, 548 U.S.
230, 241-42 (2006).4

4
Many have criticized the distinction between contributions
and expenditures because, in their view, they are “two sides of the
same First Amendment coin.” Buckley, 424 U.S. at 241 (opinion of
Burger, C.J.). Some contend that limits on contributions and
expenditures are both suspect under the First Amendment. See id.;
id. at 290 (opinion of Blackmun, J); see also Shrink Mo. Gov’t
PAC, 528 U.S. at 410 (Thomas, J., dissenting). Others argue that
the interest in limiting contributions similarly justifies restricting
expenditures. See Buckley, 424 U.S. at 260-62 (opinion of White,
J.). The Court thus far has rebuffed both critiques and adhered to
the Buckley divide. See Randall, 548 U.S. at 242 (“Over the last 30
years, in considering the constitutionality of a host of different
 9

In maintaining this line between (i) contributions to
candidates or parties and (ii) expenditures, the Court has
acknowledged that a citizen’s or group’s large expenditure –
for example, in financing advertisements or get-out-the-vote
activities – may confer some benefit on a candidate and
thereby give influence to the spender. But the Court
nonetheless has consistently dismissed the notion that
expenditures implicate the anti-corruption interest. See
Buckley, 424 U.S. at 47 (expenditures not “a quid pro quo for
improper commitments from the candidate”); see also
McConnell, 540 U.S. at 153 (“mere political favoritism or
opportunity for influence alone is insufficient to justify
regulation”); id. at 156-57 n.51 (Congress could not regulate
talk show hosts or newspaper editors “on the sole basis that
their activities conferred a benefit on the candidate”);
NCPAC, 470 U.S. at 498 (“exchange of political favors for
uncoordinated expenditures remains a hypothetical possibility
and nothing more”).

Fifth, the Court has been somewhat more tolerant of
regulation of for-profit corporations and labor unions. The
Court has permitted statutory limits on contributions that for-
profit corporations and unions make from their general
treasuries to candidates and parties.5 More controversially,
the Court has carved out a significant exception to Buckley’s
holding on expenditures: The Court has upheld laws that
prohibit for-profit corporations and unions from making
expenditures for activities expressly advocating the election

campaign finance statutes, this Court has repeatedly adhered to
Buckley’s constraints, including those on expenditure limits.”).
5
The Court also has ruled that the Government may bar
certain non-profit as well as for-profit corporations from making
direct contributions to candidates or parties. See FEC v. Beaumont,
539 U.S. 146, 159-60 (2003).
 10
or defeat of a federal candidate. The Court has permitted
those expenditure limits on the ground that they restrain the
“corrosive and distorting effects of immense aggregations of
wealth that are accumulated with the help of the corporate
form and that have little or no correlation to the public’s
support for the corporation’s political ideas.” Austin, 494
U.S. at 660; see also McConnell, 540 U.S. at 204-05; but see
First Nat’l Bank of Boston v. Bellotti, 435 U.S. 765, 776-77
(1978); Buckley, 424 U.S. at 48-49.6

To sum up so far: In reconciling the competing interests,
the Supreme Court has generally approved statutory limits on
contributions to candidates and political parties as consistent
with the First Amendment. The Court has rejected
expenditure limits on individuals, groups, candidates, and
parties, even though expenditures may confer benefits on
candidates. And the Court has upheld limits on for-profit
corporations’ and unions’ use of their general treasury funds
to make campaign contributions to candidates or political
parties or to make expenditures for activities expressly
advocating the election or defeat of federal candidates.

B

This case does not involve regulation of candidates,
parties, or for-profit corporations. Rather, this case concerns

6
The Supreme Court is presently considering whether to
overrule Austin (and McConnell’s reliance on it) to the extent
Austin permitted the Government to limit for-profit corporations’
and unions’ expenditures. See Citizens United v. FEC, No. 08-205
(S. Ct. reargued Sept. 9, 2009); cf. Austin, 494 U.S. at 702
(Kennedy, J., dissenting) (“Today’s decision abandons [Buckley’s]
distinction and threatens once-protected political speech.”). The
regulations at issue here violate the First Amendment with or
without Austin on the books. See infra note 11.
 11
the FEC’s regulation of non-profit entities that are not
connected to a candidate, party, or for-profit corporation. We
thus must consider how the constitutional principles outlined
above apply to non-profits – and in particular to three
different kinds of non-profits: (i) those that only make
expenditures; (ii) those that only make contributions to
candidates or parties; and (iii) those that do both. For
purposes of the First Amendment analysis, the central issue
turns out to be whether independent non-profits are treated
like individual citizens (who under Buckley have the right to
spend unlimited money to support their preferred candidates)
or like political parties (which under McConnell do not have
the right to raise and spend unlimited soft money).7

1

The first relevant category of non-profit entities consists
of those that only make expenditures for political activities
such as advertisements, get-out-the-vote efforts, and voter

7
In referring to non-profit entities, we mean non-connected
non-profit corporations (usually advocacy or ideological or
politically oriented non-profits) that engage in election-related
activities and register with the Internal Revenue Service under 26
U.S.C. § 527 or § 501(c), as well as unincorporated non-profit
groups. “Non-connected” means that the non-profit is not a
candidate committee, a party committee, or a committee established
by a corporation or labor union. See 11 C.F.R. § 106.6(a). “Non-
connected” for purposes of this opinion also excludes so-called
leadership PACs.
Some non-profits register with the FEC as political
committees; others do not. Our constitutional analysis of donations
to and spending by non-connected non-profits applies regardless
whether a non-profit has registered as a political committee with the
FEC. See infra note 15.
 12
registration drives. Non-profits in this category make no
contributions to federal candidates or parties.

The Supreme Court’s case law establishes that those non-
profit entities, like individual citizens, are constitutionally
entitled to raise and spend unlimited money in support of
candidates for elected office – with the narrow exception that,
under Austin, the Government may restrict to some degree
how non-profits spend donations received from the general
treasuries of for-profit corporations or unions. See Cal. Med.
Ass’n v. FEC, 453 U.S. 182, 202-03 (1981) (opinion of
Blackmun, J.); see also FEC v. Mass. Citizens for Life, Inc.
(MCFL), 479 U.S. 238, 259-65 (1986); NCPAC, 470 U.S. at
501; Citizens Against Rent Control, 454 U.S. at 296-99;
Buckley, 424 U.S. at 47; N.C. Right to Life, Inc. v. Leake, 525
F.3d 274, 292-93 (4th Cir. 2008).

Those principles were initially articulated in Cal-Med.
There, Justice Blackmun determined that “contributions to
political committees can be limited only if those contributions
implicate the governmental interest in preventing actual or
potential corruption, and if the limitation is no broader than
necessary to achieve that interest.” Cal-Med, 453 U.S. at 203
(opinion of Blackmun, J.). Applying that standard, he found
that “contributions to a committee that makes only
independent expenditures pose no such threat” of “actual or
potential corruption.” Id. “By pooling their resources,
adherents of an association amplify their own voices; the
association is but the medium through which its individual
members seek to make more effective the expression of their
own views.” Id. (citation and internal quotation marks
omitted). Justice Blackmun thus concluded that Government
 13
may not limit contributions to a non-profit that only makes
expenditures.8

The Court reinforced those principles a year later in
Citizens Against Rent Control. There, the Court struck down
limits on donations to a non-profit committee seeking to
defeat a ballot measure. See Citizens Against Rent Control,
454 U.S. at 296-99. Building on the established right of
individuals to make unlimited expenditures, the Court stated
that there are “of course, some activities, legal if engaged in
by one, yet illegal if performed in concert with others, but
political expression is not one of them.” Id. at 296. The
Court further reasoned: “Placing limits on contributions
which in turn limit expenditures plainly impairs freedom of
expression.” Id. at 299. In the Court’s words, to place “a
Spartan limit – or indeed any limit – on individuals wishing to
band together to advance their views on a ballot measure,
while placing none on individuals acting alone, is clearly a
restraint on the right of association.” Id. at 296.
8
In Cal-Med, there was no majority opinion on the First
Amendment issue. Under the Marks principle, Justice Blackmun’s
opinion in Cal-Med appears to be controlling. See Marks v. United
States, 430 U.S. 188, 193 (1977); cf., e.g., Regents of Univ. of Cal.
v. Bakke, 438 U.S. 265, 269-320 (1978) (opinion of Powell, J.).
Even if Justice Blackmun’s opinion were not binding under the
Marks principle or even if his discussion of expenditure-only non-
profits were considered dicta, his opinion’s principles have been
followed in subsequent decisions such as Citizens Against Rent
Control. In that regard, we note that Justice Marshall’s opinion in
Cal-Med did not decide how the First Amendment applies to
contributions to a non-profit that only makes expenditures. See
Cal-Med, 453 U.S. at 197 n.17 (opinion of Marshall, J.) (“American
Civil Liberties Union suggests that § 441a(a)(1)(C) would violate
the First Amendment if construed to limit the amount individuals
could jointly expend to express their political views. We need not
consider this hypothetical application . . . .”).
 14

In NCPAC, the Court reiterated that the Government may
not limit the spending of non-profits. The Court invalidated a
law that restricted a group’s expenditures in support of a
candidate who had accepted public financing. See NCPAC,
470 U.S. at 501. The Court stated that citizens’ “collective
action in pooling their resources to amplify their voices” is
“entitled to full First Amendment protection . . . .” Id. at 495.

In MCFL, the Court again underscored that non-profit
advocacy groups are generally entitled to raise and spend
unlimited money on elections. The Court invalidated an
expenditure limit imposed on a non-profit corporation that
had distributed a newsletter promoting pro-life candidates.
The Court noted that individuals “contribute to a political
organization in part because they regard such a contribution
as a more effective means of advocacy than spending the
money under their own personal direction.” MCFL, 479 U.S.
at 261. The Court added that “[v]oluntary political
associations do not suddenly present the specter of corruption
merely by assuming the corporate form.” Id. at 263; see also
Austin, 494 U.S. at 701 (Kennedy, J., dissenting) (MCFL held
that “a nonprofit corporation engaged in political discussion
of candidates and elections has the full protection of the First
Amendment”). Adhering to MCFL, the McConnell Court
ruled that BCRA’s ban on certain electioneering
communications could not validly be applied to non-profit
corporations. See McConnell, 540 U.S. at 210-11.

The principles set forth in Cal-Med, Citizens Against
Rent Control, NCPAC, and MCFL are rooted in the Court’s
consistent holdings beginning with Buckley that individual
citizens may spend money without limit (apart from the limit
on their own contributions to candidates or parties) in support
of the election of particular candidates. After all, if one
 15
person is constitutionally entitled to spend $1 million to run
advertisements supporting a candidate (as Buckley held), it
logically follows that 100 people are constitutionally entitled
to donate $10,000 each to a non-profit group that will run
advertisements supporting a candidate.9 Put another way: “If
the First Amendment prohibits any limitation on how much
money an independent political committee can spend on an
independent-expenditure campaign, how can it permit limits
on donations to committees that make only independent
expenditures?” Richard Briffault, The 527 Problem and the
Buckley Problem, 73 GEO. WASH. L. REV. 949, 982 (2005);
see also Edward B. Foley, The “Major Purpose” Test:
Distinguishing Between Election-Focused and Issue-Focused
Groups, 31 N. KY. L. REV. 341, 343 (2004) (stating “baseline
proposition that it would be unconstitutional to limit the
contributions that individuals may give to ideological groups
to be used for electoral purposes”); Note, The
Unconstitutionality of Limitations on Contributions to
Political Committees in the 1976 Federal Election Campaign
Act Amendments, 86 YALE L.J. 953 (1977).

These Supreme Court decisions reflect, moreover, the
commonsense proposition that regulation of non-profits does
not fit within the anti-corruption rationale, which constitutes

9
To be sure, some cases suggest that the First Amendment
interest in donating to someone else for speech, while important, is
less weighty than the First Amendment interest in speaking oneself.
But those cases involve contributions to candidates or parties. See
Buckley, 424 U.S. at 21. With respect to donations to groups other
than candidates or political parties, the Court has said that there are
“of course, some activities, legal if engaged in by one, yet illegal if
performed in concert with others, but political expression is not one
of them.” Citizens Against Rent Control, 454 U.S. at 296; see also
MCFL, 479 U.S. at 261; NCPAC, 470 U.S. at 495; Cal-Med, 453
U.S. at 203 (opinion of Blackmun, J.); NAACP, 357 U.S.at 460.
 16
the sole basis for regulating campaign contributions and
expenditures. See Davis, 128 S. Ct. at 2773. As the Court has
explained the anti-corruption principle, mere donations to
non-profit groups cannot corrupt candidates and
officeholders. In the words of the Fourth Circuit, it is
“implausible that contributions to independent expenditure
political committees are corrupting.” N.C. Right to Life, 525
F.3d at 293 (internal quotation marks omitted). And to the
extent a non-profit then spends its donations on activities such
as advertisements, get-out-the-vote efforts, and voter
registration drives, those expenditures are not considered
corrupting, even though they may generate gratitude from and
influence with officeholders and candidates. Rather, under
Buckley, those expenditures are constitutionally protected.
Therefore, limiting donations to and spending by non-profits
in order to prevent corruption of candidates and officeholders
represents a kind of “prophylaxis-upon-prophylaxis”
regulation to which the Supreme Court has emphatically
stated, “Enough is enough.” FEC v. Wis. Right to Life, Inc.
(WRTL), 551 U.S. 449, 478-79 (2007) (controlling opinion of
Roberts, C.J.).

Writing for the Fourth Circuit, Judge Wilkinson recently
summarized the relevant Supreme Court precedents,
concluding that “the Court has never held that it is
constitutional to apply contribution limits to political
committees that make solely independent expenditures.” N.C.
Right to Life, 525 F.3d at 292. Those non-profit groups
receive full First Amendment protection and are entitled to
receive donations and make expenditures because they “offer
an opportunity for ordinary citizens to band together to speak
on the issue or issues most important to them.” Id. at 295.
We agree with Judge Wilkinson’s assessment of the state of
the law.
 17
2

The second relevant category of non-profits consists of
those that only make contributions to federal candidates or
political parties and make no expenditures. Given the
constitutionally permissible caps on an individual donor’s
contributions to candidates or parties, the Supreme Court has
acknowledged the risk that individuals might use non-profits
to evade those limits. In order to prevent circumvention of
limits on an individual donor’s contributions to candidates
and parties, the Court has held that non-profit entities can be
required to make their own contributions to candidates and
parties, as well as pay associated administrative expenses, out
of a hard-money account that is subject to source and amount
restrictions. See Cal-Med, 453 U.S. at 198-99 (opinion of
Marshall, J.); id. at 203-04 (opinion of Blackmun, J.). As a
majority of the Court pointed out in Cal-Med, doing so
prevents non-profits from being used as “conduits” for illegal
contributions to parties and candidates and thus prevents
“evasion of the limitations on contributions” to a candidate.
Id. at 203 (opinion of Blackmun, J.); see also id. at 198
(opinion of Marshall, J.) (limit on donations to non-profit
prevents evasion of “$1,000 limit on contributions to
candidates . . . by channeling funds” through the non-profit);
Cal. Med. Ass’n v. FEC, 641 F.2d 619, 625 (9th Cir. 1980)
(Kennedy, J.) (non-profit committee is “natural conduit for
candidate contributions and . . . the essential purpose of the
provision here in question is to limit those contributions, not
to limit expenditures for any other type of political
advocacy”) (emphasis added).10
10
The requirement that certain administrative expenses be
funded in part with hard money prevents a contributor from
essentially taking control of a non-profit and thereby circumventing
limits on individual contributions to candidates. See Cal-Med, 453
U.S. at 198-99 n.19 (opinion of Marshall, J.); id. at 203 (opinion of
 18

Consistent with Cal-Med’s ruling, FECA limits
contributors to donating a maximum of $5000 per year to a
non-profit’s hard-money account. A non-profit in turn may
contribute to a candidate or party only from that hard-money
account. See 2 U.S.C. § 441a(a)(1)(C). And an individual’s
contribution to a non-profit’s hard-money account may count
against the individual’s aggregate annual contribution limits.
See 2 U.S.C. § 441a(a)(3).

3

What about a non-profit entity that falls into both
categories – in other words, a non-profit that makes
expenditures and makes contributions to candidates or
parties? EMILY’s List is a good example of such a hybrid
non-profit: It makes expenditures for advertisements, get-out-
the-vote efforts, and voter registration drives; it also makes
direct contributions to candidates and parties. In all of its
activities, its mission is to promote and safeguard abortion
rights and to support the election of pro-choice Democratic
women to federal, state, and local offices nationwide.

The constitutional principles that govern such a hybrid
non-profit entity follow ineluctably from the well-established
principles governing the other two categories of non-profits.
To prevent circumvention of contribution limits by individual
donors, non-profit entities may be required to make their own
contributions to federal candidates and parties out of a hard-
money account – that is, an account subject to source and

Blackmun, J.). But as discussed above, the Cal-Med Court never
stated that non-profits could be required to use hard money for
advertisements, get-out-the-vote activities, and voter registration
drives; indeed, Justice Blackmun’s opinion stated the opposite.
 19
amount limitations ($5000 annually per contributor).
Similarly, non-profits also may be compelled to use their
hard-money accounts to pay an appropriately tailored share of
administrative expenses associated with their contributions.
See Cal-Med, 453 U.S. at 198-99 n.19 (opinion of Marshall,
J.). But non-profit entities are entitled to make their
expenditures – such as advertisements, get-out-the-vote
efforts, and voter registration drives – out of a soft-money or
general treasury account that is not subject to source and
amount limits. Stated another way: A non-profit that makes
expenditures to support federal candidates does not suddenly
forfeit its First Amendment rights when it decides also to
make direct contributions to parties or candidates. Rather, it
simply must ensure, to avoid circumvention of individual
contribution limits by its donors, that its contributions to
parties or candidates come from a hard-money account.11

11
One additional wrinkle: To the extent a non-profit receives
donations from for-profit corporations or unions, those donations
cannot be placed in the non-profit’s hard-money account (because
for-profit corporate or union donations cannot be the source of
contributions to parties or candidates). Moreover, under Austin, the
soft-money account into which such donations are deposited cannot
be used to fund express-advocacy election activities that for-profit
corporations and unions are themselves banned from conducting.
Cf. WRTL, 551 U.S. at 476-77; MCFL, 479 U.S. at 259-64; FEC v.
NRA, 254 F.3d 173, 191-92 (D.C. Cir. 2001). Justice Souter
recently summarized these points: A “nonprofit may use its general
treasury to pay for clearly electioneering communications so long
as it declines to serve as a conduit for money from business
corporations and unions (and thus qualifies for the MCFL
exception).” WRTL, 551 U.S. at 521 (Souter, J., dissenting)
(internal quotation marks omitted). If Austin were overruled, then
non-profits would be able to make unlimited express-advocacy
expenditures from their soft-money accounts even if they accepted
donations from for-profit corporations or unions to those accounts.
 20
C

How does McConnell affect the above principles
governing non-profits? McConnell upheld congressionally
imposed limits on political parties receiving or spending soft
money. Some have argued that the Government can similarly
restrict soft-money contributions to and spending by non-
profits. In this case, the District Court accepted that reasoning
in ruling for the FEC; it found non-profits similarly situated to
political parties for purposes of the First Amendment analysis.

In our judgment, however, McConnell does not support
such regulation of non-profits. McConnell affirmed BCRA’s
limits on contributions to political parties because of the close
ties between candidates and parties and the extensive record
evidence of what it deemed a threat of actual or apparent
corruption – specifically, the access to federal officials and
candidates that large soft-money contributors to political
parties received in exchange for their contributions. The
Court said that it was “not unwarranted for Congress to
conclude that the selling of access gives rise to the appearance
of corruption.” McConnell, 540 U.S. at 154. The Court
expressly based its conclusion on the “close relationship
between federal officeholders and the national parties, as well
as the means by which parties have traded on that relationship
. . . .” Id.12

12
See generally McConnell, 540 U.S. at 130 (“both parties
promised and provided special access to candidates and senior
Government officials in exchange for large soft-money
contributions”); id. at 145 (“special relationship and unity of
interest” that candidates and officeholders share with parties); id. at
146 (“The evidence in the record shows that candidates and donors
alike have in fact exploited the soft-money loophole, the former to
increase their prospects of election and the latter to create debt on
the part of officeholders, with the national parties serving as willing
 21

Unlike the political parties examined in McConnell, there
is no record evidence that non-profit entities have sold access
to federal candidates and officeholders in exchange for large
contributions. See also Craig Holman, The Bipartisan
Campaign Reform Act: Limits and Opportunities for Non-

intermediaries.”); id. at 150 (“The record in the present cases is
replete with similar examples of national party committees peddling
access to federal candidates and officeholders in exchange for large
soft-money donations.”); id. at 151 (“So pervasive is this practice
that the six national party committees actually furnish their own
menus of opportunities for access to would-be soft-money donors,
with increased prices reflecting an increased level of access.”); id.
at 152 (“close ties that candidates and officeholders have with their
parties”); id. at 153-54 (“As the record demonstrates, it is the
manner in which parties have sold access to federal candidates and
officeholders that has given rise to the appearance of undue
influence.”); id. at 155 (“no meaningful separation between the
national party committees and the public officials who control
them”) (internal quotation marks omitted); id. (“Given this close
connection and alignment of interests, large soft-money
contributions to national parties are likely to create actual or
apparent indebtedness on the part of federal officeholders”); id.
(“This close affiliation has also placed national parties in a position
to sell access to federal officeholders in exchange for soft-money
contributions”); id. (“Access to federal officeholders is the most
valuable favor the national party committees are able to give in
exchange for large donations.”); id. at 156 n.51 (“[T]he record
demonstrates close ties between federal officeholders and the state
and local committees of their parties. That close relationship makes
state and local parties effective conduits for donors desiring to
corrupt federal candidates and officeholders. Thus, in upholding
§§ 323(b), (d), and (f), we rely not only on the fact that they
regulate contributions used to fund activities influencing federal
elections, but also that they regulate contributions to, or at the
behest of, entities uniquely positioned to serve as conduits for
corruption.”).
 22
Profit Groups in Federal Elections, 31 N. KY. L. REV. 243,
280 (2004) (“Today’s electioneering non-profit groups . . .
can make no such promises of access in exchange for a soft
money contribution.”).

More fundamentally, non-profit groups do not have the
same inherent relationship with federal candidates and
officeholders that political parties do. The McConnell Court
identified numerous “real-world differences between political
parties and interest groups.” 540 U.S. at 188. “Interest
groups do not select slates of candidates for elections.
Interest groups do not determine who will serve on legislative
committees, elect congressional leadership, or organize
legislative caucuses. Political parties have influence and
power in the Legislature that vastly exceeds that of any
interest group. As a result, it is hardly surprising that party
affiliation is the primary way by which voters identify
candidates, or that parties in turn have special access to and
relationships with federal officeholders.” Id. As noted in
McConnell, Congress recognized these differences and
enacted a statutory scheme under which “[i]nterest groups . . .
remain free to raise soft money to fund voter registration,
GOTV activities, mailings,” and advertising. Id. at 187.

In sum, it will not work to simply transport McConnell’s
holding from the political party context to the non-profit
setting. On this question as well, we agree with Judge
Wilkinson: “It is . . . not an exaggeration to say that
McConnell views political parties as different in kind than
independent expenditure committees.” N.C. Right to Life, 525
F.3d at 293.

For non-profit entities, the most pertinent Supreme Court
precedents remain Buckley, Cal-Med, Citizens Against Rent
Control, NCPAC, and MCFL. As discussed above, those
 23
cases ultimately stand for the proposition that non-profit
groups may accept unlimited donations to their soft-money
accounts. And subject to the one Austin-based exception,
non-profit groups – like individual citizens – may spend
unlimited amounts out of their soft-money accounts for
election-related activities such as advertisements, get-out-the-
vote efforts, and voter registration drives.13
13
Some have suggested that footnote 48 of the McConnell
opinion, in the course of discussing contributions to parties, subtly
re-interpreted Cal-Med to permit restrictions on large soft-money
donations to non-profits. See, e.g., Edward B. Foley & Donald
Tobin, The New Loophole?: 527s, Political Committees, and
McCain-Feingold, BNA MONEY & POL. REP., Jan. 7, 2004;
Memorandum from Prof. Daniel R. Ortiz, Univ. of Va. School of
Law, to Democracy 21 and the Campaign Legal Center (Apr. 9,
2004). We decline to adopt that expansive reading of footnote 48.
First, as explained by one leading election-law expert, such a
reading would require overruling the Supreme Court’s longstanding
dichotomy between limits on contributions and expenditures. See
Richard L. Hasen, Buckley is Dead, Long Live Buckley, 153 U. PA.
L. REV. 31, 70 (2004). Limits on donations to non-profit entities
are analytically akin to limits on expenditures by the donors. See
Cal-Med, 453 U.S. at 202 (opinion of Blackmun, J.); see also
Briffault, 73 GEO. WASH. L. REV. at 982 (“[I]f George Soros’s
direct expenditure of $23 million on anti-Bush or pro-Kerry ads is
constitutionally protected, how does he forfeit that protection if he
combines his $23 million with $20 million from Peter Lewis and
maybe another $10 million from some slightly smaller fry in a fund
that takes out essentially the same ads and supports the same voter
drives?”). For that reason, a broad interpretation of footnote 48
would mean “the entire Buckley edifice . . . falls.” Hasen, 153 U.
PA. L. REV. at 70. “Is that what the Court really intended buried in
a few sentences of a footnote in one of the longest cases in Supreme
Court history?” Id. We think not.
Second, footnote 48 simply cited Cal-Med together with
Buckley in the course of establishing the constitutionality of limits
on contributions to political parties, not to non-profits (which the
 24
III

We now consider whether the 2004 FEC regulations at
issue in this case comport with the relevant constitutional
principles. They do not.

The fundamental flaw, as counsel for EMILY’s List
succinctly stated at oral argument, is that the Commission
improperly “brought to bear what was essentially a political
party analysis to a non-connected, independent committee
which is not under the control of, or associated with

Court had no need to address). In the key concluding sentence in
the footnote, the Court rejected the idea that the government could
only regulate “parties as pass-throughs.” McConnell, 540 U.S. at
152 n.48 (emphasis added). Moreover, footnote 48 was responding
to a point in Justice Kennedy’s dissent that had nothing to do with
non-profits. See Briffault, 73 GEO. WASH. L. REV. at 986
(“importantly, the McConnell footnote was written in the course of
the Court’s analysis of BCRA’s application of contribution limits to
the activities of political parties”). We would unfairly wrench
footnote 48 from its context were we to adopt the broad
interpretation some have proposed.
Third, in a later passage in the McConnell opinion, the Court
explained that, under the statute, “[i]nterest groups . . . remain free
to raise soft money to fund voter registration, GOTV activities,
mailings,” and advertisements. 540 U.S. at 187. That passage –
and the accompanying discussion – would make little sense if
footnote 48 were read to equate non-profits with political parties.
Fourth, the Fourth Circuit in North Carolina Right to Life
refused to adopt this broad reading of footnote 48; it eschewed the
dissenting judge’s extensive reliance on it. See 525 F.3d at 333-34
(Michael, J., dissenting).
In short, we decline to read this footnote addressing a different
issue in McConnell to indirectly (i) overrule Buckley, (ii) discard
Justice Blackmun’s opinion in Cal-Med, and (iii) equate non-profits
with political parties, contrary to other discussion in McConnell.
 25
candidates in the fashion of a political party.” Tr. of Oral
Arg. at 4.

A

The rules set forth in §§ 106.6(c), 106.6(f), and 100.57
contain five relevant provisions. In our judgment, the
provisions are not closely drawn to meet an important
governmental interest.14

14
We need not decide whether the regulations are subject to
the strictest scrutiny applicable to spending restrictions or the still
“rigorous” but slightly lesser “closely drawn” scrutiny applicable to
contribution restrictions. See generally Davis v. FEC, 128 S. Ct.
2759, 2770-72 (2008); FEC v. Wis. Right to Life, Inc. (WRTL), 551
U.S. 449, 464 (2007); Buckley v. Valeo, 424 U.S. 1, 29 (1976).
Under either permutation of this “exacting” scrutiny, Buckley, 424
U.S. at 16, the regulations violate the First Amendment. That said,
the allocation and “mere reference” regulations of §§ 106.6(c) and
106.6(f) are best considered spending restrictions under the analysis
set forth in Wisconsin Right to Life. 551 U.S. at 457, 477 n.9, 478-
79; see also Cal. Med. Ass’n v. FEC, 453 U.S. 182, 203 (1981)
(opinion of Blackmun, J.) (limits on donations to non-profits
subject to strict scrutiny). In Wisconsin Right to Life, the Court
indicated that forcing an entity to spend out of a segregated fund
subject to source and amount limitations, rather than its general
treasury, was a spending restriction. 551 U.S. at 477 n.9. So too
here. Unlike BCRA’s rules for political parties, moreover, these
regulations do not limit how much someone can contribute to
EMILY’s List or other covered non-profits. Rather, these
regulations force non-profit entities to pay for a large percentage of
their varied political activities out of hard-money accounts subject
to source and amount ($5000) limits rather than out of soft-money
accounts that may receive unlimited donations. Through this
mechanism, the regulations limit how much non-profits ultimately
can spend on advertisements, get-out-the-vote efforts, and voter
registration drives. These regulations therefore “reduce[] the
 26

Under the Supreme Court’s precedents, non-profit
entities may be required to use their hard-money accounts for
their own contributions to candidates and parties and for an
appropriately tailored share of administrative expenses
associated with such contributions. But as explained above,
non-profits may not be forced to use their hard-money
accounts for expenditures such as advertisements, get-out-the-
vote efforts, and voter registration drives. Non-profits – like
individual citizens – are entitled to spend and raise unlimited
money for those activities. The FEC’s five new regulatory
provisions flout those principles.

First, the regulations require covered15 non-profit entities
to use their hard-money accounts to pay at least 50% of the

quantity of expression” for groups like EMILY’s List “by
restricting the number of issues discussed, the depth of their
exploration, and the size of the audience reached.” Buckley, 424
U.S. at 19. As a rough analogy, consider a law that requires home
buyers to pay a 50% cash down payment to obtain a mortgage.
That kind of law would significantly limit how much buyers could
afford to spend for a new house. A similar dynamic is at play as a
result of these regulations.
15
The regulations apply only to those non-profits that must
register with the FEC as political committees – namely, groups that
receive or spend more than $1000 annually for the purpose of
influencing a federal election and whose “major purpose” involves
federal elections. Buckley, 424 U.S. at 79; see 2 U.S.C. §§ 431-
434, 441a; supra note 7. Our constitutional analysis of donations
and spending limits applies both to non-connected non-profits
registered as political committees with the FEC and to non-
connected non-profits that are not so registered. The fact that a
non-profit spends a certain amount or percentage of its money in
relation to federal elections cannot be a basis, at least under the
anti-corruption rationale, for restricting its ability to accept large
donations to support those expenditures. That conclusion follows
 27
costs of their generic get-out-the-vote efforts and voter
registration activities. 11 C.F.R. § 106.6(c). By “generic,”
the regulations mean those activities that refer to a party but
do not promote or oppose a particular candidate. See id. §
100.25. This provision violates the First Amendment because
non-profits are constitutionally entitled to pay 100% of the
costs of such voter drive activities out of their soft-money
accounts. See Cal. Med. Ass’n v. FEC, 453 U.S. 182, 203
(1981) (opinion of Blackmun, J.); see also FEC v. Mass.
Citizens for Life, Inc. (MCFL), 479 U.S. 238, 259-63 (1986);
FEC v. Nat’l Conservative PAC (NCPAC), 470 U.S. 480, 501
(1985); Citizens Against Rent Control v. City of Berkeley, 454
U.S. 290, 298-99 (1981); Buckley v. Valeo, 424 U.S. 1, 45-48
(1976).

Second, the regulations mandate that covered non-profits
use their hard-money accounts for 50% of any generic
communications that refer to a party without referring to a
candidate, for example, “Support the Democratic party.” 11
C.F.R. § 106.6(c). This provision likewise violates the First
Amendment because non-profits are constitutionally entitled
to pay 100% of the costs of such communications out of their
soft-money accounts. See Cal-Med, 453 U.S. at 203 (opinion
of Blackmun, J.); see also MCFL, 479 U.S. at 259-63;
NCPAC, 470 U.S. at 501; Citizens Against Rent Control, 454
U.S. at 298-99; Buckley, 424 U.S. at 45-48.

from the Supreme Court’s consistent holdings that large
expenditures are constitutionally protected and the corresponding
principle that non-profits are constitutionally entitled to accept large
donations to their soft-money accounts to support advertisements,
get-out-the-vote efforts, and voter registration activities. Of course,
because of the lesser First Amendment protection against
disclosure, the major purpose test is permissible under current
precedent for determining non-profits’ disclosure obligations. See
Buckley, 424 U.S. at 79.
 28

Third, the regulations direct covered non-profit entities to
use their hard-money accounts to pay at least 50% of all
administrative expenses. 11 C.F.R. § 106.6(c). Those
administrative expenses include rent, utilities, office supplies,
and salaries, among other costs. But a non-profit may be
forced to use hard money for, at most, a percentage of
administrative expenses that “closely” corresponds to the
percentage of activities relating to its contributions as
compared to its advertisements, get-out-the-vote efforts, and
voter registration activities. See Davis v. FEC, 128 S. Ct.
2759, 2770 (2008) (campaign finance regulations must at
least be “closely drawn” to further an important governmental
interest); Cal-Med, 453 U.S. at 198-99 n.19 (opinion of
Marshall, J.) (rejecting argument that non-profit was entitled
to pay its “entire” administrative expenses with unlimited
donations or soft-money account) (emphasis added). The
tailoring must ensure that a hybrid non-profit is not unduly
advantaged as compared to a non-profit that makes only
contributions (and thus must fund certain administrative
expenses with hard money) and is not unduly disadvantaged
as compared to a non-profit that makes only expenditures
(and thus may fund its administrative expenses with soft
money). Section 106.6(c) does not attempt or purport to
allocate administrative expenses in that way. And the “desire
for a bright-line rule . . . hardly constitutes the compelling
state interest necessary to justify any infringement on First
Amendment freedom.” FEC v. Wis. Right to Life, Inc.
(WRTL), 551 U.S. 449, 479 (2007) (internal quotation marks
omitted) (controlling opinion of Roberts, C.J.).

Fourth, the regulations compel covered non-profit entities
to use their hard-money accounts to pay 100% of the costs of
advertisements or other communications that “refer” to a
federal candidate. 11 C.F.R. § 106.6(f)(1). If an
 29
advertisement or communication refers to a state candidate as
well as a federal candidate, the non-profit must pay for it with
a percentage of its hard-money account as determined by time
and space allocation. See id. § 106.6(f)(3). Here again, the
problem is that non-profits are constitutionally entitled to pay
100% of the costs of their advertisements and other
communications out of a soft-money account. See MCFL,
479 U.S. at 251; NCPAC, 470 U.S. at 501; Citizens Against
Rent Control, 454 U.S. at 299; Cal-Med, 453 U.S. at 203
(opinion of Blackmun, J.); Buckley, 424 U.S. at 45-48.

Fifth, the regulations create a new regime for solicitations
indicating that donated funds will be used to support or
oppose the election of a clearly identified federal candidate.
11 C.F.R. § 100.57. The regulations require that donations in
response to such solicitations be treated as 100% hard money.
Id. § 100.57(a)-(b)(1). This means that donations in response
to such solicitations are subject to a $5000 cap. If a
solicitation also refers to a state or local candidate, at least
50% of the responsive donations must go to the hard-money
account. Id. § 100.57(b)(2). This provision is badly flawed.
Non-profits are entitled to raise money for their soft-money
accounts to help support their preferred candidates, yet this
regulation prohibits non-profits from saying as much in their
solicitations. “Such notions run afoul of the fundamental rule
of protection under the First Amendment, that a speaker has
the autonomy to choose the content of his own message.”
WRTL, 551 U.S. at 477 n.9 (internal quotation marks
omitted); Davis, 128 S. Ct. at 2771 (provision that requires
choice between “unfettered political speech” and
“discriminatory fundraising limitations” violates First
Amendment).
 30
B

In short, the new FEC regulations do not pass muster
under the Supreme Court’s First Amendment precedents. The
regulations are not “closely drawn” to serve a cognizable anti-
corruption interest. See Davis, 128 S. Ct. at 2770-71; WRTL,
551 U.S. at 478-80; NCPAC, 470 U.S. at 496-97; Citizens
Against Rent Control, 454 U.S. at 296-97; Buckley, 424 U.S.
at 26-27, 45-48. Donations to and spending by a non-profit
cannot corrupt a candidate or officeholder, at least in the
absence of some McConnell-like evidence establishing such
corruption or the appearance thereof. See N.C. Right to Life,
Inc. v. Leake, 525 F.3d 274, 292-93 (4th Cir. 2008); see also
Richard Briffault, The 527 Problem and the Buckley
Problem, 73 GEO. WASH. L. REV. 949, 999 (2005) (“The 527s
do not fit easily within Buckley’s anticorruption paradigm, at
least as the Supreme Court has defined corruption until
now.”); Gregg D. Polsky & Guy-Uriel E. Charles, Regulating
Section 527 Organizations, 73 GEO. WASH. L. REV. 1000,
1027-35 (2005).

Of course, the fact that the regulations do not serve a
cognizable anti-corruption interest is not surprising because
the decision to more tightly regulate entities like EMILY’s
List arose out of an entirely different concern: the influence of
non-profits that raise and spend large amounts of money and
thereby affect federal elections. See, e.g., Comments of
Democracy 21, Campaign Legal Center & Center for
Responsive Politics in Response to Notice of Proposed
Rulemaking, No. 2004-6, at 1-2 (Apr. 5, 2004) (criticizing
“the spending of tens of millions of dollars of soft money
explicitly for the purpose of influencing the presidential
election by section 527 groups”). Responding to such
complaints, the FEC adopted these new regulations to tamp
down spending by non-profits and thereby better equalize the
 31
voices of citizens and groups who participate in the political
process. Large donations to and spending by non-profits
prompted these regulations, and limiting non-profits’
expenditures is their intended and predictable effect.

But the Supreme Court’s First Amendment cases have
repeatedly repudiated this equalization rationale as a basis for
regulating campaign-related contributions or expenditures.
See Davis, 128 S. Ct. at 2773; Buckley, 424 U.S. at 48-49.
Under current law, therefore, these regulations are
unsupportable. The concern with “large individual donations
to the 527s is that they permit a tiny group of Americans – the
wealthiest . . . – to play an enormous role in the electoral
process . . . . Buckley, however, rejected the protection of
political equality as a basis for limiting the role of money in
election campaigns.” Briffault, 73 GEO. WASH. L. REV. at
954 (internal quotation marks omitted).

As a lower court, we must strictly adhere to the Supreme
Court’s precedents. The regulations contravene the First
Amendment as it has been interpreted thus far by the Supreme
Court.16

C

As some commentators point out, it might seem
incongruous to permit non-profits to receive and spend large
soft-money donations when political parties and candidates
cannot. See Samuel Issacharoff & Pamela S. Karlan, The
Hydraulics of Campaign Finance Reform, 77 TEX. L. REV.
16
This case does not involve reporting and disclosure
obligations. The Government has a freer hand in imposing
reporting and disclosure requirements than it does in limiting
contributions and expenditures. See McConnell v. FEC, 540 U.S.
93, 121-22 (2003); Buckley, 424 U.S. at 64-68.
 32
1705, 1715 (1999). But this perceived anomaly has existed to
some extent since Buckley, which recognized that contribution
limitations “alone would not reduce the greater potential voice
of affluent persons and well-financed groups, who would
remain free to spend unlimited sums directly to promote
candidates and policies they favor in an effort to persuade
voters.” Buckley, 424 U.S. at 26 n.26. And McConnell
similarly took note of the fact that, even after that decision
upholding regulations on contributions to parties, “[i]nterest
groups . . . remain free to raise soft money to fund voter
registration, GOTV activities, mailings,” and advertisements.
McConnell v. FEC, 540 U.S. 93, 187 (2003).

If eliminating this perceived asymmetry is deemed
necessary, the constitutionally permitted legislative solution,
as the Court stated in an analogous situation in Davis, is “to
raise or eliminate” limits on contributions to parties or
candidates. 128 S. Ct. at 2774. But it is not permissible, at
least under current Supreme Court precedents, to remove the
incongruity by placing these limits on spending by or
donations to non-profits.

IV

In addition to its First Amendment challenge to the five
regulatory provisions, EMILY’s List alternatively contends
that three of the five provisions exceed the FEC’s statutory
authority. See 5 U.S.C. § 706(2)(C) (agency may not act “in
excess of statutory jurisdiction, authority, or limitations, or
short of statutory right.”). We agree.

When enacting BCRA in 2002, Congress did not
authorize the FEC to restrict donations to or spending by non-
profits – even though Congress was aware that BCRA’s
restrictions on political parties meant that independent non-
 33
profit groups would become more influential in the electoral
process. Indeed, Senator Lieberman, speaking in the Senate
at the time, anticipated that “at least some of the soft money
donors who will no longer be able to give to political parties
will be looking for other ways to influence our elections.
Donations to 527 groups will probably top many of their
lists.” 148 CONG. REC. S10779 (daily ed. Oct. 17, 2002)
(statement of Sen. Lieberman). He was right. Yet in the
seven years since BCRA was enacted, Congress still has not
imposed limits on non-profits, apparently because of
continuing constitutional and policy concerns about regulating
them in such a manner.

The statutory question, therefore, is whether the FEC’s
authority under the long-standing Federal Election Campaign
Act justifies the challenged regulations.

Under FECA, the FEC’s authority extends only to
regulating donations and expenditures made “for the purpose
of influencing any election for Federal office.” 2 U.S.C.
§ 431(8)(A)(i). As the Supreme Court has explained,
“[d]onations made solely for the purpose of influencing state
or local elections are therefore unaffected by FECA’s
requirements and prohibitions.” McConnell v. FEC, 540 U.S.
93, 122 (2003).

Under FECA, in other words, the FEC possesses
statutory authority to require a non-profit to use its hard-
money account to pay for federal activities, generic activities,
and mixed federal-state-local activities. See id. at 122-23.17

17
As explained earlier in this opinion, those approaches run
into severe First Amendment obstacles. For purposes of this
discussion, however, we analyze the statute as written without
regard to constitutional implications.
 34
But the FEC exceeds its statutory authority when it requires
non-profits to use hard money for exclusively state and local
election activities. See id. at 122; Chevron USA, Inc. v.
Natural Res. Def. Council, 467 U.S. 837, 842-43 (1984) (step
one).

The three regulatory provisions that EMILY’s List
challenges under FECA cross the statute’s boundaries.

EMILY’s List targets one of the provisions in § 106.6(c)
as exceeding the FEC’s statutory authority – namely, the part
requiring covered non-profits to use their hard-money
accounts to pay for 50% of their administrative expenses.
This requirement applies even if more than 50% of a non-
profit’s administrative expenses are exclusively associated
with state and local elections. That poses a problem because
the FEC possesses no authority under FECA to require non-
profits to use their hard-money accounts for their exclusively
state and local election activities. We thus concur with
EMILY’s List that this provision is overbroad and
“federalizes the funding and reporting of a large portion of
such a committee’s nonfederal receipts and disbursements,
which are not made for the purpose of influencing federal
elections.” EMILY’s List Br. at 39.18

18
As discussed above, § 106.6(c) also requires non-profits to
use their federal or hard-money accounts to pay for (i) at least 50%
of their generic get-out-the vote and voter registration activities and
(ii) at least 50% of their generic communications, which refer to a
party but not a candidate. In its brief, EMILY’s List does not raise
statutory challenges to those two provisions. See EMILY’s List Br.
at 35-40; id. at 38 (challenging under the statute only that provision
in § 106.6(c) that sets forth a “‘Minimum Percentages’ Rule for
Administrative Costs”). Presumably, EMILY’s List has not
challenged these two provisions under FECA because McConnell
indicated that these generic activities qualify under the statute as
 35

Next, EMILY’s List argues that § 106.6(f) exceeds the
FEC’s statutory authority. Recall that this provision requires
covered non-profits to use hard money for all or part of their
public communications that merely “refer” to federal
candidates. See 11 C.F.R. 106.6(f). The FEC runs roughshod
over the limits on its statutory authority when it presumes that
any public communications that merely “refer” to a federal
candidate necessarily seek to influence a federal election.19

For example, § 106.6(f) would compel a covered non-
profit to use some hard money to pay for an advertisement
running only in California in which Senator Jones from Maine
endorses Candidate Smith for Governor of California. The
sole purpose of such an advertisement is to influence the state
election in California – a matter entirely outside the FEC’s
statutory authority.

activities and communications “for the purpose of influencing”
federal elections. See McConnell, 540 U.S. at 123 (“Although a
literal reading of FECA’s definition of ‘contribution’ would have
required such activities to be funded with hard money, the FEC
ruled that political parties could fund mixed-purpose activities –
including get-out-the-vote drives and generic party advertising – in
part with soft money.”) (emphasis added).
19
The original FEC proposal was far narrower and would have
covered only communications that promote, attack, support, or
oppose federal candidates, similar to BCRA’s requirement for state
and local parties. See 69 Fed. Reg. at 11,753, 11,757-11,758 (Mar.
11, 2004) (notice of proposed rulemaking). After initially
considering that limited proposal, the FEC ultimately decided to
regulate broadly and to saddle non-profits with even greater
restrictions than Congress in BCRA chose to impose on state and
local parties.
 36
An incident illustrating § 106.6(f)’s statutory flaws
occurred in 2005. EMILY’s List sought to run advertisements
featuring Senator Stabenow in order to support Democratic
women candidates for state legislative offices. At the time,
Senator Stabenow was a candidate for reelection to the U.S.
Senate in Michigan. EMILY’s List represented that the
communication would not be distributed in Michigan, would
not reference Senator Stabenow’s federal candidacy, would
not solicit funds for her federal candidacy, and would not
refer to any clearly identified non-federal candidate. Rather,
it would support non-federal Democratic women candidates
as a class. Nonetheless, the FEC determined that the mere
reference to Senator Stabenow meant that EMILY’s List had
to pay for the communications with 100% hard money. See
FEC Adv. Op. 2005-13, at 3-4 (Oct. 20, 2005).

Finally, EMILY’s List argues that the solicitation rule set
forth in § 100.57 also exceeds the FEC’s statutory power. We
agree. To reiterate, § 100.57 requires covered non-profits to
treat as hard-money “contributions” all funds given in
response to solicitations indicating that “any portion” of the
funds received will be used to support or oppose the election
of a federal candidate. 11 C.F.R. §§ 100.57(a)-(b)(1)
(emphasis added). If the communication indicates that the
funds will support or oppose both a federal and non-federal
candidate, then at least 50% of those funds must be treated as
hard money. See id. § 100.57(b)(2). The statutory defect in
the rule is that, depending on the particular solicitation at
issue, it requires covered non-profits to treat as hard money
certain donations that are not actually made “for the purpose
of influencing” federal elections.

Consider a fundraising pitch in which a non-profit such
as EMILY’s List tells donors that only 10% of their gift will
be used to support identified federal candidates, with the rest
 37
to exclusively support state and local candidates. Each donor
fully and correctly understands that only a small portion of his
or her gift will be used “for the purpose of influencing”
federal elections. And yet, § 100.57 requires that at least 50%
of donations in response to such a solicitation be classified as
a hard-money donation subject to the $5000 cap – thereby
simultaneously creating a separate $5000 cap on soft-money
donations given in response to such a solicitation. This may
require a non-profit to decline or return funds it receives for
purely state and local elections. That is not permissible under
FECA.

In short, there is a significant mismatch between these
challenged provisions and the FEC’s authority under FECA.
Therefore, we conclude that §§ 106.6(f) and 100.57, as well
as the provision in § 106.6(c) that applies to administrative
expenses, exceed the FEC’s statutory authority.20

20
EMILY’s List separately argues that three of the five
regulatory provisions at issue in this case are also arbitrary and
capricious under the Administrative Procedure Act. EMILY’s List
Br. at 40-44. We are less persuaded by EMILY’s List’s free-
standing arbitrary and capricious argument. Putting aside the
constitutional and statutory-authority problems with the challenged
rules, the provisions are not otherwise arbitrary and capricious.
Agencies generally do not violate the APA’s deferential arbitrary-
and-capricious standard when they employ bright-line rules for
reasons of administrative convenience, so long as those rules fall
within a zone of reasonableness and are reasonably explained. See,
e.g., ExxonMobil Gas Mktg. Co. v. FERC, 297 F.3d 1071, 1084
(D.C. Cir. 2002); WorldCom, Inc. v. FCC, 238 F.3d 449, 461-62
(D.C. Cir. 2001).
EMILY’s List does not bring a challenge under either FECA
or the APA to § 106.6(c)’s requirement that covered non-profits
pay at least 50% of the cost of their generic communications out of
their hard-money accounts. See EMILY’s List Br. at 35-44.
 38
V

Before concluding, we add a few words regarding the
concurring opinion.

To begin, it is important to emphasize the area of
agreement between the opinion of the Court and the
concurrence. All three judges on the panel have determined
that §§ 106.6(c), 106.6(f), and 100.57 are unlawful and must
be vacated.

The concurrence advances two main points: (i) that under
McConnell, the Federal Government constitutionally may
regulate non-profits like political parties; and (ii) that we
should not address the First Amendment issue in this case.
Neither argument is convincing.

First, the concurrence contends that “regulation of
political parties is not McConnell’s theme,” and it reads
McConnell to support regulation not only of political parties
but also of independent non-profit groups. Concurring Op. at
19. As we have explained at length above, we do not find that
a persuasive interpretation of McConnell. In upholding Title I
of BCRA, the McConnell Court relied heavily on the “unity
of interest,” “close relationship,” and “close ties” among
candidates, officeholders, and political parties. The
concurrence identifies no similar unity of interest between
non-profits, on the one hand, and candidates, officeholders, or
parties on the other. The McConnell Court also based its
decision on the substantial record evidence of parties selling
access in exchange for soft-money contributions. The Court
repeatedly emphasized that “Congress must show concrete

EMILY’s List raises only a constitutional challenge to that
provision.
 39
evidence that a particular type of financial transaction is
corrupting or gives rise to the appearance of corruption . . . . It
has done so here.” McConnell, 540 U.S. at 185-86 n.72. The
concurrence cites no similar record evidence showing that
non-profits sell access to officeholders and candidates in
exchange for large soft-money contributions.

In our judgment, “McConnell views political parties as
different in kind than independent expenditure committees.”
N.C. Right to Life v. Leake, 525 F.3d 274, 293 (4th Cir. 2008).
We therefore disagree with the concurrence’s attempt to
stretch McConnell’s reasoning from political parties to non-
profits.21

21
The concurrence notes that McConnell upheld § 323(f) of
BCRA, which prohibits state and local candidates and officeholders
from using soft money for communications that promote, support,
attack, or oppose a federal candidate. We fail to see how this
aspect of McConnell justifies upholding limits on non-profits.
McConnell, as we read it, relied in part on the fact that
officeholders, candidates, and parties at all levels share a close
relationship and apparent unity of interest. And the Court also
based its conclusion with respect to state and local candidates and
officeholders – as elsewhere – on “the record in this litigation,” 540
U.S. at 185, emphasizing that “Congress must show concrete
evidence that a particular type of financial transaction is corrupting
or gives rise to the appearance of corruption . . . . It has done so
here.” Id. at 185-86 n.72.
The concurrence also raises concern about the activities of
non-profit committees that are “closely aligned” with federal
candidates. Concurring Op. at 23. But our constitutional analysis
of non-profits applies only to non-connected non-profits. See 11
C.F.R. § 106.6(a); supra note 7. Moreover, expenditures by
individuals or non-profits that are coordinated with a candidate may
be considered contributions to that candidate.
 40
Relatedly, the concurrence disputes our reading of Cal-
Med. See Concurring Op. at 24. But our analysis of that case,
including our reliance on Justice Blackmun’s opinion, tracks
the persuasive reasoning of the Fourth Circuit in North
Carolina Right to Life, of several other courts, and of
numerous commentators. See, e.g., N.C. Right to Life, 525
F.3d at 292; see also, e.g., Comm. on Jobs Candidate
Advocacy Fund v. Herrera, No. C 07-03199, 2007 WL
2790351, *4 (N.D. Cal. Sept. 20, 2007); Wash. State
Republican Party v. Wash. State Pub. Disclosure Comm’n, 4
P.3d 808, 825 (Wash. 2000); Richard Briffault, The 527
Problem and the Buckley Problem, 73 GEO. WASH. L. REV.
949, 982-85 (2005); John C. Eastman, Strictly Scrutinizing
Campaign Finance Restrictions (and the Courts that Judge
Them), 50 CATH. U. L. REV. 13, 37 (2000); Gregg D. Polsky
& Guy-Uriel E. Charles, Regulating Section 527
Organizations, 73 GEO. WASH. L. REV. 1000, 1031 (2005).
Moreover, the concurrence does not substantively address the
several post-Cal-Med cases that similarly recognize the right
of non-profits to raise and spend money to support their
agendas and preferred candidates. See, e.g., FEC v. Mass.
Citizens for Life, Inc. (MCFL), 479 U.S. 238, 259-65 (1986);
FEC v. Nat’l Conservative PAC (NCPAC), 470 U.S. 480, 501
(1985); Citizens Against Rent Control v. City of Berkeley, 454
U.S. 290, 296-99 (1981).

The concurrence further contends that we have not
received on-point briefing on the constitutional issue. We
again respectfully disagree. The briefs and oral argument
focused first and most extensively on the First Amendment
and McConnell – and debated the key question in this case:
For First Amendment purposes, are non-profits more like
individual citizens (who under Buckley have the right to spend
unlimited money to support their preferred candidates) or
more like political parties (which under McConnell do not)?
 41
See Buckley v. Valeo, 424 U.S. 1, 44-51 (1976); see also
McConnell, 540 U.S. at 155-56. Moreover, Cal-Med was
cited and discussed often in the briefs and at oral argument.
See Tr. of Oral Arg. 17-18, 32-33; FEC Br. at 21, 27, 28, 34;
Amicus Br. at 21, 24; EMILY’s List Reply Br. at 11, 19-20.
Indeed, the FEC’s brief noted that Cal-Med was a case
“chiefly relied upon.” FEC Br. at iv. In short, the briefs and
oral argument focused on and grappled with the critical issues
posed by the First Amendment challenge.

The concurrence suggests, however, that our holding
goes further than the submission of EMILY’s List. But
EMILY’s List forcefully argued that these regulations
“violate the First Amendment of the United States
Constitution,” contended that McConnell and Cal-Med do not
support the FEC’s approach, and asked the Court to vacate the
new regulations in their entirety. EMILY’s List Br. at 19. In
deciding this case, we have set forth the relevant
constitutional principles as we discern them, and we then have
applied those principles to the challenged regulations. In so
doing, we have concluded that the regulations violate the First
Amendment; we therefore have vacated the regulations,
which is precisely the relief EMILY’s List sought in
advancing its First Amendment claims.

Second, apart from its substantive disagreement with our
First Amendment analysis, the concurrence states that we
should resolve this case on statutory grounds alone, and
claims that it is “gratuitous” for us to address the First
Amendment. We respectfully but firmly disagree.

The threshold problem with the concurrence’s preferred
statutory-only approach is that EMILY’s List raises a
statutory challenge to only three of the five provisions at issue
here. EMILY’s List does not advance a statutory challenge to
 42
the provision in § 106.6(c) requiring that covered non-profits
use their federal or hard-money accounts to pay for at least
50% of their generic get-out-the-vote and voter registration
activities. Nor does EMILY’s List raise a statutory challenge
to the provision in § 106.6(c) requiring that covered non-
profits use hard money to pay for at least 50% of their generic
communications. Compare EMILY’s List Br. at 38-39
(discussing only administrative expenses provision of §
106.6(c) in statutory section of brief) with EMILY’s List Br.
at 32 (raising constitutional challenges to all provisions of §
106.6(c)). Indeed, footnote 11 of EMILY’s List’s brief all but
concedes that, under the statute, the FEC may require use of
hard money for these generic activities.

EMILY’s List’s decision not to target these two
provisions of § 106.6(c) on statutory grounds appears wise.
Such an argument would be very difficult to square with
McConnell’s several pointed statements that FECA permits
the FEC to treat generic activities as entirely federal for
purposes of contribution and expenditure limits. In fact,
McConnell harshly criticized the FEC for not having
previously treated political parties’ generic activities as
entirely federal activities subject to FECA’s limits. See
McConnell, 540 U.S. at 142 (by allowing generic activities to
be funded largely with soft money, FEC’s allocation regime
“subverted” original FECA scheme); id. at 167 (FECA
scheme “eroded” by FEC’s allocation regime); see also id. at
142 n.44.22

22
The concurrence finds “perplexing[]” our reading of
McConnell’s statutory discussion. Concurring Op. at 10. We think
it’s straightforward. McConnell said the statutory phrase “for the
purpose of influencing” federal elections covers generic activities.
McConnell, 540 U.S. at 167. That seems to foreclose any statutory
challenge to the new regulatory provisions applicable to generic
 43

Under the circumstances, we have no choice but to
address EMILY’s List’s First Amendment argument. The
concurrence apparently wants us to address a statutory
argument that EMILY’s List did not raise and then to accept
that statutory claim even though we find it unpersuasive and
inconsistent with precedent. We respectfully decline the
concurrence’s proposal.23

activities. But that of course does not resolve EMILY’s List’s
constitutional challenge.
23
Even if EMILY’s List had put forward meritorious statutory
challenges to all of the regulatory provisions that it has challenged
under the First Amendment, we still would possess discretion to
rule in the alternative on both statutory and constitutional grounds.
The avoidance principle cited by the concurrence is prudential, not
jurisdictional. And it is not uncommon for lower courts to rule on
alternative statutory and constitutional grounds when appropriate.
See, e.g., United States v. Thomas, No. 07-3080, 2009 WL
2152429, *3-5 (D.C. Cir. July 21, 2009); Time Warner Entm’t Co.,
L.P. v. FCC, 240 F.3d 1126, 1128 (D.C. Cir. 2001); see generally
Richard H. Fallon, Jr. & Daniel J. Meltzer, New Law, Non-
Retroactivity, and Constitutional Remedies, 104 HARV. L. REV.
1731, 1801 (1991) (“Simply as a routine matter, the Supreme
Court, in common with the lower federal courts, may choose to
discuss either or both of alternative grounds for reaching a
decision.”).
Given that the complaint in this case was filed four and a half
years ago, that the parties and the District Court overwhelmingly
focused their attention on the constitutional issue, that resolution of
the case only on statutory grounds would not alleviate the
continuing legal uncertainty, and that this is an area of law
demanding prompt and clear judicial decisionmaking, it would not
be an inappropriate exercise of judicial discretion for an
intermediate court to resolve this case on alternative constitutional
and statutory grounds.
 44

***

The FEC rules challenged by EMILY’s List –
§§ 106.6(c), 106.6(f), and 100.57 – violate the First
Amendment. Sections 106.6(f) and 100.57 also exceed the
FEC’s authority under the Federal Election Campaign Act, as
does the provision of § 106.6(c) that applies to administrative
expenses. The FEC may not enforce §§ 106.6(c), 106.6(f), or
100.57. We reverse the judgment of the District Court and
direct it to enter judgment for EMILY’s List and to vacate the
challenged regulations.

So ordered.

In any event, we need not cross that discretionary bridge here
because, as we have explained, we must address the Constitution’s
application to non-profits’ election-related spending and
fundraising in order to resolve the appeal.
 BROWN, Circuit Judge, concurring in part: “If there is
one doctrine more deeply rooted than any other in the process
of constitutional adjudication, it is that we ought not to pass
on questions of constitutionality . . . unless such adjudication
is unavoidable.” Spector Motor Serv., Inc. v. McLaughlin,
323 U.S. 101, 105 (1944). “Thus, if a case can be decided on
either of two grounds, one involving a constitutional question,
the other a question of statutory construction or general law,
the Court will decide only the latter.” Ashwander v. TVA, 297
U.S. 288, 347 (1936) (Brandeis, J., concurring). Because
these regulations must be vacated as contrary to the statute,
we need not and should not reach the First Amendment issue.
But if we’re going to answer an unnecessary constitutional
question, we at least ought to get it right. In light of
McConnell v. FEC, 540 U.S. 93 (2003), I have grave doubts
about the court’s analysis, which bears at most a passing
resemblance to the parties’ briefs, and which will profoundly
affect campaign finance law in this circuit. I thus respectfully
concur only with Part IV of the court’s opinion, except for
footnotes 17, 18 and 20.

I.

A.

Though I do not join their First Amendment holding, I
agree with my colleagues’ conclusion that we must vacate the
regulations challenged here (the Multiple Candidate
Allocation Regulation, 11 C.F.R. § 106.6(f); the Solicitation
Regulation, 11 C.F.R. § 100.57; and the Administrative Costs
Allocation Regulation, 11 C.F.R. § 106.6(c)). I begin with
first principles. The Federal Election Commission (FEC) is
an agency within the Executive Branch. The Executive
Branch cannot make law, but instead executes laws enacted
by the Legislative Branch. In executing the law, the FEC may
issue “necessary” rules, 2 U.S.C. § 437d(a)(8), but, as with all
agencies, the FEC acts contrary to law if it promulgates
 2
regulations “in excess of [its] statutory jurisdiction,” 5 U.S.C.
§ 706(2)(C). See also La. Pub. Serv. Comm’n v. FCC, 476
U.S. 355, 374 (1986) (“[A]n agency literally has no power to
act . . . unless and until Congress confers power upon it.”).

By the plain language of the Federal Election Campaign
Act (FECA), the FEC lacks the power it now asserts. To fall
within FEC jurisdiction, a “gift, subscription, loan, advance,
or deposit of money or anything of value” must be provided
to a political committee “for the purpose of influencing any
election for Federal office,” 2 U.S.C. § 431(8)(A)(i), and any
money expended by such a committee must also have been
done for that same purpose, id. § 431(9)(A)(i). There is no
other reasonable way to read Congress’s words. For the FEC
to have any role, money must be used for the “purpose”—
defined as an “objective, goal, or end,” BLACK’S LAW
DICTIONARY 1356 (9th ed. 2009)—of “influencing” an
“election for Federal office.” The inescapable corollary is the
FEC has no authority over money given or spent “solely for
the purpose of influencing state or local elections,” an activity
“unaffected by FECA’s requirements and prohibitions.”
McConnell, 540 U.S. at 122.

Here, the FEC has set aside Congress’s command that the
agency’s jurisdiction be bounded by the “purpose” for which
money is spent. Instead of strictly minding this jurisdictional
marker, the FEC conclusively presumes a federal purpose
drives any spending that might influence a federal election.1
The question though is not whether spending influences a
federal election, but whether it was spent for that reason.
1
See, e.g., Political Committee Status, Definition of Contribution,
and Allocation for Separate Segregated Funds and Nonconnected
Committees, 69 Fed. Reg. 68,056, 68,062 (Nov. 23, 2004) (Final
Rules) (“[R]eferences solely to a political party inherently influence
both Federal and non-Federal elections.”).
 3
Otherwise, the word “purpose” becomes superfluous, a result
that this court cannot accept, e.g., Reiter v. Sonotone Corp.,
442 U.S. 330, 339 (1979), especially for a jurisdictional
provision like this one, see, e.g., N. Am. Van Lines, Inc. v.
NLRB, 869 F.2d 596, 598 (D.C. Cir. 1989). Under FECA,
federal effects are simply not enough.2

Nor does labeling spending that may affect both state and
federal elections as “mixed-purpose” somehow solve the
FEC’s problem. Regulating on the basis of such a label still
assumes there must be a federal purpose behind any spending
that might influence, even tangentially, a federal campaign.
Because that necessary assumption is false, these regulations
remain invalid. Only after a federal purpose—mixed or
otherwise—is identified does the FEC’s power come into
play. If a federal purpose can be shown, then allocation ratios
like those promulgated here may well be appropriate under
FECA, but just asserting that there must be a federal purpose
skips the threshold jurisdictional question.3
2
Cf. 42 U.S.C. § 1973c (under the Voting Rights Act, certain
jurisdictions cannot alter their voting procedures without showing
the change “neither has the purpose nor will have the effect of
denying or abridging the right to vote on account of race or color”
(emphasis added)).
3
It is also no defense to say that under Buckley v. Valeo, 424 U.S. 1
(1976), everything political committees do reflects federal
purposes. See EMILY’s List v. FEC, 569 F. Supp. 2d 18, 44
(D.D.C. 2008) (“EMILY’s List is undoubtedly correct that its status
as a political committee does not automatically give the FEC
authority to regulate its legitimately nonfederal election
activities.”). In Buckley, the Court explained “[e]xpenditures of . . .
‘political committees’ . . . can be assumed to fall within the core
area sought to be addressed by Congress.” 424 U.S. at 79. This
“assum[ption],” however, is rebutted when it is unreasonable to
posit a federal purpose. See Akins v. FEC, 101 F.3d 731, 742 (D.C.
 4

B.

These regulations give short shrift to the “purpose” of
spending and so must be vacated. Indeed, we have already
rejected the FEC’s view. Just fours years ago, in Shays v.
FEC, we held FECA requires that “to qualify as ‘expenditure’
in the first place, spending must be undertaken ‘for the
purpose of influencing’ a federal election” and agreed with
the FEC that “time, place, and content may be critical indicia
of communicative purpose.” 414 F.3d 76, 99 (D.C. Cir.
2005). Though a federal

election-related intent is obvious . . . in statements
urging voters to “elect” or “defeat” a specified
candidate or party, the same may not be true of ads
identifying a federal politician but focusing on
pending legislation—a proposed budget, for example,
or government reform initiatives—and appearing three
years before the next election. Nor is such purpose
necessarily evident in statements referring, say, to a
Connecticut senator but running only in San Francisco
media markets.

Id.

Shays confirms what FECA says: context matters.
Referencing a federal candidate “may” reveal a federal
purpose, but if an ad will not be aired to her constituents or if
it will run “years before the next election,” then absent some
persuasive indicia of a federal purpose, a reference by itself to

Cir. 1996) (en banc) vac’d on other grounds 524 U.S. 1 (1998)
(Buckley only “creat[es] a presumption” of a federal purpose).
 5
the candidate does not trigger FEC jurisdiction. This was the
FEC’s position in Shays, and it should be the FEC’s position
now. FECA’s unambiguous text requires no less.

Under the Multiple Candidate Allocation Regulation,
political committees must use hard money for “[p]ublic
communications that refer to one or more clearly identified
Federal candidates, regardless of whether there is reference to
a political party, but do not refer to any clearly identified non-
Federal candidates.” 11 C.F.R. § 106.6(f)(1). The rule is
categorical. Even if a communication only says “Vote No on
State Ballot Initiative 123—They Waste Enough Taxes in
Washington, D.C.,” merely referring (perhaps by means of a
montage of grainy black-and-white photos) to United States
Senators of both parties with a penchant for pork-barrel
spending, the ad is per se deemed to have a federal purpose.
This is true even if those big spenders hail from distant states,
are not up for reelection for years, and the spot will not be
shown anywhere near their voters. Whether such an ad could
affect a federal race is doubtful, but the FEC goes further and
says these ads always reflect a federal purpose. In an age
when even pizza shops and used-car dealers invoke the
stereotype of wasteful federal spending to sell their wares, the
FEC’s lack of sophistication is startling.

The FEC’s approach also ignores that a state campaign
may be more effective if the campaigning group can mention
a federal official’s endorsement. Many federal politicians are
of national stature, particularly those associated with hot
button political issues. If, for example, a referendum would
make it more difficult to get an abortion, a pro-choice group
may trumpet a statement denouncing it from a prominent pro-
choice United States senator, while a pro-life group may
respond with a statement from an equally prominent pro-life
senator. To say, as the FEC does, that citing these statements
 6
can only be done for the purpose of influencing federal
elections is to pretend away the power of celebrity. Just as
Michael J. Fox’s name might be used in commercials for a
stem cell amendment because his association makes for
effective politics, see Alfonso Serrano, Stem Cell Opponents
To Air Celebrity Ad, CBSNEWS.COM, Oct. 25, 2006,
http://www.cbsnews.com/stories/2006/10/25/politics/main212
2383.shtml, an out-of-state and out-of-cycle Senator Debbie
Stabenow’s endorsement might be used to support state
candidates because her association makes for effective state
politics, particularly to targeted demographics.4

The Solicitation Regulation respects Congress’s language
no better. The regulation declares any donation “made by any
person in response to any communication is a contribution to
the person making the communication if the communication
indicates that any portion of the funds received will be used to
support or oppose the election of a clearly identified Federal
candidate.” 11 C.F.R. § 100.57(a). This rule applies even if a
solicitation’s banner headline says only a certain percentage
of the donation will be used for federal activities. Hence,
even if someone gives $1000 in response to a solicitation that
unambiguously says 90% of what is received will be spent on
local elections, the FEC asserts jurisdiction over the entire
gift. If a gift is made subject to this disclaimer, how is the full
$1000 given for the purpose of influencing a federal election?

The Solicitation Regulation also says “[i]f the solicitation
does not refer to any clearly identified non-Federal
4
“[R]eferring to Senator Stabenow might well inspire recipients
outside of her home state to contribute to her campaign, and thus
influence her federal election, or might otherwise raise her national
profile and ultimately influence her election,” EMILY’s List, 569 F.
Supp. 2d. at 48, but if such speculative brainstorming can satisfy
FECA’s “purpose” requirement, there is no “purpose” requirement.
 7
candidates, but does refer to a political party, in addition to [a]
clearly identified Federal candidate,” then the entire gift
becomes subject to the FEC’s authority, id. § 100.57(b)(1),
and “[i]f the solicitation refers to one or more clearly
identified non-Federal candidates, in addition to [a] clearly
identified Federal candidate . . . , at least fifty percent (50%)
of the total funds received are contributions [subject to the
FEC], whether or not the solicitation refers to a political
party,” id. § 100.57(b)(2). Even if the solicitation is
unmistakable that the entire gift will be spent on local
elections, the FEC nonetheless still claims jurisdiction over at
least some of the resulting donations.

This blindingly-bright line suffers from the same flaw as
the Multiple Candidate Allocation Regulation: it assumes
merely referencing a federal candidate always unmasks a
purpose of influencing a federal election and assumes those
who give money in response to such a solicitation also
unfailingly do so for the same purpose. That’s just not true.
Instead, a federal politician’s name can be used for reasons
tied solely to state electioneering. Again, consider an out-of-
state and out-of-cycle Senator Stabenow. If she were to say
“EMILY’s List supported a Democrat like me when I was
running for state office, and I’m asking you to support
EMILY’s List now so it can continue to work on behalf of
women who are seeking state office,” then under the
Solicitation Regulation, the entire amount of any donations is
subject to the FEC.5 That result conflicts with Congress’s
“purpose” requirement.
5
My hypothetical is similar to one EMILY’s List posed to the FEC,
with the only material difference being the inclusion of a party
label. While the FEC said EMILY’s List’s solicitation was fine,
see FEC Advisory Op. 2005-13, at 5–6 (Oct. 20, 2005), it could not
have said the same for mine: including “Democrat”—an important
label in state politics too—causes the gift to be hard money.
 8
Finally, the Administrative Costs Allocation Regulation is
contrary to law. Under 11 C.F.R. § 106.6(c), committees
must use “at least 50 percent Federal funds” for
“administrative expenses, costs of generic voter drives, and
costs of public communications that refer to any political
party.” An obvious problem is this rule applies in odd-
numbered years when there are no federal elections. If later
this November EMILY’s List were to say “vote a straight
Democratic ticket for the city counsel” in a broadcast in Walla
Walla, Washington, no less than half of the cost would have
to be expensed to a federal account, even though there is no
federal race to influence. The FEC is explicit: if a
communication mentions a political party, then context is
irrelevant. See FEC Advisory Op. 2005-13, at 4–5 (Oct. 20,
2005) (the duty to “pay the costs of public communications
that refer to a political party with at least 50 percent Federal
funds does not change based on the activities of [the
committee] in the particular State”).

Contrary to this regulation’s premise, moreover, certain
“administrative expenses” do not always reflect a federal
purpose, mixed or otherwise. A committee, for example, that
opposes human cloning (and thus supports many different
state and federal candidates and laws throughout the nation)
may launch an outpost in a state that is considering an anti-
cloning measure and organize a voter drive there,6 even
though the group has no intention of participating in any
federal election. By this regulation, a full half of the costs

6
A “generic voter drive” includes “any . . . activities that urge the
general public to . . . support candidates . . . associated with a
particular issue, without mentioning a specific candidate.” 11
C.F.R. § 106.6(b)(1)(iii) (emphasis added). Thus, if a committee
says “support candidates for the General Assembly who oppose
human cloning” in an odd-numbered year, hard money is required.
 9
must be expensed to the committee’s federal account. In fact,
the FEC would require the committee to use hard money for a
leaflet that says “both Democrats and Republicans” endorse
the initiative. Such a leaflet is not “generic party advertising,”
McConnell, 540 U.S. at 123; it is an ad for a state ballot
initiative, not a political party, and it defies reason to say
otherwise.7 FECA does not require this absurdity.

The court says EMILY’s List has waived part of its
statutory claim. First, my colleagues concede EMILY’s List
has challenged every other subpart of these regulations, and
agree the FEC has exceeded its statutory powers, but say
because EMILY’s List only mentions “administrative
expenses” in one section of its brief, it waives its argument
about “costs of generic voter drives” and “costs of public
communications that refer to any political party.” 11 C.F.R.
§ 106.6(c). This is so even though the clauses are in the exact
same sentence of the exact same regulation, and even though
they violate the exact same section of FECA for the exact
same reason. This is perplexing. EMILY’s List
comprehensively says “[n]or are the regulations permitted by
FECA. FECA was passed to regulate contributions and
expenditures made with ‘the purpose of influencing any
election for Federal office.’” EMILY’s List Br. at 17
(emphasis added). And if there is any doubt, EMILY’s List

7
“Generic party advertising” is not in the United States Code or the
Code of Federal Regulations. It was used by the McConnell district
court to mean, naturally, ads that support a party. See 251 F. Supp.
2d 176, 199 (D.D.C. 2003) (per curiam) (“[N]ational parties
expended $14 million in nonfederal funds for ‘generic’ party
advertising, consisting predominantly of television advertisements
that did not mention candidates names, but urged viewers to simply
vote for a particular party or stressed themes from the presidential
campaigns.”); id. at 654 (separate opinion of Kollar-Kotelly, J.)
(“generic party advertising (that is, ‘Vote Republican!’)”).
 10
opens its reply brief by saying “[t]he regulations at issue in
this case violate the First Amendment, Chevron[], and the
Administrative Procedure Act.” EMILY’s List Reply Br. at 1
(emphasis added). EMILY’s List also contends § 106.6(c)
requires using hard money for ads that state “both Democrats
and Republicans” support a measure. Id. at 16. It goes
without saying that EMILY’s List never argues “the
regulations all violate FECA, except for the second and third
clauses in the second sentence of 11 C.F.R. § 106.6(c).”8

Second and even more perplexingly, in order to buttress
its waiver argument, the same judges that read McConnell
narrowly as a case that “views political parties as different in
kind than independent expenditure committees,” Maj. Op. at
22, concludes EMILY’s List could not have successfully
challenged § 106.6(c)’s regulation of generic activities on
statutory grounds because McConnell specifically approved
regulation of contribution and expenditure limits for these
funds. Id. at 41–42. If this were correct, it would mean the
FEC can constitutionally regulate a committee like EMILY’s
List, and the court’s constitutional analysis is critically
undermined.

8
Even if we assume EMILY’s List has inadequately raised this
issue, waiver is a prudential doctrine—not jurisdictional. E.g.,
Mitchell v. Fishbein, 377 F.3d 157, 164–65 (2d Cir. 2004). And
though my colleagues are mistaken on waiver, if they were truly
concerned about this, we could order additional briefing. E.g., U.S.
Nat’l Bank v. Indep. Ins. Agents of Am., Inc., 508 U.S. 439, 444–48
(1993) (holding the D.C. Circuit did not err in ordering
supplemental briefing on a subject not raised because “when an
issue or claim is properly before the court, the court is not limited to
the particular legal theories advanced by the parties, but rather
retains the independent power to identify and apply the proper
construction of governing law”).
 11

C.

No one disputes the FEC can craft bright-line rules. An
“objective test,” Orloski v. FEC, 795 F.2d 156, 162 (D.C. Cir.
1986), in fact, may be constitutionally required, see FEC v.
Wis. Right to Life, Inc., 551 U.S. 449, 467–69 (2007) (opinion
of Roberts, C.J.). Communications calling for the election or
defeat of federal candidates certainly fall within the agency’s
authority, and the FEC may have a bit more leeway to
regulate ads directed at federal electorates or aired during
federal elections. Shays, 414 F.3d at 99. FECA’s
unambiguous text, however, forbids the Commission from
doing what it has done here: promulgating proxies for
“purpose” that wholly ignore all relevant contextual clues.
The regulations consequently must be vacated as contrary to
congressional will.

II.

A.

Because this case can be decided on statutory grounds,
we need not reach the constitutional question, and so should
not reach the constitutional question. Our precedent is not
wishy-washy: “Federal courts should not decide constitutional
questions unless it is necessary to do so. Before reaching a
constitutional question, a federal court should therefore
consider whether there is a nonconstitutional ground for
deciding the case, and if there is, dispose of the case on that
ground.” Kalka v. Hawk, 215 F.3d 90, 97 (D.C. Cir. 2000).
See also Meredith Corp. v. FCC, 809 F.2d 863, 870 (D.C. Cir.
1987); United States v. Thomas, 572 F.3d 945, 952 (D.C. Cir.
2009) (Ginsburg, J., concurring in part). My colleagues duck
this rule, preferring to summon the awesome power of
 12
Marbury v. Madison. But in their eagerness to play John
Marshall, they do not follow him. The Great Chief Justice
himself cautioned: “No questions can be brought before a
judicial tribunal of greater delicacy than those which involve
the constitutionality of a legislative act. If they become
indispensably necessary to the case, the court must meet and
decide them,” but if not, “a just respect for the legislature
requires, that the obligation of its laws should not be
unnecessarily and wantonly assailed.” Ex parte Randolph, 20
F. Cas. 242, 254 (C.C.D. Va. 1833) (No. 11,558).

The reasons to be “keenly mindful of our institutional
role” and “fully appreciate” the solemnity of constitutional
adjudication are obvious. Nw. Austin Mun. Util. Dist. No.
One v. Holder, 129 S. Ct. 2504, 2513 (2009). Meekness, for
one, compels us to recognize we are not the Constitution’s
only friend—each branch swears an oath to uphold it. See
U.S. CONST. art. II, § 1, cl. 8; Rostker v. Goldberg, 453 U.S.
57, 64 (1981). And if we misread a statute, Congress can fix
it; not so with the Constitution.

The court, however, is not content just answering a
gratuitous constitutional question. Its holding is broader than
even the plaintiff requests. Instead of arguing nonprofits have
a constitutional right to pay for ads attacking federal
candidates with soft money, EMILY’s List more modestly
challenges the regulations as the “functional equivalent of
spending limits, prohibiting EMILY’s List from supporting
state and local candidates in certain ways when its federal
funds are exhausted” and claims they are not properly tailored
because they “restrict vast amounts of nonfederal activity.”
EMILY’s List Br. at 17 (summary of argument) (emphasis
added). The court holds, nonetheless, that EMILY’s List is
constitutionally entitled to pay 100% of the costs of its
 13
advertisements out of a soft-money account, even for ads that
attack or promote federal candidates. Maj. Op. at 28–29.9

Because EMILY’s List’s actual claims are not bold
enough, the court sua sponte spins a more aggressive
argument—making its waiver charge all the more curious.
Nowhere does any party refer to Justice Blackmun’s separate
opinion in California Medical Association v. FEC, 453 U.S.
182 (1981) (“Cal-Med”). Nor does EMILY’s List mention
FEC v. National Conservative PAC, 470 U.S. 480 (1985)
(“NCPAC”), FEC v. Massachusetts Citizens for Life, Inc., 479
U.S. 238 (1986) (“MCFL”), or North Carolina Right to Life,
Inc. v. Leake, 525 F.3d 274 (4th Cir. 2008)—in other words,
the cases upon which the court’s holding depends, see, e.g.,
Maj. Op. at 11–16, 22–23, 26.10 None of the law review

9
Consistent with its briefing, during oral argument counsel for
EMILY’s List was more circumspect than the court is today:

The Court: Would your position preclude say regulation
of get out the vote drives, or voter
registration, or that sort of thing? Do you
think that would be beyond the FEC’s
purview?

EMILY’s List: No, Your Honor, we don’t take that position,
we take the position that reasonable
regulations to account for the federal election
related impact of that activity are
permissible.

Tr. of Oral Arg. at 9. Cf. Maj. Op. at 27 (“[N]on-profits are
constitutionally entitled to pay 100% of the costs of such voter
drive activities out of their soft-money accounts.”).
10
In fact, EMILY’s List does not mention Cal-Med until its reply
brief, and neither party cites Cal-Med for a proposition integral to
 14
articles the court relies on are cited—much less discussed—
either.11 And far from hashing out McConnell’s “footnote
48,” id. at 23 n.13, neither party mentions it. The court today
issues an expansive constitutional decision without the benefit
of on-point briefing. Before deciding a complicated First
Amendment issue, we ought to ask for the parties’ views.

In attempting to connect its decision to the parties’ views,
the court artfully re-imagines the “key question in this case,”
describing it as whether “non-profits [are] more like
individual citizens (who under Buckley have the right to spend
unlimited money . . . ) or more like political parties (which
under McConnell do not).” Maj. Op. at 40. The court notes
“Cal-Med was cited and discussed often in the briefs and at
oral argument.” Id. at 41. Cal-Med was cited and discussed
by the parties, but never to support the proposition for which
the court now relies on it. For instance, when asked by the
court to respond to the FEC’s reliance on Cal-Med, counsel

the court’s holding. EMILY’s List does cite, once, as a “see also,”
Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290
(1981), but for a background principle. EMILY’s List Br. at 23–24.
The FEC cites MCFL in a footnote for an immaterial point, FEC
Br. at 23 n.17, but does not mention NCPAC, Citizens Against Rent
Control, or North Carolina Right to Life.
11
Many of the articles do not support the court, and not one
squarely does. E.g., Edward B. Foley, The “Major Purpose” Test:
Distinguishing Between Election-Focused and Issue-Focused
Groups, 31 N. KY. L. REV. 341, 344 (2004) (“[I]n my judgment
contributions to political committees should be classified under the
First Amendment with contributions to political parties . . . .”);
Richard L. Hasen, Buckley is Dead, Long Live Buckley, 153 U. PA.
L. REV. 31, 72 (2004) (explaining that because of the “considerable
deference” it displayed, “[l]ower courts showing fidelity to
McConnell will have a difficult time striking down most campaign
finance regulation”).
 15
for EMILY’s List offered this succinct critique of the court’s
holding:

The Court: Can you deal with [FEC’s counsel’s]
response on [Cal-Med]?

EMILY’s List: Yes. I mean, Cal[-]Med is
mysteriously produced here for the
FEC’s position. Cal[-]Med didn’t
raise any of the issues in this case.
Cal[-]Med was a simple question of
whether a committee that was
making contributions to federal
candidates had to observe a limit on
contributions made to that federal
program. That set a law now, that’s
certainly what EMILY’s List does. I
don’t think it bears at all on this
invasion of our state and local
programs through the promulgation
of these excessive federal regulatory
schemes.

Tr. of Oral Arg. at 32–33. How true.

But even if it were judicially proper for me to do so, and
even if the issues were briefed, I doubt I could join the court’s
opinion in full. This is not because I dislike its outcome.
Indeed, I agree with what seems to be the unstated premise: if
the Supreme Court’s cases made any sense, the First
Amendment would protect much more than pornography,
profanity, and pyrotechnics. See United States v. Playboy
Entm’t Group, Inc., 529 U.S. 803 (2000); Cohen v.
California, 403 U.S. 15 (1971); Texas v. Johnson, 491 U.S.
397 (1989). The amendment’s “purpose,” after all, is “to
 16
preserve an uninhibited marketplace of ideas in which truth
will ultimately prevail,” FCC v. League of Women Voters,
468 U.S. 364, 377 (1984)—a principle that “has its fullest and
most urgent application to speech uttered during a campaign
for political office,” Eu v. S.F. County Democratic Cent.
Comm., 489 U.S. 214, 223 (1989). If those beautifully fierce
words “Congress shall make no law” are to do anything but
condemn our constitutionalism as a failed experiment, then at
least political speech in all its forms should be free of
government constraint.

My colleagues’ distaste for the FEC’s handiwork is to
their credit. It shows they take the First Amendment
seriously. And they are right, of course, that if constitutional
law were better acquainted with the Constitution, regulations
such as these would never survive Article III scrutiny. If an
advertisement criticizes the President of the United States, the
Speaker of the House of Representatives, or the Chair of the
Senate Committee on Foreign Relations, it can be a felony
punishable by up to five years in prison to pay for that ad
using money the federal government doesn’t know about or
that comes from sources the federal government deems to
have already given enough. See 2 U.S.C. § 437g(d). The
First Amendment, logically construed, cannot condone such a
weighty burden on political speech at the same time it forbids
penalizing the production of “virtual child pornography,”
Ashcroft v. Free Speech Coal., 535 U.S. 234 (2002).

I also agree with the court that “corruption” should only
be understood in terms of quid pro quo—not a free-floating
unease about money in politics. Once “corruption” is
disconnected from “pay to play,” Congress has carte blanche
to stifle speech, a license that is particularly pernicious as our
overweening government ever enlarges itself. Power—
government power—is what generates passion in politics.
 17
Money only measures its depth. The more power is at stake,
the more money will be used to shield, deflect, or co-opt it.
So long as the government can take and redistribute a man’s
livelihood, there will always be money in politics. One man’s
corruption is another man’s political accountability.

But there is a rub. We sit on a lower court and “must
follow the binding Supreme Court precedent” until the Court
itself overrules it. We the People Found., Inc. v. United
States, 485 F.3d 140, 144 (D.C. Cir. 2007). Though we do
not read the court above’s precedent unduly expansively, we
also do not drag our feet: “it is not our role to fight a rear-
guard action” against the logical implications of the Court’s
cases. United States v. Gardellini, 545 F.3d 1089, 1096 (D.C.
Cir. 2008). Instead, we take its holdings as we find them,
applying them to cover all they fairly address. Stare decisis
means nothing if we only are bound by those cases with
which we already agree. Like it or not, we cannot ignore
Supreme Court precedent.

Modesty is dictated by the difficulty of applying
McConnell’s facial generalizations to real world events. It is
hard to say exactly how contribution limits on hybrid
committees, like EMILY’s List, should be analyzed after
McConnell. McConnell involved a wide-ranging facial
challenge addressing the constitutionality of BCRA. Trying
to extrapolate from that case to this one is risky and reason
enough to avoid the constitutional bog. Suffice it to say that
the Supreme Court majority, the dissenters, and the
commentators all have read McConnell as a maximalist
opinion. See 540 U.S. at 192–93 (“[O]ur decisions in Buckley
and MCFL were specific to the statutory language before us;
they in no way drew a constitutional boundary that forever
fixed the permissible scope of provisions regulating
campaign-related speech.”); id. at 263 (Scalia, J., dissenting)
 18
(“We have witnessed merely the second scene of Act I of
what promises to be a lengthy tragedy.”); id. at 264 (Thomas,
J., dissenting) (“[T]he Court today upholds what can only be
described as the most significant abridgment of the freedoms
of speech and association since the Civil War.”); id. at 294
(Kennedy, J., dissenting) (“This new definition of corruption
sweeps away all protections for speech that lie in its path.”);
id. at 357 (Rehnquist, C.J., dissenting) (“Today’s decision, by
not requiring tailored restrictions, has significantly reduced
the protection for political speech having little or nothing to
do with corruption or the appearance of corruption.”); see also
Richard Briffault, The 527 Problem . . . and the Buckley
Problem, 73 GEO. WASH. L. REV. 949, 970 (2005)
(“McConnell, however, transformed the constitutional
landscape.”).

In both holding and discussion, the McConnell Court
sided with the censor, going so far as to rely on theoretical
anticipation to uphold a speech restriction, see 540 U.S. at
185 (upholding BCRA § 323(f), which forbids state and local
officeholders and candidates from using soft money to
support or attack federal candidates, based on the “eminently
reasonable prediction that . . . state and local candidates and
officeholders will become the next conduits for the soft-
money funding of sham issue advertising”), and ending with
an invitation for even more congressional action, id. at 224
(“We are under no illusion that BCRA will be the last
congressional statement on the matter.”). This is not the
modus operandi of a tentative tribunal; the Court knew what it
was doing, and said so. After McConnell, if these regulations
are within the FEC’s statutory power, then there is no obvious
reason they facially violate the First Amendment.12

12
This a facial challenge. For such claims, “exercising judicial
restraint . . . frees the Court not only from unnecessary
 19

It is true McConnell upheld, against an equal protection
challenge, a provision of BCRA regulating political parties.
But regulation of political parties is not McConnell’s theme.
The Court broadly recognized and deferred to governmental
interests in preventing corruption, the appearance of
corruption, and circumvention of election regulations.
McConnell, 540 U.S. at 136–37, 143–45, 152–53. Arguably,
this expansive corruption/circumvention/conduit rationale is
broad enough to encompass some limits on independent
expenditure committees, particularly for those political
committees with a self-proclaimed electoral mission. See
Briffault, 73 GEO. WASH. L. REV. at 986–87; Hasen, 153 U.
PA. L. REV. at 67–68. EMILY’s List is a multicandidate
political committee that has as its primary purpose electing
ideologically compatible candidates, EMILY’s List Br. at 3. 13

pronouncement on constitutional issues, but also from premature
interpretations of statutes in areas where their constitutional
application might be cloudy.” Wash. State Grange v. Wash. State
Republican Party, 128 S. Ct. 1184, 1190–91 (2008). As a rule, a
law is facially repugnant only if it “is unconstitutional in all of its
applications,” but there is a narrow First Amendment exception
where “a law may be overturned [if] a substantial number of its
applications are unconstitutional, judged in relation to the statute’s
plainly legitimate sweep,” id. at 1191 n.6. Those seeking this
“strong medicine,” id., however, face a “heavy burden,”
McConnell, 540 U.S. at 207.
13
In defining its mission, EMILY’s List explains that it “is
committed to a three-pronged strategy to elect pro-choice
Democratic women: recruiting and funding viable women
candidates; helping them build and run effective campaign
organizations; and mobilizing women voters to help elect
progressive candidates across the nation.” EMILY’s List, Our
Mission, http://emilyslist.org/about/mission/ (last visited Aug. 28,
2009).
 20
As such, it supports Democratic candidates for federal office.
See Matthew B. Stannard, Cash is Key for Tauscher’s
Replacement, S.F. CHRON., Aug. 24, 2009, at A1, available at
2009 WLNR 16474721 (noting that a candidate in a federal
primary race “has touted the endorsement of EMILY’s List, a
national fundraising organization that supported [former U.S.
Representative Ellen Tauscher] and backs Democratic women
who support abortion rights.”). Thus, EMILY’s List is much
more like a political party than the Sierra Club or the Society
for the Prevention of Cruelty to Animals.

Whether, under the high court’s current precedents,
EMILY’s List could be regulated exactly like a political party
is unknown. That it should not be regulated more harshly
than a political party seems to be the committee’s complaint
with the FEC regulations challenged in this court. EMILY’s
List Br. at 39 (“Bizarrely, [the regulation] also treats nonparty
PACs more harshly than any other type of committee, save
national parties and candidates themselves.”). My point is not
that McConnell mandates such treatment; only that nothing in
the opinion’s logic clearly precludes it. The court does not
think this is “a persuasive interpretation of McConnell.” Maj.
Op. at 38. Perhaps the court is right. But reading the case, as
the court does, to sanction First Amendment immunity for all
non-connected nonprofits seems even more implausible.

B.

Precedent holds allocation and solicitation rules are
contribution limits. This is key, as such limits receive less
than “strict scrutiny,” given they “‘entail only a marginal
restriction upon the contributor’s ability to engage in free
communication.’” McConnell, 540 U.S. at 134–35 (quoting
Buckley, 424 U.S. at 20)). The “‘overall effect’ of dollar
limits on contributions is [also] ‘merely to require candidates
 21
and political committees to raise funds from a greater number
of persons.’” Id. at 136 (quoting Buckley, 424 U.S. at 21–22).
Unlike strict scrutiny, which requires narrow tailoring to serve
compelling governmental interests, a contribution limit is
“valid [if] it satisfies the lesser demand of being closely
drawn to match a sufficiently important interest.” Id.

After McConnell, allocation and solicitation rules are
subject only to this lesser scrutiny. Facing BCRA § 323(a),
which forbids national parties from soliciting and spending
soft money, and § 323(b), which forbids state parties from
spending soft money on “federal election activities,” the
Court declined to apply strict scrutiny. 540 U.S. at 138–39.
The Court held “neither provision in any way limits the total
amount of money parties can spend. Rather, they simply limit
the source and individual amount of donations. That they do
so by prohibiting the spending of soft money does not render
them expenditure limitations.” Id. at 139. We instead ask
“whether the mechanism adopted to implement the
contribution limit, or to prevent circumvention of that limit,
burdens speech in a way that a direct restriction on the
contribution itself would not.” Id. at 138–39. Using the
Court’s standard, I agree with the district court that while the
FEC’s regulations “may affect the manner in which EMILY’s
List must fund the speech in which it chooses to engage, they
do not in any way limit the political speech that EMILY’s List
may undertake.” EMILY’s List, 569 F. Supp. 2d at 39.

The issue we confront then is whether these regulations,
facially, are closely drawn to match an important interest. To
answer, we again ought to look to McConnell. In upholding
BCRA § 323, the Court noted the “interests that underlie
contribution limits—interests in preventing both the actual
corruption threatened by large financial contributions and the
eroding of public confidence in the electoral process through
 22
the appearance of corruption.” McConnell, 540 U.S. at 136.
The Court bluntly held these interests are “not limited . . . to
the elimination of cash-for-vote exchanges,” but “extend to
the broader threat from politicians too compliant with the
wishes of large contributors.” Id. at 143. Congress must be
able to “address [these] more subtle but equally dispiriting
forms of corruption” by “remov[ing] the temptation” of “large
financial contributions.” Id. at 153. To combat “cynical
assumption[s],” Congress can “regulate the appearance of
undue influence,” with “undue influence” defined as “a sense
of obligation” or “grat[itude],” id. at 144–45. Importantly,
Congress also has an interest in preventing the circumvention
of these limits—and so can use broad prophylaxes—because
“candidates, donors, and parties test the limits of the current
law.” Id. at 144. In sum, McConnell defines the “interest” so
broadly it is hard to imagine regulations that are not properly
drawn to it. See id. at 356–57 (Rehnquist, C.J., dissenting).

Following from such an encompassing statement of the
interest, the McConnell Court facially upheld many onerous
restrictions on the use of soft money. The Court emphasized,
for example, that even a complete ban on soliciting nonfederal
funds would still “leave open ample opportunities for
soliciting federal funds,” and noted such restrictions “increase
the dissemination of information by forcing parties,
candidates, and officeholders to solicit from a wider array of
potential donors.” Id. at 139–40. The Court also upheld a
ban on any use of soft money by national parties, even for
those “minor parties” that are unlikely to have any tangible
electoral success, remarking only that “a nascent or struggling
minor party can bring an as-applied challenge if § 323(a)
prevents it from amassing the resources necessary for
effective advocacy.” Id. at 159. While national parties do not
always act on behalf of or in concert with federal candidates,
a prophylactic prohibition on any soft money spending and
 23
soliciting by them is facially valid, given the “close
connection and alignment of interests” between national
parties and federal candidates. Id. at 155.

The Court’s discussion of national parties by itself raises
difficulties for the court, especially because the Court already
seemingly has held there is a “close connection and alignment
of interests” between committees like EMILY’s List and
federal candidates. In Cal-Med, the Court sustained
contribution limits to multicandidate committees. Four
justices adopted the Conference Report’s conclusion that
these committees may “‘appear to be separate entities
pursuing their own ends, but are actually a means for
advancing a candidate’s campaign,’” 453 U.S. at 199 n.18
(plurality opinion) (quoting H.R. Conf. Rep. No. 94-1057, pp.
57–58 (1976)). Justice Blackmun penned a concurring
opinion, but did not disclaim the Conference Report or
disagree with the Court’s ultimate holding. In fact, he
expressly said “contributions to multicandidate committees
may be limited to $5,000 per year as a means of preventing
evasion [of contribution limits],” though he noted in dicta that
his conclusion would be different if the committee “makes
only independent expenditures” and so does not “pose a
perceived threat of actual or potential corruption.” Id. at 203
(opinion of Blackmun, J.) (emphasis added). If Congress can
forbid all allocation and solicitation of soft money by national
parties, how is it unconstitutional to forbid only some
allocation and solicitation of soft money by multicandidate
committees that are also closely aligned with federal
candidates?

The court disputes this reading of Cal-Med, claiming
there are not just two types of political committees (ones that
only make independent expenditures, and all others), but
actually three: (1) those that only make independent
 24
expenditures; (2) those that only contribute to candidates; and
(3) those that make independent expenditures and contribute
to candidates. Such political committees are, in the court’s
view, entitled to raise and spend “unlimited money” for
advertisements, get-out-the-vote efforts, and voter registration
drives. Maj. Op. at 26. If these hybrid committees contribute
to federal candidates, they must use hard money, says the
court, but all other spending can be with soft money.

This novel argument is not without considerable charm,
but one must read Cal-Med with a squint to see that holding.14
There is no indication in Cal-Med that the committee did not
make independent expenditures, but the Court still sustained
the statute, without announcing the distinction the court draws
today. The Court has consistently cited Cal-Med for the
unqualified proposition that it is constitutional to limit
contributions to multicandidate committees. See, e.g., FEC v.

14
The court relies heavily on Justice Blackmun’s concurring
opinion in Cal-Med, arguing that it is controlling. But the opinion
is controlling, if at all, only for “points that can be said to be fairly
subsumed within the reasoning of the plurality.” John C. Eastman,
Strictly Scrutinizing Campaign Finance Restrictions (and the
Courts that Judge Them), 50 CATH. U. L. REV. 13, 37 (2000); see
Marks v. United States, 430 U.S. 188, 193 (1977). This circuit has
clarified it is only the narrowest opinion’s overlap with the broader
opinion that counts. “Marks is workable—one opinion can be
meaningfully regarded as ‘narrower’ than another—only when one
opinion is a logical subset of other, broader opinions. In essence,
the narrowest opinion must represent a common denominator of the
Court’s reasoning; it must embody a position implicitly approved
by at least five Justices who support the judgment.” King v.
Palmer, 950 F.2d 771, 781 (D.C. Cir. 1991) (en banc). Presumably
then, the controlling part of Justice Blackmun’s opinion is the
holding that the FEC may constitutionally regulate contributions to
fund independent political expenditures without contravening the
First Amendment—no more and no less.
 25
Colorado Republican Fed. Campaign Comm., 533 U.S. 431,
441–42 (2001); FEC v. NRA Political Victory Fund, 513 U.S.
88, 97 (1994). See also Buckley, 424 U.S. at 38.

The infamous “footnote 48 of the McConnell opinion,”
Maj. Op. at 23 n.13, also flatly contradicts the court:

Justice KENNEDY’s contention that Buckley limits
Congress to regulating contributions to a candidate
ignores Buckley itself. There, we upheld FECA’s
$25,000 limit on aggregate yearly contributions to
candidates, political committees, and party committees
out of recognition that FECA’s $1,000 limit on
candidate contributions would be meaningless if
individuals could instead make “huge contributions to
the candidate’s political party.” Likewise, in [Cal-
Med], we upheld FECA’s $5,000 limit on
contributions to multicandidate political committees.
It is no answer to say that such limits were justified as
a means of preventing individuals from using parties
and political committees as pass-throughs to
circumvent FECA’s $1,000 limit on individual
contributions to candidates. Given FECA’s definition
of “contribution,” the $5,000 and $25,000 limits
restricted not only the source and amount of funds
available to parties and political committees to make
candidate contributions, but also the source and
amount of funds available to engage in express
advocacy and numerous other noncoordinated
expenditures. If indeed the First Amendment
prohibited Congress from regulating contributions to
fund the latter, the otherwise-easy-to-remedy
exploitation of parties as pass-throughs (e.g., a strict
limit on donations that could be used to fund candidate
 26
contributions) would have provided insufficient
justification for such overbroad legislation.

McConnell, 540 U.S. at 152 n.48.

That the Court may have created a doctrinal anomaly
might suggest, as the court argues, see Maj. Op. at 23 n.13,
that it could not possibly have meant what it said, but that is a
hard argument to make. Often cases are in tension as doctrine
works itself pure. Our duty as an intermediate court is not to
tell the Court what it ought to have said, but to abide by what
it did say.

But even leaving aside its treatment of national parties,
McConnell further undermines the court. Recognizing the
dynamic—indeed, Sisyphean—character of campaign finance
law, the Court noted “[m]oney, like water, will always find an
outlet.” 540 U.S. at 224. Upon realizing structural forces
inherent in a republic inevitably create incentives for those
subject to regulation to petition for relief and to campaign
against those who are disinclined to grant it, the Court did not
retreat to a more manageable and less burdensome “quid pro
quo” standard. Id. at 296 (Kennedy, J., dissenting). Instead,
after upholding § 323(a), the Court emphasized money would
now “corrupt” federal races by other, more subtle routes, so
Congress, in anticipation of this new corruption, can enact
broad anti-circumvention measures. E.g., id. at 165–66.
Indeed, without pointing to any evidence that local officials
(e.g., county assessors) are connected to federal candidates
(e.g., for President of the United States) or have been used to
circumvent the law, the Court held Congress prophylactically
can regulate them without facially offending the Constitution.
See id. at 184–85. If the First Amendment is flexible enough
to allow regulating local officials because contributions might
flow through them to federal candidates, then why can’t the
 27
FEC also make an “eminently reasonable prediction” that
committees like EMILY’s List—which actually do campaign
for candidates and give them money15—will be the next route
for corruption, and regulate accordingly? Id. at 185.

The court sidesteps McConnell by saying EMILY’s List
is a nonprofit, not a political party, and so has more
constitutional rights.16 But EMILY’s List is not just a
nonprofit; it is a multicandidate political committee that
campaigns for and contributes money to federal candidates.
In upholding § 323 in full, including § 323(f), McConnell
blessed restrictions on local officeholders, who are not, of
course, political parties. The rule then cannot be that parties
and federal candidates are in one column, and everyone else is
in another, because that does not explain McConnell. Instead,
there is a spectrum, N.C. Right to Life, 525 F.3d at 291, so we
should ask not whether an entity is a nonprofit, but instead
where it falls on the spectrum. Specifically, is EMILY’s List
more or less likely than a local official to act as a conduit to
federal candidates? “Common sense” says such committees

15
Britt Cocanour, EMILY’s List Chief of Staff, avowed her
committee “has helped to elect sixty-eight Democratic women to
Congress, thirteen to the U.S. Senate, eight to governorships, and
over 350 to other state and local offices.” Joint Appendix at 70.
16
The court suggests McConnell guarantees “‘interest groups’” the
right “‘to raise soft money to fund voter registration, GOTV
activities, mailings,’ and advertising.” Maj. Op. at 22 (quoting 540
U.S. at 187). But McConnell only notes BCRA can treat interest
groups differently than parties without violating Due Process. See
540 U.S. at 188. There is a difference between noting Congress has
not regulated and holding Congress cannot regulate. See N.C. Right
to Life, 525 F.3d at 333–34 (Michael, J., dissenting) (discussing
McConnell’s reference to “interest groups”).
 28
are the more natural path to “corruption,” McConnell, 540
U.S. at 297 (Kennedy, J., dissenting), but if my colleagues are
correct, Congress cannot regulate them the same way it
regulates county directors of animal control.17 Precedent is
plain: local officials must use hard money to attack federal
candidates, see id. at 184–85, but as the court now resolves
this case, multicandidate political committees cannot be so
limited. Can that be right?

The court’s opinion is boldly creative, and will, if
followed, have profound results on campaign finance
regulation. This case means:

1. Multicandidate political committees can spend
unlimited amounts of soft money to run ads attacking
or supporting federal candidates and political parties.
2. These committees can spend unlimited amounts of
soft money on get-out-the-vote activities that support
federal candidates and political parties.
3. These committees can solicit soft money by saying:
“Just like you, we want [federal candidate] to win.
You have already donated all the law allows to
[federal candidate], but there is no limit on how much
you can give to us to support [federal candidate].”
4. Congress can do nothing about any of this.

These results are in tension—perhaps irreconcilable tension—
with McConnell.

17
Though these regulations go further than BCRA § 323(f), by the
court’s opinion, it would not matter if they were exactly the same:
nonprofits are categorically distinct. See Maj. Op. at 22.
 29

C.

Recall how the Court in McConnell concluded its opinion:
“‘To say that Congress is without power to pass appropriate
legislation to safeguard an election from the improper use of
money to influence the result is to deny the nation in a vital
particular the power of self protection.’” 540 U.S. at 223–24
(quoting Burroughs v. United States, 290 U.S. 534, 545
(1934)). This “conviction” compelled the Court to uphold
“Congress’s most recent effort to confine the ill effects of
aggregated wealth on our political system” by
“control[ing]. . . soft money.” Id. at 224. But the Court was
“under no illusion that BCRA will be the last congressional
statement on the matter”—“[m]oney, like water, will always
find an outlet.” Id.

While I have argued courts should not unnecessarily assail
legislative acts, political speech is the core of what the First
Amendment protects. From Buckley to McConnell the Court
has relied on an ad hoc empiricism ill-suited to the complex
interactions of democratic politics. The government has
unlimited resources, public and private, for touting its policy
agenda. Those on the outside—whether voices of opposition,
encouragement, or innovation—must rely on private wealth to
make their voices heard. An increasingly anomalous
campaign finance jurisprudence only impoverishes this
essential debate. McConnell’s careless invocation of access
and influence (two integral aspects of political participation)
as synonyms for corruption is instructive. Such an expansive,
self-referential, and amorphous definition of corruption,
coupled with lax standards of scrutiny and a willingness to
accept as “evidence” any plausible theory of corruption or
claim of circumvention, is likely to doom any argument for
protection of core political speech. Someday the Supreme
 30
Court may be persuaded to reconsider this approach. But that
cannot be our task.

***

This should have been a straightforward application of
administrative law, not unlike countless agency cases decided
by this circuit every year. Congress has enacted a statute; the
agency has violated it; the rules must be vacated; done. I
would enforce the statute as written and call it a day. A good
rule of thumb is we often do more for the law when we do
less with the law. Per that rule, I concur only in part with the
court’s opinion.