Amicus Curiae Brief — McConnell v. Federal Election Commission

Supreme Court brief2003

Ask Donna

What actually matters in this document.

Text

j ~

/e

— a

No. 02-1674 & Consolidated Cases

IN THE

Supreme Court of the Anited States

|

MITCH MCCONNELL, e7 al., a

Appellant

OFFICE OF

Vv.

FEDERAL ELECTION COMMISSION, ef al..,

Appellees.

On Appeal from the United States District Court

for the District of Columbia

BRIEF OF AMICUS CURIAE THE LEAGUE

OF WOMEN VOTERS OF THE UNITED STATES

IN SUPPORT OF APPELLEES

LLOYD LEONARD DANIEL R. ORTIZ

LEAGUE OF WOMEN VOTERS Counsel of Record

OF THE UNITED STATES UNIVERSITY OF VIRGINIA

1730 M Street, N.W. SCHOOL OF LAW

Suite 1000 580 Massie Rd.

Washington, DC 20036 Charlottesville, VA 22903

(202) 229-1965 (434) 924-3127

Counsel for Amicus Curiae

WILSON-EPES PRINTING Co., INC. — (202) 789-0096 -— WASHINGTON, D.C. 20001

[ Sasseme Ca

ty] i

AUG 5 -

TABLE OF AUTHORITIES

TABLE OF CONTENTS

INTEREST OF AMICUS CURITAE...........scccssssssesseeeeeees

SUMMARY OF ARGUMENT ..............ccscssscssseseceeseees

FITTS ccscnssenssnmssciesnstanrennsnnnnsnnteseenuiienininnsnaaiano

I.

BCRA’S REGULATION OF ELECTION-

EERING COMMUNICATIONS, UNLIKE

REGULATION UNDER THE PRIOR

“MAGIC WORDS” TEST, PREVENTS

CORRUPTION AND THE APPEARANCE

OF CORRUPTION WITHOUT IMPER-

MISSIBLY BURDENING SPEECH ................

A. Disclosing Election Spending and Stop-

ping Corporations and Unions from

Spending From Their General Treasures to

Influence Federal Elections Represent Two

Central Safeguards Against Corruption and

the Appearance of Corruption in Federal

SID a cccnsscscnsesensnensmnnnsonnnieniebienianasdansiien

B. Identifying Advertisements Designed to

Influence Federal Elections Through

“Magic Words” Undermines These Two

Central Safeguards Against Corruption.......

THE BUYING TIME STUDIES ON WHICH

A MAJORITY OF THE DISTRICT COURT

RELIED IN UPHOLDING SEVERAL OF

BCRA’S CENTRAL PROVISIONS ARE

RELIABLE AND STRONGLY SUPPORT

BCRA’S PRIMARY DEFINITION OF

ELECTIONEERING COMMUNICATIONS ....

(i)

18

. ii

TABLE OF CONTENTS—Continued TABLE OF AUTHORITIES

Page CASES Page

A. A Majority of the District Court Clearly Buckley v. Valeo, 424 U.S. 1 (1976) v..cccscessesesseeeee 4, 6,8

Rejected the Plaintiffs’ Attacks on the Communist Party v. Control Board, 367 U.S. |

Buying Time Studie ......s..-cssseeeseessneeeeeneee IB RE 6

B. Plaintiffs Are Wrong in Claiming that the FEC v. Beaumont, 123 S.Ct. 2200 (2003) ............. 7,8

Authors of the Buying Time Studies First Nat'l Bank of Boston v. Bellotti, 435 U.S.

Improperly Recoded Data ...........:ssssessseeesses 23 765 Ee 8

! United States v. Auto. Workers, 352 U.S. 567

C. The Buying Time Studies Strongly | _ e 7

Support BCRA’s Primary Definition of

Electioneering Communications...............++. 26 STATUTES

ica icdinniteeainteieernnsesconceensnsnnee 29 Act of Jan. 26, 1907 (Tillman Act), ch. 420, 34

IE TITIT bassiciesesicibasditennnnnnanneniaenaneueenanemainnsensnseesn 7

Act of June 25th, 1910, ch. 392, 36 Stat. 822........ 6

Bipartisan Campaign Reform Act (BCRA) of

2002, Pub. L. No. 107-155, 116 Stat. 81

IIT inctsenssictuiiedienidinusidiidninetnsennninessensenncasssonesadsin 3

Federal Corrupt Practices Act, 1925, ch. 368, tit.

Ba i Sy HI cncecncncssscccencnsassesosessconss 6

Federal Election Campaign Act (FECA) of 1971,

PUD Ee. POD, FS- 22S, BG BAM, S ncccccccrccsccscescosccssess 6

OTHER AUTHORITIES

148 Cong. Rec. $2116 (daily ed. March 20,

STI ibihibihidehnethmactbaninsscesituiinieniianimeieminienmnnnnenpassese 9

' 148 Cong. Rec. $2117 (daily ed. March 20,

IIIT iidenntiaisiesdneebinienaniiedisepnedennctsnemnaieemnesteseneese 9

148 Cong. Rec. $2141 (daily ed. March 20,

ITI heitsiicsiadensebiseniinseinsidadencenitsacnsansnsaneeessestnsseenseceoce 9

Annenberg Public Policy Center, /ssue

Advertising in the 1999-2000 Election Cycle

I aidhinstiesiniestaieieestemiencieneanntenmninenneneantasseninscanen 21,22

Deborah Beck, et. al., Annenberg Public Policy

Center, /ssue Advertising During the 1996

SI TIT iii icisenidiandnsiniihansinethinnsinensonedens | 2

iV

TABLE OF AUTHORITIES—Continued

Erika Falk, Annenberg Public Policy Center,

Legislative Issue Advertising in the 107th

Congress (2003) ....cccsscrsesesesseesecesseosecescvsosssssoeues

Jonathan S. Krasno & Daniel E. Seltz, Brennan

Center for Justice, Buying Time: Television

Advertising in the 1998 Congressional

Bic titns Ce ccnctesecissenscsniinennsntenmencnasin

Leslie Wayne, A Back Door for the Conservative

Donor, N.Y. Times, May 22, 1997, at B10.......

Louis D. Brandeis, Other People’s Money

(National Home Library Foundation Ed.

DOI cccacececensccnsesctscdvessensinvatsimennmennnsmnensemestntnteitn

Senate Comm. on Governmental Affairs, /nves-

tigation of Illegal or Improper Activities in

_ Connection with 1996 Federal Election Cam-

paigns, S. Rep. No. 105-177 (1998) ........seeeeeeee

Page

IN THE

Supreme Court of the Gnited States

No. 02-1674 & Consolidated Cases

MITCH MCCONNELL, et ai.,

Appellants,

Vv.

FEDERAL ELECTION COMMISSION, et ai.,

Appellees.

On Appeal from the United States District Court

for the District of Columbia

BRIEF OF AMICUS CURIAE THE LEAGUE

OF WOMEN VOTERS OF THE UNITED STATES

IN SUPPORT OF APPELLEES

INTEREST OF AMICUS CURIAE'

The League of Women Voters of the United States (the

“League”) is a nonpartisan, community-based organization

that encourages informed and active participation of citizens

in government and seeks to influence public policy through

education and advocacy. The League is organized in more

than a thousand communities in every state and has more than

125,000 members and supporters nationwide. The League

' The Department of Justice has consented to the filing of this brief in a

letter which accompanies this filing. All other parties have filed written

consents to the filing of amicus briefs with the Court. No counsel for any

party authored this brief in whole or in part and no person or entity other

than the amicus curiae, its members, or counsel made any monetary

contribution to its preparation or submission.

2

was organized in 1919 as an outgrowth of the 72-year

struggle to win voting rights for women in the United States.

One of the League’s primary goals is to promote an open

governmental system that is representative, accountable, and

responsive, that protects individual liberties established by the

Constitution, and that assures opportunities for citizen

participation in government decision making. To further this

goal, the League has been a leader in seeking campaign

finance reform at the state, local, and federal levels for more

than two decades. The League thus has a strong and direct

interest in this case.

SUMMARY OF ARGUMENT

In late October of 1996, the following television adver-

tisement ran repeatedly in Arkansas:

Senate candidate Winston Bryant’s budget as Attor-

ney General increased seventy-one percent. Bryant has

taken taxpayer funded junkets to the Virgin Islands,

Alaska, and Arizona. And spent about $100,000 on new

furniture. Unfortunately, as the state’s top law enforce-

ment official, he’s never opposed the parole of any

convicted murderers. And almost 4,000 Arkansas pris-

oners have been sent back to prison for crimes com-

mitted while they were out on parole. Winston Bryant:

government waste, political junkets, soft on crime.

Deborah Beck, et. al., Annenberg Public Policy Center, /ssue

Advocacy During the 1996 Campaign 24 (1997). Winston

Bryant was running for an open seat in the United States

Senate. The tax-exempt organization that sponsored the ad,

Citizens for the Republic Education Fund, spent more than

$300,000 on a last-minute television blitz of Little Rock and

Jonesboro. Bryant lost. /d. Such advertisements come as

little surprise, of course. They are a powerful part of many

campaign strategies.

3

What is surprising is that until the enactment of the

Bipartisan Campaign Reform Act (BCRA), Pub. L. No. 107-

155, 116 Stat. 81 (2002), such advertisements were not

treated as electioneering speech under federal law. Despite

their appearance on the eve of the election, their unmistakable

mention of a particular federal candidate, their targeting of

the candidate’s electorate, and their appearance on television,

the courts would have treated such ads as “issue” advocacy.

By avoiding the use of “magic words”—express terms like

“vote for,” “elect,” “support,” “vote against,” “defeat,” and

“reject” —they fell outside of coverage. Even if they were

designed to promote a particular candidate, continuously

featured that candidate, swayed voters to the candidate’s

favor, and did not even mention a traditional issue, they were

treated as simple “issue” ads.

7” «66 7” «66

“Issue” advocacy thus became a peculiar term of legal art.

It indicated nothing about what an advertisement said or

concerned. It signified only that a few particular words were

missing. An ad could educate on a burning social issue or

shamelessly shill for a candidate. But whatever its content, so

long as no magic words appeared, it was completely immune

from the sort of reasonable source and disclosure rules that

have long applied to campaign advocacy.

By making the absence of a few particular words all that

mattered, the “magic words” approach undermined two of the

longest-standing and most central architectural features of

American campaign finance regulation: the disclosure of

political spending and the bar against direct corporate and

union spending to influence federal elections. By omitting a

few particular words, individuals and entities could spend

unlimited amounts to campaign for federal candidates without

ever having to disclose their activities. In the Bryant blitz, for

example, the citizens of Arkansas did not know the tax-

exempt sponsor’s identity, let alone the individuals or

corporations or unions that may have funded the tax-exempt’s

4

political activities. See, e.g., Leslie Wayne, A Back Door for

the Conservative Donor, N.Y. Times, May 22 1997 at B10.

Only after much research by the press did the identity of the

tax-exempt itself become clear. No research, however, was

able to uncover the identity of any individuals, corporations,

or unions that may have donated money to the tax-exempt for

these particular advertisements. As this Court has repeatedly

observed: “Publicity is justly commended as the remedy for

social and industrial diseases. Sunlight is said to be the best

of disinfectants; electric light the most efficient policeman.”

Buckley v. Valeo, 424 U.S. 1, 67 (1976) (quoting L. Brandeis,

Other People’s Money 62 (Nat'l! Home Library Found. ed.

1933)). But the absence of “magic words” bars all disclosure.

Without “magic words,” moreover, corporations and

unions could have used the tax-exempt as a conduit to run

this ad throughout the campaign season, or, indeed, have paid

for it themselves. So long as they did not use “magic words”

nothing stopped them from spending whatever they wanted to

advocate the election or defeat of particular federal candi-

dates. The addition of two words—‘vote against”—at the

end of the Bryant ad would have made all the difference. The

“magic words” test would certainly have prevented a corpo-

ration or union from funding that. But that is of little concern

to the corporate or union sponsor. If anything, using such

words makes ads less effective.

Understanding this history—indeed, documenting it in a

detailed factual record—Congress enacted BCRA, which

sought to return campaign finance regulation to steadier

footing. By applying an objective test which looked at

whether a clearly identified federal candidate was mentioned,

the medium, the targeting, and the timing of the commu-

nication, BCRA aimed to reach “issue” ads that looked the

same as—and presumably were as effective as—the ads that

candidates ran on their own behalf. This represented modest

change. It merely sought to subject to longstanding dis-

5

closure and source requirements those advertisements that

looked like the campaign ads candidates themselves run.

Such a small and sensible step would prevent circumvention

of this country’s basic safeguards against corruption and the

appearance of corruption.

A majority of judges on the district court found that sev-

eral empirical studies supported this modest step. Of these

studies, Plaintiffs challenge two: the Buying Time studies.

These particular studies surveyed political television adver-

tising from the 1998 and 2000 election cycles and concluded

that very few issue ads would have been mistakenly caught

by BCRA’s primary definition of “electioneering commu-

nication.” After carefully considering Plaintiffs’ general

claims that the Buying Time studies were biased and

untrustworthy, a majority of the judges properly rejected

those claims. Plaintiffs’ more detailed methodological claims

also fail. They complain first that some of the data under-

lying the Buying Time studies was improperly recoded. The

recoding they complain of, however, was approved by their

own expert and nothing suggests that the recoding they

themselves propose and on which they base many of their -

own numbers more accurately reflects the actual judgments of

the original coders, let alone the content of the ads

themselves. The contrary, in fact, is more likely true. Their

second complaint goes to the choice of metric employed by

one of the studies. Although they assert it is unhelpful to

overbreadth analysis, the opposite is the case. It strongly

demonstrates the care and caution Congress took in entering

this sensitive area.

6

ARGUMENT

I. BCRA’S REGULATION OF ELECTIONEER-

ING COMMUNICATIONS, UNLIKE REGULA-

TION UNDER THE PRIOR “MAGIC WORDS”

TEST, PREVENTS CORRUPTION AND THE

APPEARANCE OF CORRUPTION WITHOUT

IMPERMISSIBLY BURDENING SPEECH.

A. Disclosing Election Spending and Stopping

Corporations and Unions from Spending from

Their General Treasuries to Influence Federal

Elections Represent Two Central Safeguards

Against Corruption and the Appearance of

Corruption in Federal Elections.

Congress first required the disclosure of certain campaign

contributions in 1910, and has since expanded this require-

ment to a much wider range of campaign spending.” This

Court has recognized that such disclosure serves three

important interests that underpin the “free functioning of our

national institutions.” Buckley v. Valeo, 424 U.S. 1, 66

(1976) (quoting Communist Party v. Control Board, 367

U.S. 1, 97 (1961)). First, disclosure provides voters with

information about where money comes from and how it is

spent. Such information helps voters better evaluate

candidates and determine where they stand. /d. at 66-67.

Second, disclosure prevents actual corruption and the

appearance of corruption. By giving voters information about

the ties of elected officials to their financial supporters, it

deters contributions made with an eye to special favors and,

should any be made, allows voters more easily to detect them.

Id. at 67. Third, disclosure is “an essential means of

? See e.g., Act of June 25th, 1910, ch. 392, 36 Stat. 822; Federal

Corrupt Practices Act, 1925, ch. 368, tit. 3, 43 Stat. 1053, 1070-79;

Federal Election Campaign Act of 1971 (FECA), Pub L. No. 92-225, 86

Stat. 3.

7

gathering the data necessary to detect violations of .. .

contribution limitations,” id. at 68, which are independently

necessary to prevent corruption. In short, disclosure repre-

sents one of the most important means of securing the public

legitimacy of our political institutions because it provides

voters important political information, because it directly

deters corruption, and because it allows for the enforcement

of independent safeguards, like contribution limits.

Source requirements preventing business corporations and

unions from spending money from their general treasuries to

influence federal elections are even longer-standing. Since-

1907, Congress has carefully regulated corporate political

activity. In that year, Congress passed the Tillman Act,

which made it “unlawful for any national bank, or any

corporation organized by authority of any laws of Congress,

to make a money contribution in connection with any election

to any political office.” Act of Jan. 26, 1907, ch. 420 34 Stat.

864. Over the years, this rule was extended to cover both

contributions and independent expenditures by both business

corporations and unions. See, e.g., United States vy. Auto.

Workers, 352 U.S. 567, 570-84 (1957) (recounting history).

From the beginning, the law has focused on the “special

characteristics of the corporate structure that threaten the

integrity of the political process.” FEC v. Beaumont, 123

S.Ct. 2200, 2206 (2003) (internal citations and quotation

marks omitted). Because “[s]tate law grants corporations

special advantages . . . that enhance their ability to attract

capital and to deploy their resources in ways that maximize

the return on their shareholders’ investments . . . corporations

. .. [Can] use resources amassed in the economic marketplace

to obtain an unfair advantage in the political marketplace.”

Id. (internal citations and quotation marks omitted). As this

Court has long recognized, “barring corporate earnings from

conversion into political war chests . . . prevent[s] corruption

or the appearance of corruption,” id. (internal citations and

quotation marks omitted), the importance of which “has never

8

been doubted,” First Nat'l Bank of Boston v. Bellotti, 435

U.S. 765, 788 n.26 (1978). Barring corporate contributions

and expenditures also helps prevent the use of corporations

“as conduits for circumvention of valid contribution limits.”

Beaumont, 123 S.Ct. at 2207 (internal citations, brackets, and

quotation marks omitted). Together the disclosure and source

requirements barring corporate and union spending to

influence federal elections are two of the primary bulwarks

against corruption.

B. Identifying Advertisements Designed to Influ-

ence Federal Elections Through “Magic

Words” Undermines These Two Central Safe-

guards Against Corruption.

In Buckley, this Court considered the statutory meaning of

“expenditures . . . advocating the election or defeat of [a

clearly identified] candidate.” 434 U.S. at 41-51. In order to

avoid possible void-for-vagueness difficulties (in the case of

spenders other than candidates and political committees,

including political parties), this Court “limited [the term] to

communications that include explicit words of advocacy of

election or defeat of a candidate,” id. at 43, and in a footnote

“restrict{ed its] application . . . to communications containing

express words of advocacy of election or defeat, such as ‘vote

for,’ ‘elect,’ ‘support,’ ‘cast your ballot for,’ ‘Smith for

Congress,’ ‘vote against,’ ‘defeat,’ ‘reject.’”” Jd. at 44 n.52.

Nowhere did this Court indicate that such “magic words”

were constitutionally required for Congress ever to impose

disclosure and source requirements. Rather, this Court

narrowly held that given the uncertain reach of this particular

term, it had to be given more objective contours. Otherwise,

speakers would not know ahead of time what they could and

could not say. They would always be at the mercy of their

audience, for no matter how carefully they crafted their

speech to avoid advocacy, their audience could interpret it

differently. /d. at 42-44. The lesson of Buckley, in other

——

oe

9

words, is not that the First Amendment requires the use of

“magic words,” but rather that covered spending must be

defined objectively—just as BCRA’s primary definition of

“electioneering communications” does. Definitions that rely

largely on subjective audience understanding or speaker

intent are constitutionally problematic.

After carefully surveying the landscape of political

advertising, Congress found that the “magic words” test did

not work.” First, few modern campaign ads use any “magic

words.” Uncontroverted evidence before the district court*

showed that only 11.4 percent of the 433,811 advertisements

aired by candidates during the 2000 federal election met the

express advocacy test laid down in Buckley. Supp. App. at

659sa, 763sa-764sa (Kollar-Kotelly, J.). Second, candidates

and political consultants have found it very easy to evade

magic words in their advertising. In 1996, for example, the

Republican National Committee (“RNC”) ran the following

ad, called “The Story,” on television before the election:

Audio of We have a moral obligation to give our

Bob Dole: children an America with the opportunity and

values of the nation we grew up in.

Voice Over: — Bob Dole grew up in Russell, Kansas. From

his parents he learned the value of hard work,

honesty and responsibility. So when his

* See, e.g., 148 Cong. Rec. $2116 (daily ed. March 20, 2002) (state-

ment of Sen. Levin) (“magic words . . . are not a complete test of what

constitutes electioneering ads”); id. at $2117 (statement of Sen. Jeffords)

(“the ‘magic words’ standard . . . has been made useless”); id. at $2141

(statement of Sen. McCain) (“even a casual observer would concede that

‘magic words’ is a dramatically underinclusive test’).

* The filing of a Joint Appendix in this case has been deferred. Cita-

tions to the original record are made where necessary pursuant to this

Court's Rule 26.4(b). The opinions of the district court are reprinted in

the Joint Supplemental Appendix to Jurisdictional Statements, cited herein

as Supp. App. at Isa-1382sa.

10

country called . . . he answered. He was

seriously wounded in combat. Paralyzed, he

underwent nine operations.

Audio of I went around looking for a miracle that

Bob Dole: would make me whole again.

Voice Over: | The doctors said he’d never walk again. But

after 39 months, he proved them wrong.

Audio of He persevered, he never gave up. He fought

Elizabeth his way back from total paralysis.

Dole:

Voice Over: Like many Americans, his life experience

and values were as a strong moral compass.

The principle of work to replace welfare.

The principle of accountability to strengthen

_ our criminal justice system. The principle of

discipline to end wasteful Washington

spending.

Voice of It all comes down to values. What you

Bob Dole: believe in. What you sacrifice for. And what

you stand for.

Senate Comm. on Governmental Affairs, /nvestigation of

Illegal or Improper Activities in Connection with 1996

Federal Election Campaigns, S. Rep. No. 105-177, at 4014

(1998). “The Story” focused entirely on the personality traits

and affect of Bob Dole, not on any issues. Although the

RNC’s political director thought that it “need[ed] to be run,”

he worried that “[mJaking this spot pass the issue advocacy

test may take some doing.” /d. at 4015. He need not have

worried. As Bob Dole himself explained to Ted Koppel

when asked why such an ad would be considered “generic,”

i.e., issue advocacy, “It doesn’t say ‘Bob Dole for president.’

. . . [I]t talks about the Bob Dole story. It also talks about

issues. It never mentions the word that I’m—it never says

—™

—_— oe

that I’m running for president, though I hope that it’s fairly

obvious, since I’m the only one in the picture!” /d. at 4154.

The audience laughed. The absence of four little words made

all the difference.

Third, “magic words” make advertisements, if anything,

less effective. The uncontroverted testimony of political

consultants in the district court was that “magic words” were

clumsy and weakened political advertisements. Republican —

political consultant Douglas L. Bailey, for example, testified

that “[iJn the modern world of 30 second political adver-

tisements, it is rarely advisable to use such clumsy words as

‘vote for’ or ‘vote against.’ .. . . All advertising professionals

understand that the most effective advertising leads the

viewer to his or her own conclusion without forcing it down

their throat.” Supp. App. at 660sa (Kollar-Kotelly, J.).

Likewise, Democratic political consultant Raymond Strother

testified that

media consultants prefer putting across electioneering

messages without using words such as ‘vote for.’ Good

media consultants never tell people to vote for Senator

X; rather you make your case and let the voters come to

their own conclusions. In my experience, it actually

proves less effective to instruct viewers what you want

them to do.

Id. at 662sa. In this respect, political advertising is just like

commercial advertising. As two experts pointed out,

the practices of political advertisers are not dissimilar

from those of commercial advertisers. Car ads rarely

exhort viewers to “buy” a Chevrolet, nor do soft drink

ads urge people to “drink” their product. . . . This

atmospheric approach to commercial advertising—

where the product is presented in various tableaus—has

become increasingly popular. It serves the general

strategy of advertisers to present viewers with a variety

of reasons to choose their product, hoping that they will

latch onto one. Too heavy-handed an approach might

12

interfere with this process by raising viewers’ defenses.

Political ads seem to follow the same strategy, hoping

that citizens will grow to prefer a candidate without

being told to troop to the polls.

Id. at 663-64sa (quoting Krasno and Soraf Expert Report at

54). Like any good advertiser, those advertising candidates

avoid “magic words” because they want to make their pitch

as effective as possible.

The “magic words” test, then, ironically identifies as cam-

paign advertising those ineffective ads that candidates

themselves seldom use. It classifies as pure issue advocacy,

by contrast, exactly the kinds of effective campaign adver-

tising the candidates themselves do use. It gets the real world

backwards. In a sense, the “magic words” test works only to

protect political advertisers from making stupid decisions.

Because it compels disclosure of and keeps corporations and

unions from using treasury funds to engage in the least

effective forms of candidate advocacy, it encourages more

effective forms. Can that be the purpose of the First

Amendment?

By classifying nearly all campaign advocacy by indepen-

dent spenders as issue advocacy, the “magic words” test

clearly undermines ordinary disclosure requirements, which

give voters important information and prevent corruption. In

three 2000 Republican primaries, for example, an organi-

zation called “Republicans for Clean Air’ sponsored ads

praising then-Governor Bush and criticizing Senator McCain.

After the first of these primaries, reporters discovered that

“Republicans for Clean Air” consisted of two brothers,

Charles and Sam Wyly, who were long-term friends of

Governor Bush. One was, in fact, an authorized fundraiser

for the Bush campaign. According to the press, the Wylys

spent $2.5 million on the ads. When their identities were

uncovered in a later primary, much bad publicity ensued and

the Bush campaign felt compelled to distance itself from

them. In short, the “magic words” test allowed this major

13

advertising campaign to be conducted in secrecy and the

eventual public reaction shows how greatly the identity of

the sponsors mattered to voters. See Supp. App. at 97-99sa

(per curiam).

As political consultant Raymond Strother stated,

one of the biggest problems that a candidate’s media

consultant now faces is the lack of disclosure associated

with third parties running these ads. A few years ago,

Jill Docking ran for the United States Senate against

Sam Brownback in Kansas. . . . In the last two weeks of

a very tight election, an unidentifiable group came in and

poured a million dollars into the race. They ran [ads]

throughout Kansas telling voters that Jill wasn’t a

Christian, and all we could find was a fax machine. We

had no idea where the money came from. I have had

similar experiences in other races as well. Without

knowing who is funding the groups that run these ads,

we are often unable to correct the distortions.

Id. at 93sa-94sa. Without disclosure, voters cannot evaluate

candidates by seeing who supports them, cannot easily recog-

nize corruption in any special favors given to the funders of

these ads, and cannot tell whether the spending effectively

represents a circumvention of the contribution limitations.

Similarly, the “magic words” test allows corporations and

unions to evade the prohibitions against spending directly

from their general treasuries to influence federal elec-

tions. By carefully avoiding “magic words,” for example,

the National Rifle Association (“NRA”), a § 501(c)(4) tax-

exempt entity, successfully evaded the campaign finance

law’s source requirements. It sponsored much advertising to

influence the 2000 federal elections. Executive Vice Presi-

dent of the NRA, Wayne LaPierre, stated in a fundraising

letter that he “spent what it took [in 2000] to defeat Al Gore,

which amounted to millions more than we had on hand.”

Supp. App. at 693sa-694sa (Kollar-Kotelly, J.). He later

testified that “[w]Je took some money out of the reserves to

14

cover the deficit . . . . [The Gore advertising] was probably ...

the main contributing factor.” /d. The NRA ran several

different Gore ads—some funded out of its general treasury

funds and some funded by the NRA Political Victory Fund,

its affiliated political action committee. The small differ-

ences between some of the ads run by the NRA and those run

by its PAC show what little difference there is between

advertisements that do and do not use “magic words.” The

ads were virtually identical, The ones paid for from

the corporation’s general treasury simply omitted a few

express terms:

NRA PAC Advertisement NRA Direct (Non-PAC)

Advertisement

HESTON: Other issues

may come and go, but no

issue is aS important as our

freedom. And the day of

reckoning is at hand.

HESTON:

Did you know that right

now in federal court, Al

Gore’s Justice Department

is arguing that the Second

Amendment gives you no

right to own any firearm?

No handgun, no rifle, no

shotgun.

Did you know that right

now in federal court, Al

Gore’s Justice Department

is arguing that the Second

Amendment gives you no

right to own any firearm?

No handgun, no rifle, no

shotgun.

And when Al Gore’s top

government lawyers make it

to the U.S. Supreme Court

to argue their point, they

can have three new judges

handpicked by Al Gore if

he wins this election.

And when Al Gore’s top

government lawyers make it

to the U.S. Supreme Court

to argue their point, they

can have three new judges

handpicked by Al Gore if

he wins this election.

Imagine . . . what would

Supreme Court Justices

Hillary Clinton, Charlie

Schumer, and Dianne Fein-

stein do to your gun rights?

And what you _ think

wouldn’t matter any more.

Because the Supreme Court

has the final say on what

the Constitution means.

When Al Gore’s Su-

preme Court agrees with Al

Gore’s Justice Depart-

ment and bans private

ownership of firearms,

that’s the end of your Sec-

ond Amendment rights.

Please, vote freedom

first. Vote George W.

Bush for President.

ANNCR: Paid for by

the NRA Political

Victory Fund and not

authorized by any

candidate or candidate’s

committee.

15

Imagine . . . what would-

Supreme Court Justices

Hillary Clinton, Charlie

Schumer, and Dianne Fein-

stein do to your gun rights?

And what you _ think

wouldn’t matter any more.

Because the Supreme Court

has the final say on what

the Constitution means.

When Al Gore’s Su-

preme Court agrees with Al

Gore’s Justice Depart-

ment and _ bans private

ownership of _ firearms,

that’s the end of your

Second Amendment rights.

ANNCR: Paid for by the

National Rifle Association.

Id. at 695sa-696sa. (differences between ads bolded). NRA

Executive Vice President Wayne LaPierce described the two

scripts as “exactly the same” because, he admitted, the

reference to “the day of reckoning . . . at hand” in the NRA ad

was to the day of the 2000 election. /d. at 697sa.

16

Uncontroverted testimony in the district court showed,

moreover, that candidates track these advertisements and who

funds them. Lobbyist Wright Andrews testified that

[s]ophisticated political donors—particularly lobbyists,

PAC directors, and other political insiders acting on

behalf of specific interest groups—are not in the

business of dispensing their money purely on ideological

or charitable grounds. Rather, these political donors

typically are trying to wisely invest their resources to

maximize political return. . . . [Issue ads] are noticed by

the elected officials on whose behalf, or against whom,

[they] are run. An effective advertising campaign may

have far more effect on a member than a direct campaign

contribution or even a large soft money donation to his

or her political party that is used for political purposes in

his or her district or state.

Id. at 709sa-710sa. So helpful are these ads, in fact, that

politicians who benefit from this type of corporate and union

general treasury fund spending often raise money for these

organizations. /d. at 712sa. Congressman Ric Keller, for

example, for whose 2000 campaign the Club for Growth had

funded advertising, signed a Club for Growth fundraising

letter which stated: “The Club for Growth selected my race as

one of its top priorities. .... Since the Club targets the most

competitive races in the country, your membership in the

Club will help Republicans keep control of Congress.” Id.

(emphasis in original).

_ Although the number of entities funding these advertise-

ments has grown, only a handful account for most of the

influence. In the 1999-2000 election cycle, for example, the

Republican and Democratic parties accounted for 31 percent

of this spending; Citizens for Better Medicare, 13 percent; the

Coalition to Protect America’s Health Care, 6 percent; the

U.S. Chamber of Commerce, 5 percent; AFL-CIO, 4 percent;

NRA, 4 percent; and U.S. Term Limits, 4 percent. /d. at

655sa. Just the six leading nonparty entities accounted for 36

17

percent of the spending—more than the parties themselves

(31 percent) and more than all other organizations combined

(33 percent). Two of these six entities, in fact, represented

the same type of special interest. Far from being “a grass-

roots generated group of citizens,” Citizens for Better

Medicare was “an arm of the Pharmaceutical Research and

Manufacturers Association” and the ‘Coalition to Protect

America’s Health Care worked under the umbrella of the

American Hospital Association, Erika Falk, Annenberg

Public Policy Center, Legislative Issue Advertising in the

107th Congress 62, 64 (2003) (internal citations omitted). As

these figures show, far from broadening the spectrum of

political debate, the ability to spend from general treasury

funds constricts it. It concentrates the threat of corruption.

The “magic words” test, then, has undermined two of the

central protections against corruption and the appearance of

corruption—the disclosure of campaign spending and the bar

against funding electioneering activity from general union

and corporate treasuries—that FECA was intended to support.

The test paradoxically allows corporations and unions to use

their treasuries to fund only the kinds of advertisements that

candidates now usually employ to promote their own

candidacy—those without “magic words.” And it dis-

courages only the kinds of advertisements that candidates

themselves now seldom use because they think them

ineffective—those with “magic words.” The test, in short

allows corporate earnings and union dues to be easily

converted into political “war chests,” the influence of which

is often invisible to voters.

18

Il. THE BUYING TIME STUDIES ON WHICH A

MAJORITY OF THE DISTRICT COURT

RELIED IN UPHOLDING SEVERAL OF

BCRA’S CENTRAL PROVISIONS ARE RELI-

ABLE AND STRONGLY SUPPORT BCRA’S

PRIMARY DEFINITION OF ELECTIONEER-

ING COMMUNICTATIONS.

A. A Majority of the District Court Clearly

Rejected the Plaintiffs’ Attacks on the Buying

Time Studies.

Plaintiffs rest much of their attack on BCRA’s handling of

“electioneering communications” on the supposed unreli-

ability of parts of two particular reports. These reports, the

Buying Time studies, surveyed television ads from the 1998

and 2000 election cycles and concluded that very few issue

ads would be affected by BCRA’s primary definition.

BCRA’s primary definition would, in the view of these

studies, leave the vast majority of actual issue ads untouched.

The 1998 study, for example, concluded, in part, that only 7%

percent of all issue ads televised that year met BCRA’s

primary definition. Jonathan S. Krasno & Daniel E. Seltz,

Brennan Center for Justice, Buying Time: Television Adver-

tising in the 1998 Congressional Elections, 109 (2000); see

also Supp. App. at 1348sa n.213 (Leon, J.) (noting that Dr.

Krasno’s current estimate is 6.1%).

Plaintiffs understandably attack these studies because they

stand so powerfully against them. Among other things,

Plaintiffs claim that “the district court, after detailed and

laborious consideration, unanimously rejected the conclusions

of the Buying Time reports,” McConnell Br. at 53, that “the

Buying Time reports were entirely and irredeemably biased,

id. at 53 n.18, and that “the reports’ conclusions about

BCRA’s insignificant effect on so-called ‘genuine’ issue ads

are insupportable,” id. at 55. Nothing could be further from

the truth.

19

In their individual opinions, two of the three judges on the

district court strongly rejected Plaintiffs’ broad-brushed

attacks. As Judge Leon put it, “I find that although the

Buying Time studies contain some flaws and shortcomings,

... those shortcomings do not detract from the studies’

credibility and reliability.” Supp. App. at 1335sa. Judge

Kollar-Kotelly went even further. She described Plaintiffs’

expert's attempts to discredit the studies as “a pifiata party: if

one hits the pifiata enough, it will eventually crack apart.” /d.

at 764sa. In the Plaintiffs’ case, however, the expert clearly

failed to crack the pifiata. Like Judge Leon, Judge Kollar-

Kotelly extensively relied on the studies’ central findings.

She also thought the Plaintiffs’ failure to do any independent

studies or even to try to replicate the study they criticized

highly significant:

Although some of [the pifiata] “hits” have merit, . . .

neither Plaintiffs nor Dr. Gibson[, their expert,] have

attempted to conduct their own similar study, or even

replicate a discrete portion of the Buying Time studies,

despite the fact that the underlying materials were

provided to them by Defendants. Presenting the Court

with contradictory results from such a study would have

been far more persuasive than the recalculations of

incorrect versions of the Buying Time data sets and the

often conjectural and speculative criticism proffered by

Plaintiffs and Dr. Gibson.

Id. In fact, the Plaintiffs’ complete failure to offer their own

studies puts them in an awkward position. Although they

continually attack the Buying Time reports, they find them-

selves having to rely extensively on the underlying data—

quarrelling with it in some places, changing it in others, but

all the time using it to support their own arguments on

overbreadth. See, e.g., McConnell Br. at 56 (massaging data

to argue that “64% of ads mentioning a candidate in the 60

days before the 1998 general election were coded . . . as

20

genuine”). Their own extensive reliance on the studies

undermines their broadside attacks.

A majority of the judges on the district court also strongly

rejected any claim of bias. Judge Leon, for example, found

that “while . . . the primary purpose of the Buying Time

studies was to further campaign finance reform, I do not find

that this fact has skewed the results of the study.” Supp. App.

at 1335sa. Judge Kollar-Kotelly was even more forceful. “I

would not,” she found, “discount the [Buying Time] studies

because they were approached with a particular result in

mind. The testimony shows that policy perspectives and

effective scientific research are not mutually exclusive.” /d. at

764sa. And, as to the particular actions which Dr. Gibson

pointed to as evidence of bias—the “cleaning” of data and

recoding—Judge Kollar-Kotelly correctly noted that “(t]he

‘cleaning’ of the data that Dr. Gibson finds suspicious

appears . . . to be a necessary function for databases of the

size produced for the Buying Time reports and not a function

of bias. Fixing miscodings and resolving the ‘cookie cutter’

issues required such actions.” /d. at 764sa-765sa. Judges

Leon and Kollar-Kotelly both noted, moreover, that one of

Dr. Gibson’s major complaints with the studies, his inability

to reproduce their conclusions from the underlying data sets,

resulted from his own use of the wrong data set. /d. at 1336sa

(Leon, J.) (“[W]hile Dr. Gibson maintains that his inability to

replicate the Buying Time . . . results ‘undermines . . . any

confidence one should place in the findings,’ his inability

seems attributable to his using the incorrect data set.”); id. at

765sa (Kollar-Kotelly. J.) (“confusion . . . as to the correct

database to use to analyze the studies’ findings decreases the

utility of Dr. Gibson’s Expert Report”) (internal citations

omitted). Judge Kollar-Kotelly, in fact, found that this

confusion also “undermines the notion that the Buying Time

authors manipulated the data in order to achieve their desired

results.” Jd. at 765sa. The charge of bias is simply empty

of merit.

2)

As a majority of judges pointed out, moreover, even if the

Plaintiffs’ attacks were all credible, they would leave many of

the conclusions of the Buying Time studies untouched. See

Supp. App. at 1333sa-1334sa (Leon, J.); id. at 765sa (Kollar-

Kotelly, J.). First, as Dr. Gibson himself admitted, “Entirely

objective characteristics of the ads . . . present few threats to

reliability.” Gibson Expert Report at 11. These include such

central findings as (1) that “magic words” are rarely used in

political advertising, (2) that group-sponsored ads that men-

tion candidates tend to be concentrated before an election,

and (3) that advertisements sponsored by parties and interest

groups comprise a significant and increasing portion of

political advertising broadcast in federal races. Supp. App. at

1 333sa-1334sa (Leon, J.).

Second, Plaintiffs failed to challenge the validity of other

authoritative studies, particularly the Annenberg studies,

which support the central findings of the Buying Time reports.

Id. at 763sa (Kollar-Kotelly, J.) (“The Annenberg studies . . .

have not been challenged by anyone.”). Even if Plaintiffs did

manage to discount the Buying Time studies, these others

would support many of the same conclusions. In particular,

the Annenberg studies concluded that in the 1999-2000

election cycle, candidate campaign ads virtually dominated

all types of issue ads in the last two months of the election:

The type of issue ad that dominated depended greatly

on how close we were to the general election. .. .

Though candidate-centered issue ads always made up a

majority of issue ads, as the election approached, the

percent [of] candidate-centered spots increased and the

percent of legislative and image ads decreased, such that

by the last two months before the election almost all

televised issue spots made a case for or against a

candidate.

Annenberg Public Policy Center, /ssue Advertising in the

1999-2000 Election Cycle 14 (2001), quoted and emphasis

added in Supp. App. at 654sa (Kollar-Kotelly, J.). The

22

Annenberg studies, in fact, produced a bottom line for the

2000 election remarkably close to the one the Buying Time

reports produced for the 1998 election: “Fully 94% of issue

ads aired in after August[, 2000] made a case for or against a

candidate. Just 3.1% were legislative [issue] ads, and 2.3%

were general image ads.” /d., quoted in Supp. App. at 726sa

(Kollar-Kotelly, J.).

The findings of the uncontroverted Annenberg studies not

only bolster the conclusions of the Buying Time reports, but

also underscore the perversity of the “magic words” test.

Among the conclusions of the Annenberg studies were ( I) the

amount of money spent on “issue advocacy” is rising rapidly;

(2) instead of creating the number of voices Buckley v. Valeo

had hoped, issue advocacy allow[s] groups such as the

parties. business and labor to use soft money or general

treasury “war chests” to gain a louder voice; (3) the

distinction between issue advocacy and express advocacy Is a

fiction; and (4) “issue” advocacy, as practiced in recent

elections, masks the identity of some key players and so

deprives citizens of information about source of messages

which research tells us is a vital part of assessing message

credibility. /d. at 2, quoted in Supp. App. at 657sa (Kollar-

Kotelly, J.). In addition, the unrebutted findings of Dr.

Goldstein about the 2000 election showed that (1) interest

group advertising that mentioned candidates was highly

concentrated in battleground states; (2) no more than 12% of

candidate-sponsored advertisements in 2000 used “magic

words”; and (3) interest group advertisements that identified a

candidate in 2000 tended to be broadcast within the final 60

days of the election campaign, whereas those that did not

identify a candidate were spread more evenly throughout the

year. See Supp. App. 763sa-764sa (Kollar-Kotelly, J.). The

authoritative studies Plaintiffs did not challenge, then, support

BCRA’s overall approach to electioneering communications

just as strongly as do the studies they do attack. Even Judge

Henderson, who criticized the Buying Time studies, relied on

ee ee

23

these other studies in her opinion. See Supp. App. at 230sa-

231sa, 233sa.

B. Plaintiffs Are Wrong in Claiming that the

Authors of the Buying Time Studies Improperly

Recoded Data.

In the district court, Plaintiffs made one central challenge

to the integrity of the 1998 Buying Time study. They claimed

that the authors had improperly recoded data to recharacterize

eight “genuine” issue ads as campaign ads. They still do. See

McConnell Br. at 56. Judge Leon, however, who of the three

judges considered the Buying Time studies most closely,

found no problem here. See Supp. App. at 1344sa. He

rightly saw that “the changes [made] in the database reflect

the gradual filling in of missing data and the discovery of

internal contradictions.” /d. What happened was that the

original coders who had reviewed these eight ads had filled

out their coding sheets in a way that was_ internally

contradictory. Plaintiffs’ expert Dr. Gibson acknowledged as

much. Gibson dep. at 116-21. Since it was impossible from

the original coding sheets to tell how the coders would have

properly classified these ads, they were recoded. Dr. Gibson

agreed that that was the proper course. In fact, when asked

what he would have done under the circumstances, he

answered directly “I probably would have fired [the coder]

and had the cases re-coded,” id. at 121, which—apart from

the firing—is what the authors of the 1998 study did. When

asked to review several of these internally contradictory

coding sheets along with the ads they coded, Dr. Gibson

himself indicated that it was “highly probable” that the

original coders intended to categorize the ads as campaign

advertising, not as issue ads. /d. at 112. Dr. Gibson, then,

unlike the Plaintiffs themselves, did not believe that these

eight ads should be counted as issue ads, an assumption upon

which one of the most dramatic and misleading arguments of

the Plaintiffs rest: “that 64% of ads mentioning a candidate in

24

the 60 days before the 1998 general election were coded by

the students as ‘genuine. McConnell Br. at 56. That

number cannot stand in the face of the record and Plaintiffs’

own expert's testimony.

Plaintiffs also attack the coding of the data underlying the

Buying Time 2000 study. They incorrectly claim that 17%,

not 2.3%, of the television ads appearing within 60 days of

the general election were actual issue ads. To get there, they

must recode six of seven basically identical ads that were

widely broadcast in the 2000 campaign. This is improper. In

his expert report analyzing the Buying Time 2000 study's

methodology and use of data, Defendants’ expert Professor

Goldstein explained that one of these seven ads was

originally coded by the students_as an issue ad and the

remaining six essentially identical ads were coded by them as

campaign ads. See Goldstein Rebuttal Report at 16-17.

Although the original Buying Time 2000 investigators had

then recoded the one ad to match the six others, Professor

Goldstein had followed the original inconsistent coding in his

analysis. But the Plaintiffs argued in district court that the six

should be made to conform to the one. See Pls.’ D. Ct. Reply

Br. at 37. The Defendants and Intervenors, on the other hand,

believed it more sensible to reclassify the one outlier rather

than the large majority of advertisements. They did not

object, however, to Dr. Goldstein’s adhering to the original

inconsistent coding, the approach that led to the figure of

2.3%. Although in deposition Plaintiffs pressed Professor

Goldstein to reclassify the majority of ads, he agreed only

that under the Plaintiffs’ version of “the most conservative

estimate” the six should be made to conform to the outlier.”

* Protessor Goldstein's deposition testimony proceeded as follows:

Q: [I}f you believe [the six ads] are virtually indistinguishable

{from the one other], then wouldn't it have been fairer to

include them in your analysis, your quote ‘conservative

analysis’ in undertaking to prepare your expert report?

25

He nowhere stated that recoding six ads to conform to one

would better reflect the ads’ content or his own “conservative

estimate.” He merely confirmed that under “the most

conservative estimate” any doubt—no matter how small—

would place the ads in the genuine issue category. And it

is this “most conservative estimate” that produces the figure

of 17%. Supp. App. at 1353sa-1354sa, 1361sa (Leon, J.).

If anything, controversies over recoding reflect bias in the

Plaintiffs’ challenges to the Buying Time studies, not in the

studies themselves. At one point in his report Dr. Gibson

relied on a modified version of the 2000 database in which 30

ads had been selected for recoding from campaign ads to

issue ads by Plaintiffs’ counsel. See Gibson Expert Report

62-63. Based on that recoding, Plaintiffs argued that the

actual number of issue ads BCRA’s primary definition would

have covered “would have increased by . . . 23%.” Pls.’

Omnibus Op. Br. (McConnell Pls.) at 69 n.34. But Dr.

Gibson himself never understood why those ads should be

recoded. He received no explanation from counsel and he

could not see any reason himself.° Indeed, Dr. Gibson never

A: In retrospect, knowing that these [six] ads were originally

coded as genuine issue ads [—a counterfactual assumption—}

they should have been included in the most conservative

estimate .. . . Having said that, . . . | don’t know what they

were originally coded as.

Goldstein dep. at 160 (emphasis added).

* His deposition testimony makes this clear:

Q: Now, you refer... to this mysterious group of 30 ads and

[your report] says, “[t)his list was provided to me by counsel.”

What I wanjt] to know is how were these ads selected by

counsel for you?

A: Idon’'t know.

** *

26

even looked at the storyboards for the recoded ads. /d. at

181. He simply followed the directions of counsel.

C. The Buying Time Studies Strongly Support

BCRA’s Primary Definition of Electioneering

Communications.

Because of a slight change of focus, the 1998 and 2000

Buying Time studies employ somewhat different measures to

quantify BCRA’s fit. The studies compare a single number,

the numerator, to two other numbers, the different denom-

inators. The common numerator consists of the number of

pure issue ads mischaracterized as campaign advocacy under

BCRA’s primary definition of electioneering communica-

tions. The 1998 study compares this numerator to the overall

number of issue ads televised in that year and thus offers up a

ratio of how many “false positives” using BCRA’s primary

definition would create. For example, if BCRA’s primary

definition mistakenly tagged four issue ads as campaign ads

out of 100 issue ads televised that year, the measure would

produce 4/100 or four percent. For 1998, this measure

produces a result of 6.1% false positives, Supp. App. at

1348sa n.213 (Leon, J.); for 2000, it produces a result of

| Were] these ads ... selected to illustrate that changing the . . .

coding on relatively few ads can produce a relatively large

swing in the percentages?

Pebvcds

I take it you didn’t make any determination that these 30 ads

should have been coded as genuine issue ads; is that correct?

That's correct.

Do you know now (sic) the Plaintiffs’ lawyers went about

selecting the ads .. .?

No, I don’t.

Did you have any understanding of the way in which the 30

ads were selected?

A: No.

Gibson dep. at 179-81.

©

QF 22 LC

27

3.1%, id. at 1360sa. These numbers are uncontested by

Plaintiffs, who rest their arguments on the other measure.

The 2000 study takes a somewhat different approach. It

compares this same numerator to the total number of ads

covered by BCRA’s primary definition. This approach

provides a measure of the proportion of ads BCRA’s primary

definition classifies as campaign ads that are actually pure

issue ads. It measures not “false positives” but rather the

inaccuracy of BCRA’s primary definition just within the

relatively small universe of ads it tags. Although both

measures use the same numerator—the number of issue ads

mischaracterized as campaign advocacy by BCRA’s primary

definition—they employ quite different denominators and

thus lead to different results.’

Contrary to Plaintiffs, this denominator issue is no

“problem” at all. These two different measures answer two

’ For 1998, the Defendants and Judge Leon believe that this second

measure produces a result of 14.7%, while Plaintiffs, because they would

recode eight campaign widely televised ads as issue ads for this year,

believe it would produce a result of 64%. Compare Supp. App. at 1157sa

with McConnell Br. at 56. As discussed earlier, however, their argument

is specious. Supra at 23-24. When asked to review several of these ads

and their internally contradictory coding sheets, Dr. Gibson, Plaintiffs’

own expert, indicated that it was “highly probable” that the coder

intended to categorize the ads as candidate campaign advertising. Gibson

dep. at 112.

For 2000, the Defendants believe this same measure produces a result

of 2.3%, while Plaintiffs and Judge Leon believe it produces a result of

17%. Supp App. at 1157sa. As discussed earlier, however, this higher

figure takes seven essentially—identical ads, six of which the original

coders characterized as campaign advocacy and one of which they

characterized as issue advocacy, and characterizes all seven as issue

advocacy. Supra at 24-25. At best the 17% figure can serve only as an

upper limit. Accepting the original inconsistent coding, which itself errs

on the side of caution, produces the figure Defendants tendered: 2.3%.

The more reasonable approach of reclassifying the one outlier to conform

to the other six ads would, of course, produce a still lower figure.

28

different questions about BCRA's effect. The original, 1998

measure compares the number of issue ads covered by BCRA

to all issue ads televised in that year and so measures the

impact of BCRA on issue advertising generally. The later

measure compares the number of issue ads covered by BCRA

to all ads covered by BCRA and so measures the inaccuracy

of the primary definition within the restricted domain of those

ads it captures. Unlike the first measure, it says nothing at all

about the accuracy of the primary definition in the total

domain of political advertising.

Of these two measures, only the first bears on what

proportion of total issue advocacy is mistakenly caught by

BCRA’s primary definition. On any view, that is a very

small number—6.1 percent for 1998 and 3.1 percent for 2000.

This ratio, however, actually overstates the impact of BCRA

on issue ads for several reasons. First, the coders were given

a stark all-or-nothing choice—the ads had to be classified

simply as issue or candidate ads, not some mixture of both.

Hybrid ads touting or attacking candidates on particular

issues could thus have been coded as pure issue ads even if a

major purpose was electioneering. Second, the coders had no

context for the ads. An ad that appears on its surface to be

about issues can assume a quite different role in the context

of an actual election. An ad attacking trial lawyers, for

example, might not have coded as a candidate ad even if in a

particular race that ad had been aimed at one candidate

prominently identified with that profession. Third, the study

did not consider print ads and mass mailings, which BCRA’s

primary definition excludes from coverage. Had they been

included in the overall analysis, however, the proportion of

issue advocacy covered by BCRA’s primary definition would

have fallen greatly. The same number of televised issue ads

would have been coded as falling under BCRA’s primary

definition, of course, but they would have been among the

much larger universe of televised, print, and mass mailing

issue ads. For these reasons, then, the already low figures

29

that the 1998 measure produces for mischar-

acterized issue advocacy are upper limits. The actual

numbers would be much lower. As the Buying Time studies

show, BCRA’s primary definition in no way impermissibly

burdens speech.

CONCLUSION

For the foregoing reasons, this Court should uphold

BCRA’s primary definition of electioneering commu-

nications.

Respectfully submitted,

LLOYD LEONARD DANIEL R. ORTIZ

LEAGUE OF WOMEN VOTERS Counsel of Record

OF THE UNITED STATES UNIVERSITY OF VIRGINIA

1730 M Street, N.W. SCHOOL OF LAW

Suite 1000 580 Massie Rd.

Washington, DC 20036 Charlottesville, VA 22903

(202) 229-1965 (434) 924-3127

Counsel for Amicus Curiae

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.

A word about cookies

We need a few to keep you signed in and the library working. The rest help us see which pages people use and where they get stuck. They stay off unless you say yes.