MIT LIBRARIES
3 9080 02617 6906
\ *
* /
2
(
LIBRARIES
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium
Member
Libraries
http://www.archive.org/details/didfirmsprofitfrOOanso
5
1
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
Did Firms Profit from Soft
Money?
Stephen Ansolabehere
James M. Snyder,
Michiko Ueda
Jr.
Working Paper 04-1
January 2004
Room
E52-251
50 Memorial Drive
Cambridge,
MA 021 42
This paper can be downloaded without charge from the
Social Science Research Network
Paper Collection
3062
http://ssrn.com/abstract id=51
at
J
MASSACHUSETTS INSTITUTP
OF TECHNOLOGY
__
-'8RARIES
DID FIRMS PROFIT
FROM SOFT MONEY?
Stephen Ansolabehere
Department of Political Science
MIT
James M. Snyder,
Jr.,
Departments of Political Science and Economics
MIT
Michiko Ueda
Department of Political Science
MIT
January, 2004
The authors thank William Burke
for his excellent research assistance. Professor
and Professor Snyder thank the National Science Foundation for
its
Ansolabehere
generous financial support.
At
its
heart, the Bipartisan
firms receive from
Campaign Reform Act seeks
A series of rulings by the Federal Election
campaign contributions.
Commission created
to limit the private benefits that
the opportunity for organizations and individuals to give funds to party
accounts outside the system of direct contribution limits - so called soft money. Although
used before 1992, soft
money ballooned
parties raised approximately
Federal Elections
Campaign
money, most of which came from corporations
in
times larger than the hard contribution limits set in the
Act.
to
have profited directly and substantially from
their soft
donations. In the years leading up to the passage of BCRA, public interest groups and the
press provided
soft
in soft
at least ten
Companies were widely alleged
money
during the 1990s. In the 2000 election, the two major
$500 million
donations in excess of $ 1 00,000,
little
numerous examples of firms
money donors -
that benefited
tobacco, pharmaceutical, and
reports.
The most
emerged
in hearings before the U.S. Senate
telling evidence, cited extensively
executives and legislators testified that soft
government contracts where on the
contribute, then their
line.
companies were especially featured
by
the majority opinion in
Committee on Commerce
money
in 1998.
donations were often given
in these
McConnell
v.
FEC,
Corporate
when
valuable
In other cases donors feared that, if they did not
companies would lose competitive advantages through regulations.
These were clearly
stories
of excess. Parties and
excess of what they could raise under the hard
the soft party
oil
from public policies and were also large
money
money, allegedly received excessively
limits,
and corporations, which gave most of
large benefits at public expense in return for
their contributions.
See the report by the Federal Election Commission
their candidates received donations well in
at:
http://www.fec.gov/press/051501partyfund/051501partyfund.html.
Were
these cases typical, or exceptional? If exceptional, then the government might best
deal with the problems of corrupt practices through aggressive enforcement of anti- bribery laws. If
typical, then the
contributions
-
government might attempt
is
the extent to
money. Economists and
which donors, especially
political scientists
campaign contributions on public policy.
money
problems with blanket
restrictions
on
as they in fact did.
At issue
hard
to eliminate these
from
large corporate donors, benefited
soft
have long been puzzled about the influence of
An
extensive literature examines the association between
contributions and public policy decision- making, especially roll call voting in the U.S.
Congress. The large majority of studies find no significant effects of hard
public policy, and, in those that do find
some
association, the
money
magnitude of the
contributions on
effects
is
typically
very small.
More
much
troubling
influence.
still,
the total
amount of campaign contributing seems too small
Tullock (1973) observed that although corruption
Assuming a reasonable
plausibly large.
services that firms
buy with
their
On
amount
lot,
but
it
is
to a significant societal
the other hand, such calculations
(e.g.,
donations?
is
not
goods and
several
hundred
problem.
may be wrong. Under some
command an
it
produce
rounding error on the national
assumptions about the
extraordinarily high return on
Persson and Tabellini, 2002, pp. 187-190; Dal Bo, 2002).
empirical studies that claim to find evidence for such high returns
Is there
all
campaign contributions cannot be more than a
nature of political bargaining, companies might
investment
widely alleged,
return on investment, the total value of
million dollars per year. That might sound like a
accounts, and likely does not
is
to
(e.g.,
And
Stratmann, 1991).
evidence that firms profited substantially and systematically from their soft
We know of no
few
there are a
money
study that has looked for a systematic relationship between soft
See Ansolabehere, de Figueiredo, and Snyder (2003) for
a
summary of this
literature.
money
We address this question here, by examining
donations and policy decisions or outcomes.
corporate stock returns.
The
Bi-Partisan
Campaign Reform Act
itself provides a lens
through which to observe the
value that firms' derived from soft money. If companies profited from soft money, then investors
in firms should
have valued the firms accordingly. Companies that gave soft money and received
undue competitive advantages or large contracts
recent years than companies that gave
new
have been better investments
in
money. By the same reasoning, the imposition of
regulations and the Court's decision to uphold those regulations should have lowered the value
of the firms
that
used the soft
money
Consider a typical Fortune
average,
$50
billion,
contributions (not
this
little sofi
in return should
all
1
loopholes for profit.
00 company. Annual revenues for these companies
10% of which
roughly
is profit.
Large firms that give
soft
do) give an average of about $500,000 over two years.
are,
on
money
An
excellent return
on
investment would double the amount invested. This would account for just one one-hundredth
of one percent of the company's two-year
about.
The
fear,
however,
is
that
profit
-
difficult to notice
of contracts and services, and $100 million
and would account for
1
in profits.
donations yields $1 billion worth
This would represent a 20,000 percent
eliminate this stream of profit. In the
on investment are more modest and where
relatively small, the effect
money
percent of a company's annual profit.
The elimination of soft money would
returns
to get excited
companies receive returns thousands of times larger than the
investment. Suppose, for example, that $500,000 in soft
return,
and not much
the total value
latter case,
case,
where
of goods and services bought
on a company's stock price would be negligible -
one-hundredth of one percent. In the
first
where the returns
in the
is
range of one
are exceptionally large and the
cost of soft
money
money would be
We
to
to
government and society might be
substantial, the effect
lower the stock value of the hypothetical firm by about
1
of banning
soft
percent.
can examine the effects of BCRA using stock market data and standard event study
methodology.
3
Previous papers by Roberts (1990a, 1990b), Fisman (2001), Jayachandran (2002),
and others have found
that political events
- such
as the death
1983 and Senator James Jeffords' party switch in 2001
of Senator Henry (Scoop) Jackson
- can have
in
a noticeable impact on stock
prices.
One
of an event
important prerequisite for conducting an event study
that releases
regard because
10, 2003.
its
the
we know precisely the
new
exactly
willing to
information into the market.
date of the
BCRA as
it
We are especially
until the
very
was
to
is
uphold
4
it.
was about
deeply divided over
the following,
how
to
approach campaign finance regulation...
it
all,
a description
that the Justices
clear
was not even
5
98 1 ) for
is difficult
as likely to strike
was not
It
standard would be used to judge the challenged provisions of BCRA."
1
it
to
down
by Professor Michael C. Dorf of the Columbia University Law
lengthy oral argument which of these views will prevail. Indeed,
(
the court revealed
An example of the tentative commentary offered by campaign
School: "The four-hour oral argument in McConnell indicated, above
See Schwert
moment
"surprise" the decision was, the fact that almost no observers were
predictions suggests that they believed the court
finance law experts
As
fortunate in this
information was clearly released to the market that day. While
how much of a
make
the ability to determine the date
Supreme Court's decision on BCRA: December
Moreover, because the outcome was uncertain
decision,
know
new
is
The day
from the
clear
what
legal
after the decision,
of the method and a survey of papers employing it.
BCRA and Ansolabehere on the side
consultants on this case (Snyder on the side opposed to the
remain
in
support of it),
two of the authors had detailed knowledge of the proceedings and followed the litigation closely. Both thought the
plaintiffs were more likely to prevail.
5
Quoted from an article on the CNN web site, September 19, 2003, "The Supreme Court's campaign finance reform
argument." The article was found at http://images.cnn.com/2003/LAW/09/19/findlaw.analysis.dorf.campaign.finance/.
campaign finance expert Thomas Mann
said: "Yesterday, [the
decision that surprised many, although, certainly, not
all
Supreme Court reached] another
members of this
6
of its findings on the Bipartisan Campaign Reform Act."
clarity
The
this
view.
oral
arguments
final vote
on BCRA's
soft
money
provisions
was
panel, in the reach and
Other pieces of evidence support
as close as possible, 5 to 4.
September, Justice Rehnquist, thought to be pivotal on
in
5:4
At the
this matter, subjected the
defense to hostile lines of questioning, which signaled his likely vote against upholding key
provisions of the Act.
7
And many
observers find that Justice O'Connor, another pivotal justice,
"even more inscrutable than usual" on campaign finance questions.
The
prediction, then,
is
BCRA stopped them, then the
fallen
on December
10,
bill
signed the
bill into
6
v.
donations produced profits and
stock prices of companies that used soft
Supreme Court's
final decision, there
surprised the market. Thus,
on February
passed the
money
'
money
14, 2002; (2) the Senate
we have
passed
it
were four other events surrounding
five events in
1
1
all.
(1)
on March 20, 2002;
The U.S. House
(3) the president
law on March 27, 2002; (4) the Supreme Court heard oral argument on
Brookings Briefing, "Supreme Court Rules on Campaign Finance Case: The Legal and
FEC" December
heavily should have
2003, while those that did not should either have risen or been unaffected.
In addition to the
BCRA that might have
straightforward: If soft
is
,
Political
Impact of McConnell
2003.
Aug 31, 2003 by Charles Lane, "Rehnquist May Be Key for
Campaign Finance Chief Justice's Past Votes Leave Outcome of Challenges to McCain-Feingold Law Uncertain."
Quote by Professor Roy Schotland of the Georgetown University Law Center, from Washington Post article on Aug
31, 2003 by Charles Lane, "Rehnquist May Be Key for Campaign Finance Chief Justice's Past Votes Leave Outcome
See, for example, the Washington Post article on
of Challenges to McCain-Feingold
Law
Uncertain."
Another interesting piece of evidence is the BCRA "market" run for an undergraduate course on the Supreme Court
the University of North Carolina at Chapel Hill. About 130 students in the class traded in the market, rewarded with
The
at
betting was on whether or not the electioneering communication provisions would be upheld. The last
which trades took place, posted on December 8, were $.55 for the position that the provisions would be upheld
and $.55 for the position that the provisions would be struck down, on bets that paid $1.00- this implies beliefs very
close to 50-50. See http://www.unc.edu/courses/2003fall/poli/079/001/market/.
grades.
prices at
September
8,
2003
which, Justice Rehnquist's questioning was viewed as a signal that he would
(at
Supreme Court issued
side with plaintiffs); and (5) the
Following
the event study literature,
the Capital Asset Pricing
we
its
ruling
on December
10, 2003.
10
estimate the following equation, a modification of
Model:
J
5
j=\ s=\
where
Rii is
i
the return
type-y
indexes firms,
on firm
donor and
j's
t
indexes dates, j indexes donor status
stock for date
otherwise, and lst
"bad news" for large
soft
statistically significant;
money
and
if
=
1
We
M
if s
the market return for date
is
t
=t
and
statistically significant. (Note,
February 10, 2001 through December 12, 2003.
cases leaves 446 firms.
12
/
is
a
be negative and
money
Ru = (Pn
'
'
all
donors, then the
- Pi,t-i)l Pi,t-u
where
Fortune 500 companies for the period
Some of these companies
are not publicly traded
complicated mergers during the period under study
To measure
if firm
3 and 5 produced
1, 2,
all
1
t.)
assembled data on daily stock prices for
in
If events
event 4 produced "good news", for large soft
of firm is stock on date
and others were involved
otherwise.
donor, non-donor),
Ay =
t,
donors, then the corresponding ?/s should
corresponding I4 should be positive and
Pit is the closing price
/,
{e.g., large
the market return
We merged this with data on the soft money donations
we
used the
CSRP
for all these firms.
- dropping these
value- weighted return.
13
10
Hertzel, Martin, and Meschke (2002) studied the impact of the first three of these events on the stock returns of 40
major soft-money donors, and found no significant effects. These were probably less surprising than the last event.
The vote on final passage in the House was 240- 1 89, and the vote in the Senate was 60-40; moreover, other events that
occurred during consideration of the bill may have been equally important, such as the approval of a rule by the Rules
Committee on February 8, the adoption of the rule by the full House on February 13, and the approval of the ShaysMeehan substitute amendment over two competitors on February 13. We analyze all five events for completeness.
The first date is exactly one year prior to the first of our five events.
The total number of observations is therefore nearly 325,000.
Firm stock market price data and market data are from
University of Chicago) and Factiva. Soft
money
CRSP
(Center for Research in Security Prices,
donations are from the
web
site
at
the
of the Center for Responsive Politics
(http://www.opensecrets.org/softmoney/index.asp) and the Federal Election Commission.
The Fortune 500 companies include many of the
number of companies
large
that
gave
little
over the 4-year period 1999-2002 were
million),
AT&T ($6.8
at all,
including
money
donors, as well as a
ten largest soft
million), Freddie
SBC Communications
Mac
money donors
($6.4 million), Philip
($3.3 million), Verizon
($3.0 million), Pfizer ($2.9 million), Bristol-Myers Squibb ($2.8
and Anheuser-Busch ($2.7
money
soft
Mae
The
or no soft money.
Morris ($5.3 million), Microsoft ($4.2 million),
($3.1 million), Fannie
largest soft
million).
IBM, American
On
the other side, an impressive
Electric
Power,
Intel,
ALCOA,
list
of firms gave no
Whirlpool, and
Consolidated Edison.
We
group firms into three categories, based on their
election cycles
total soft
money
donations over the two
1999-2000 and 2001-2002: Non-Donors are those who gave $10,000 or
less,
Modest Donors those who gave between $10,000 and $250,000; and Large Donors those who gave
$250,000.
at least
gave
in excess
We
also conducted analyses in
which we isolated Million Dollar Donors, who
of $1 million over the 4-year period. The
first
group contains 216 firms; the second
group contains 142 firms, with an average contribution of about $90,0000; and the third group
contains 142 firms, with an average contribution of $1,080,000. There are 50 Million Dollar
Donors.
Were
answer
is:
soft
money donors
hurt
by the Supreme Court's decision
uphold
BCRA? The
short
Evidently not.
At the end of the trading day on December
better
to
day on Wall Street than firms
that
dropped by about .5% on December 10
day, the prices of
gave no
th
.
The
10, the firms that
soft
gave
soft
money had had
a
money. The value of the broad market index
stock prices of Large
Donors dropped by .3%
Moderate Donors dropped by .6%, and the prices of the Non-Donors
lost
that
.8% of
The Million Dollar Donors - such
their value that day.
saw
their stocks drop only
The event study
the events
on
by .1%. This
is
as
AT&T,
firms' stock market valuations
(i.e.,
BCRA and the
1
shows the estimated
their ?'s) for the three types
money and
soft
from
0.
What
of
Court's decision to uphold the law had no statistically
i.e.,
their ?'s are
more, the effects of the events on firms that gave no
is
modest amounts of soft money were not
firms that gave
effect
of donor firms. The
discernable effect on the valuation of firms that gave large amounts of soft money;
statistically indistinguishable
-
exactly the reverse of our expectations.
analysis confirms this conclusion. Table
events marking the passage of
Microsoft, and Philip Morris
statistically different
from
firms that gave large amounts of soft money. Specifically, the F- statistics at the foot of the table
reveal, in the first case, that the ?'s for the different types
from each other and, in the second case,
That
is,
that
we
of firms are not significantly different
cannot reject the hypothesis that the ?'s are
the data are consistent with the hypothesis that
all
of the events had zero
effect
0.
on the
valuations of all types of firms, If anything, the court's decision appears to have helped the Large
Donors, and hurt the Non-Donors - again, completely contrary
In short, the Bipartisan
that
Campaign Reform Act did not
affect the profitability of corporations
gave considerable amounts of soft money.
There are two possible interpretations
have
new
to expectations.
little
effect
on behavior. Investors might have expected
law. Indeed, the
and 501(c)4's.
to these findings.
14
FEC now
We believe,
faces
new
possibility
that
is
that firms will find a
challenges in dealing with committees
however, that the soft
sizable corporate donations that
One
money ban has
were the hallmark of soft money
teeth
BCRA will
way around
known
the
as 527's
and will eliminate the
in the 1990s.
A second, more profound possibility is that the premise of most of the discourse over
campaign finance
14
These
are the tax
is
simply wrong. Firms
code designations for
political
may not
care
much about
advocacy groups.
soft
money because
they do
not profit
much from
soft
money
donations.
As noted
treated their soft- money donations as investments,
rate
of return, the
total effect
on
profits
in the calculations above,
even
and these investments produced a
would be minuscule and
if firms
fairly
decent
virtually undetectable in stock
market prices.
Moreover, very few firms gave large amounts of soft money. Anyone
this issue
R.J.
over the past decade can probably
AT&T. But
Reynolds, Philip Morris, and
companies gave
money
at all,
in excess
name some of the
The lack of apparent
has followed
money donors - such
as
they are the exceptions. Only one in 25 Fortune 500
of $1 million of soft money
and half gave $10,000 or
large soft
who
in the
2000
election,
while
40%
gave no
soft
less.
financial losses associated with the
end of soft money indicates
at the
very least that the campaign contributions do not exact exceedingly large returns on investment.
Thousand- fold returns, as suggested by the Senate hearings and other anecdotes, are not borne out
in the
behavior of investors.
This conclusion raises a problem with a basic premise in the
societal interest in regulating soft
money? Probably
not.
BCRA.
Apparently,
we
Is there a
are not in a
compelling
world of
excessively large returns to campaign contributors. Firms do not appear to receive a lot for a
And, even with return on investment as large as
1
00%,
there
is
very
little
money
little.
in elections
compared with the value of all goods and services produced by the government.
There
may
be a governmental concern about cases of corruption. That concern seems more
appropriately handled through stricter enforcement of laws that prohibit bribery and quid pro quo
arrangements (Lowenstein, forthcoming). This concern, however,
not that system- wide corruption
is
is
about a few bad apples.
producing substantial private benefits from public policies.
It is
Lacking evidence of actual corruption, the Court relied heavily on concern about perceived
corruption.
But, in the absence of evidence that firms profited from soft
money
or that there
noticeable economic cost to society, one wonders on what the perception of corruption
And
if investors,
who back
their perceptions with real
gains for the largest contributors, one wonders
hard moneys
whose perceptions
—
their
own
the Court
is
is
is
a
based.
—do not perceive
relying on.
References
Ansolabehere, Stephen, John de Figueiredo, and James M. Snyder, Jr. 2003. "Why
Little Money in US Politics?" Journal of Economic Perspectives 17: 105-130.
Is
There So
Dal Bo, Ernesto. 2002. "Bribing Voters." Unpublished manuscript, University of California
at
Berkeley.
Fisman, Raymond. 2001. "Estimating the Value of Political Connections." American Economic
Review
91: 1095-1102.
J. Spencer Martin, and J. Felix Meschke. 2002. "Corporate Soft Money
Firm
Performance."
Unpublished manuscript, Arizona State University.
Donations and
Hertzel, Michael G.,
Jayachandran, Seema. 2002. "The Jeffords Effect." Unpublished manuscript, Harvard University.
William C. Heffernan and John Kleinig,
"When
Campaign Contribution a Bribe?" in
Private and Public Corruption. Lanham, MD: Rowman
Lowenstein, Daniel H. Forthcoming, 2004.
Is
a
&
Littlefield.
Persson, Torsten, and Guido Tabellini. 2002. Political Economics. Cambridge,
Roberts, Brian E. 1990a.
Benefits."
"A Dead
Senator Tells
No
American Journal of Political Science
Lies: Seniority
MA: MIT
Press.
and the Distribution of Federal
34: 31-58.
Roberts, Brian E. 1990b. "Political Institutions, Policy Expectations, and the 1980 Election:
A
Financial Market Perspective." American Journal of Political Science 34: 289-310.
Schwert, G. William. 1981. "Using Financial Data to Measure Effects of Regulation." Journal of
24: 121-158.
Law and Economics
"What Do Campaign Contributions Buy? Deciphering Causal
Money and Votes." Southern Economic Journal 51 606-620.
Stratmann, Thomas. 1991.
:
Effects of
Table
1:
Effect of
BCRA Decisions on
Key
Stock Returns
Supreme
Large Donors
Supreme
House
Senate
President
Court
Court
Passes
Passes
Signs
Argument
Decision
-.19
.32
.07
-.11
.13
(.23)
(.23)
(.23)
(.23)
(.23)
-.08
.31
.46*
-.17
-.18
(.23)
(.23)
(.23)
(.24)
(.24)
-.21
.24
.19
-.31
-.42*
(.19)
(.19)
(.19)
(.20)
(.20)
1
0.11
0.05
0.81
0.24
1.69
F- statistic 2
0.69
1.84
1.78
1.14
1.89
Moderate Donors
Non-Donors
F- statistic
Standard errors in parentheses
*
=
significant at the .05 level
for testing Ho: ?j s
=
?2s
F-statistic 2 is for testing Ho: ?i s
=
?2 S
F-statistic
1
is
=
?3.s (i-e.,
the effect of event s
is
the
same
for
all
3
types of firms)
=
?3s
=
(i.e.,
types of firms)
3837
!;
25
the effect
of event 5
is
zero for
all
3
MIT LIBRARIES
3 9080 02617 6906
wmmmmmM^MmmMM
1
III
P