Research Briefs
IN ECONOMIC POLICY
June 3, 2020 | Number 217
Quid Pro Quo?
Corporate Returns to Campaign Contributions
By Anthony Fowler, University of Chicago; Haritz Garro, Northwestern
S
University; and Jörg L . Spenkuch, Northwestern University
cholars, pundits, and political reformers have long
worried that corporate campaign contributions dis-
tort public policy. The biggest concern is that cor-
porate contributions alter the behavior of elected
officials in ways that benefit the contributing firms
while subverting the will of the electorate. Despite some
troubling anecdotes, we have little systematic evidence on
the effects of corporate campaign contributions. In a forth-
coming study in the Journal of Politics, we assemble data from
nearly 19,000 elections and nearly 3,000 firms spanning
more than three decades to understand the extent to which
campaign contributions systematically benefit firms.
One challenge to studying the effects of corporate cam-
paign contributions is that firms give strategically, and big,
successful firms are especially strategic. So naive compari-
sons of firms that give more versus firms that give less or
comparisons of elected officials who received contributions
versus those that did not would make it look as if corporate
contributions are significantly distorting policy. For example,
the companies that have contributed to more members of
Congress are often the most prosperous firms, but we have
little way of knowing how much of that success is a result of
their campaign contributions.
To more credibly estimate the distortionary effects of
corporate campaign contributions, we utilize two comple-
mentary research designs. In both cases, our goal is to esti-
mate the effect of having one additional elected official to
whom a firm has contributed on the value of the firm. If there
Editor, Jeffrey Miron, Harvard University and Cato Institute
is a quid pro quo whereby elected officials distort policy to
benefit the firms from which they received contributions,
this means that the election of an additional official with this
kind of connection should benefit the firm, and these ben-
efits should be realized in the firm’s stock price.
Our first design relies upon the arbitrary results of very
close elections. We focus on situations in which a publicly
traded firm has given money to one candidate (but not the
other) in a very close election, and we test whether a bare
victory for the supported candidate benefits the firm’s stock
price relative to a bare victory for the other candidate. By
focusing on the arbitrary results of very close elections,
we are essentially comparing situations where firms sup-
ported a barely winning candidate with those where firms
supported a barely losing candidate, which is much more of
an apples-to-apples comparison than if we had simply com-
pared firms that gave to lots of winners with firms that gave
to fewer winners.
The figure below provides a visualization of this analysis.
The vertical axis shows the cumulative abnormal stock re-
turns for firms shortly after a close election. The horizontal
axis shows the vote share of the firm-supported candidate.
Each dot corresponds to many firm-election pairs with
similar vote shares. We see that the stock returns for barely
winning firms are essentially identical to those of barely los-
ing firms, suggesting that firms do not meaningfully benefit,
on average, from having one additional connected candi-
date elected.
2
Figure 1
Cumulative abnormal stock returns and vote share of firm-favored candidates
0.02
sn)sru/tm

0.01
erk(laym
ticronlebvaneoviistsaelucm

0
eR
uC

-0.01
0.45 0.46 0.47 0.48 0.49 0.50 0.51 0.52 0.53 0.54 0.55
976154.0
565454.0
349654.0
758954.0
885264.0
579464.0
58764.0
141074.0
495274.0
535474.0
336674.0
520974.0
568084.0
53384.0
591584.0
69684.0
360984.0
139094.0
438294.0
530594.0
458694.0
884894.0
5679994.0
643005.0
307205.0
481505.0
197705.0
139905.0
972215.0
546415.0
959615.0
345915.0
205125.0
217325.0
19525.0
11825.0
227035.0
747235.0
460535.0
681735.0
904935.0
586145.0
720445.0
870645.0
799745.0
579945.0
Two-party vote share of favored candidate

Source: Authors’ calculations.

Note: Figure depicts fitted values and confidence intervals from regression
Distance of cumulative
from Earth (millionabnormal returns on two-party vote share of favored candidates.
light-years)

This particular analysis aggregates data from thousands of
firms and thousands of elections across many different set-
tings and time periods. However, if we focus on a subset of
the data where we might expect greater effects, we still find
no evidence of distortionary effects. For example, if we only
focus on gubernatorial elections or Senate elections, big-
ger or smaller firms, firms with more or fewer political con-
nections, elections before or after Citizens United v. Federal
Election Commission, and so forth, we still obtain null results.
We also implement a second design that focuses on a dif-
ferent source of variation. Specifically, we focus on Senate
campaigns where the outcome is uncertain, and we utilize
betting-market data to exploit within-campaign changes in
market beliefs about the outcome of the elections. Again,
we find little evidence that the election of a candidate to
whom the firm contributed benefits the firm. For those who
worry that very close elections are unusual and unrepresen-
tative, this design provides variation for a broader set of
high-stakes elections. However, this design also relies more
heavily on the assumption of efficient markets. Because of
their relative strengths and weaknesses, we show results
from both designs, and, reassuringly, they give nearly iden-
tical results.
The fact that we fail to obtain statistically significant evi-
dence of a distortionary effect doesn’t necessarily mean that
corporate campaign contributions have no effects. How big
could those effects potentially be? We can statistically reject
any purported effect that is greater than 0.3 percent. In other
words, if having an additional elected official who received
campaign contributions does benefit a firm, the effect is like-
ly no bigger than 0.3 percent of the firm’s value. Of course,
many firms in our analysis are quite large and many of the
campaign contributions are small, so it could still be worth-
while from the firm’s perspective to contribute. However, the
effects are not large enough for any one person in the firm
(including the CEO) to justify contributing to the corporate
political action committee (PAC) individually, and this is the
primary way in which corporate PACs raise money.
If firms don’t systematically benefit from corporate cam-
paign contributions, why do they give so much? One answer
is that they don’t. Many large firms don’t even have corpo-
rate PACs, and many of those that do make only nominal
contributions. Compared with other firm activities, such as
marketing, research and development, legal counsel, and so
forth, corporate campaign contributions are a tiny drop in the
bucket. Even though firms might give a little and get a little
 3

in return, know that there isn’t much further to gain from
campaign contributions. Another potential answer is that it’s
hard to know whether corporate campaign contributions are
worthwhile and that there are agency problems within firms
that lead them to continue making campaign contributions
even though the returns are negligible.
Although there may be some troubling cases where cor-
porate campaign contributions appear to have distorted pol-
icy, a more thorough analysis suggests that, on average, firms
do not systematically benefit from having a candidate to
whom they contributed elected. This is not to say that there
is no reason to worry about money in politics or that there
are no good justifications for campaign finance regulations,
but we find little evidence that public policy is distorted in fa-
vor of firms as a result of corporate campaign contributions.
NOTE:
This research brief is based on Anthony Fowler, Haritz Garro, and
Jörg L. Spenkuch, “Quid Pro Quo? Corporate Returns to Cam-
paign Contributions,” Journal of Politics, forthcoming, https://
www.journals.uchicago.edu/doi/full/10.1086/707307.

The views expressed in this paper are those of the author(s) and should not be attributed to the Cato Institute, its
trustees, its Sponsors, or any other person or organization. Nothing in this paper should be construed as an attempt to
aid or hinder the passage of any bill before Congress. Copyright © 2020 Cato Institute. This work by Cato Institute is
licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.