The Complicated Partisan Effects of State Election Laws
| Author | Donald P. Moynihan,Barry C. Burden,David T. Canon,Kenneth R. Mayer |
| Published date | 01 September 2017 |
| Date | 01 September 2017 |
| DOI | http://doi.org/10.1177/1065912917704513 |
| Subject Matter | Articles |
research-article2017
Article
Political Research Quarterly
2017, Vol. 70(3) 564 –576
The Complicated Partisan Effects of
© 2017 University of Utah
Reprints and permissions:
sagepub.com/journalsPermissions.nav
State Election Laws
https://doi.org/10.1177/1065912917704513
DOI: 10.1177/1065912917704513
journals.sagepub.com/home/prq
Barry C. Burden1, David T. Canon1, Kenneth R. Mayer1,
and Donald P. Moynihan1
Abstract
Conventional political wisdom holds that policies that make voting easier will increase turnout and ultimately benefit
Democratic candidates. We challenge this assumption, questioning the ability of party strategists to predict which
changes to election law will advantage them. Drawing on previous research, we theorize that voting laws affect who
votes in diverse ways depending on the specific ways that they reduce the costs of participating. We assemble datasets
of county-level vote returns in the 2004, 2008, and 2012 presidential elections and model these outcomes as a function
of early voting and registration laws, using both cross-sectional regression and difference-in-difference models. Unlike
Election Day registration, and contrary to conventional wisdom, the results show that early voting generally helps
Republicans. We conclude with implications for partisan manipulation of election laws.
Keywords
election law, early voting, voter registration
To underline the importance of a turnout strategy heavily
and empirical studies have overlooked the means by
used by his campaign, in 2012 Barack Obama became the
which turnout is affected by a particular law, if at all. We
first president to vote early. It was widely believed that
find that Election Day registration (EDR) helps Democrats
early voting, particularly among young people, first-time
as expected, but early voting, by itself, actually helps
voters, and racial and ethnic minorities, made a signifi-
Republicans. Early voting’s effect is due to its indirect
cant contribution to his two victories.1 Given this belief,
effects in reducing overall mobilization even while it low-
it is unsurprising that Republicans tried to shorten the
ers the direct costs of voting. There are several plausible
early voting period in swing states such as Florida, Ohio,
reasons why Democrats might continue to advocate for
and Wisconsin, restrictions that Democrats and affiliated
more early voting, but improving their fortunes at the
groups vigorously opposed.2 Yet virtually no one involved
polls should not be among them unless early voting is
in these legal fights questioned the assumption that early
combined with SDR or EDR. We conclude with several
voting increases turnout and the Democrats’ vote share.
suggestions for future research and what our findings indi-
Indeed, the recent legal tussles over early voting, same-
cate for partisan manipulation of election practices.
day registration (SDR), and voter identification mimic a
debate that took twenty years earlier about the introduc-
All Efforts at Increasing Turnout Are
tion of “Motor Voter” reforms. Since that time, the parti-
san divide over the effects of election laws has escalated
Not Equal
to become “the voting wars” in which parties alter laws,
It is easy to see why so many observers associate easier
go to court, and formulate campaign strategies in ways
voting practices with more votes for the Democrats. The
they believe will increase their vote share (Hasen 2012).
two have become correlated in recent elections, as shown
Efforts to restrict early voting are grouped by scholars as
in Table 1. The table presents data from the 2008 and
part of a wider political strategy to restrict voter access
2012 Survey of the Performance of American Elections
for partisan gain (Bentele and O’Brien 2013).
(SPAE). Obama’s share of the two-party vote in 2008 was
We challenge the conventional wisdom in several
ways. To begin, we suggest that party leaders will some-
1University of Wisconsin–Madison, USA
times be wrong about which side will benefit from a
Corresponding Author:
change in an election law. Not all laws making it easier to
Barry C. Burden, Department of Political Science, University of
register and vote actually increase turnout. And an increase
Wisconsin–Madison, 1050 Bascom Mall, Madison, WI 53706, USA.
in turnout does not necessarily help Democrats. Debates
Email: bcburden@wisc.edu
Burden et al.
565
Table 1. Democratic Vote Share by Voting Method.
mobilizing voters and thus underwriting the costs of par-
ticipation. The direct effects are widely understood, but
2008
2012
indirect effects are usually overlooked. This is problem-
Voted in person on Election Day
48.1%
48.5%
atic because the net effect of election laws on turnout is
Voted in person early
60.0%
51.5%
the combination of these two pathways.
Voted by mail early
52.7%
52.3%
The second theoretical foundation for our study is
n
9,124
8,679
Berinsky’s (2005) insight about the “perverse” effects of
many election laws. Although many election laws are
Source. Survey of the Performance of American Elections.
designed to “stimulate” new participants, in practice most
Data are weighted. Minor party voters are excluded.
conveniences are better at “retaining” likely voters than
bringing out new voters. Making election laws more
highest among those who voted early in person and low-
accommodating to offer things such as absentee voting
est among those who voted at an Election Day polling
often benefits regular voters who might otherwise have
place. The patterns are similar but less stark in 2012,
their voting habits disrupted but does not necessarily
when Obama’s margin of victory was smaller, but early
mobilize additional voters unless parties or other groups
voters (both mail and in-person) still voted for Obama at
take advantage of the opportunities those changes present
a rate more than 3 percentage points higher than tradi-
(Oliver 1996).
tional Election Day voters.
To illustrate how these two theories work in practice,
It is tempting to infer from these patterns that early
we contrast two popular election practices designed to
voting caused an increase in the Democratic vote share.
facilitate voter participation: EDR and early voting. Both
The 2012 Obama campaign concluded as much in its
lower the direct costs of voting. By eliminating the pre-
postelection review, surmising that the “early vote offered
election registration “closing date,” EDR allows the two-
eligible voters expanded access and convenience” and
step process of registering and voting to happen at the
“the campaign’s effort to encourage supporters to take
polling place. By allowing more days on which ballots
advantage of this opportunity was tremendously success-
may be cast, early voting provides alternatives for people
ful,” especially for “voters who the campaign had specifi-
who find traditional Election Day less convenient. But
cally prioritized for early vote: sporadic supporters”
the indirect effects of EDR and early voting are quite
(Obama for America 2013). The Republican Party post-
different.
mortem on the 2012 election mirrored these assumptions,
We hypothesize that EDR is better at “stimulation” of
identifying early voting as a Democratic advantage that
more marginal voters while early voting is better at
Republicans need to counter: “The Democrats success-
“retention” of those already likely to vote. When EDR is
fully front-loaded many of their votes this cycle, expand-
available, political parties, candidates, and social organi-
ing their early vote and absentee reach . . . They continued
zations have a great incentive to turn out all of their sup-
to expand their advantage in early voting” (Republican
porters on Election Day. Because registration is not
National Committee 2012, 29). However, the fact that
required before Election Day, a larger pool of eligible but
traditionally Democratic constituencies such as young
unregistered voters remains ripe for mobilization, making
people and African Americans prefer to vote early does
Election Day an important social event that may nudge
not necessarily mean that early voting, by itself, has a
potential nonvoters into becoming voters. As a result, the
causal effect in boosting the Democratic vote share. That
indirect effect of EDR is also positive, leading to an even
Democratic-leaning voters make disproportionate use of
bigger net positive effect on turnout. It is not surprising
the practice is a necessary but not sufficient condition for
that EDR might increase turnout and help Democrats.
the law to benefit Democratic candidates. The counter-
In contrast, early voting lowers the direct costs of par-
factual question still looms: how many Democratic and
ticipation by providing more days on which to vote. Yet,
Republican voters would have abstained without such
it does not remedy the need for preregistration, and regis-
options?
tration is a powerful predictor who will vote (Highton
We build our argument on the foundations of two
2004; Timpone 1998). The convenience of more voting
existing theories. First, we follow Burden et al.’s (2014)
days or locations is only helpful to those who are already
argument that election laws have both direct and indirect
registered and interested enough to take part. Early voting
effects on voter turnout. Direct effects...
To continue reading
Request your trial