Using Computer Simulations to Measure the Effect of Gerrymandering on Electoral Competition in the U.S. Congress

• Date: 01 August 2019
• Author: David Cottrell
• DOI: http://doi.org/10.1111/lsq.12234
• Published date: 01 August 2019

487

LEGISLATIVE STUDIES QUARTERLY, 44, 3, August 2019

DOI: 10.1111/lsq.12234

DAV ID COT TR EL L

University of Georgia

Using Computer Simulations to

Measure the Effect of Gerrymandering

on Electoral Competition in the

U.S. Congress

Recent research has leveraged computer simulations to identify the ef-

fect of gerrymandering on partisan bias in U.S. legislatures. As a result of this

method, researchers are able to distinguish between the intentional partisan bias

caused by gerrymandering and the natural partisan bias that stems from the geo-

graphic sorting of partisan voters. However, this research has yet to explore the

effect of gerrymandering on other biases like reduced electoral competition and

incumbency protection. Using a computer algorithm to design a set of districts

without political intent, I measure the extent to which the current districts have

been gerrymandered to produce safer seats in Congress. I find that gerrymander-

ing only has a minor effect on the average district, but does produce a number of

safe seats for both Democrats and Republicans. Moreover, these safe seats tend to

be located in states where a single party controls the districting process.

In recent dec ades, House elections h ave resulted in a near-

even partisan spl it of the popular vote. Yet despite partis an parity

at the national level, elections at the d istrict level are relatively un-

competitive. Re-elect ion rates in the House have dropped b elow

90% only twic e in the past 40 years , and electoral comp etition

has been in de cline since the 1970s (Abramowitz, Alexander, and

Gunning 20 06). In the most recent 2018 midterm election, 91%

of incumbents who ran for re-election were re- elected. De spite

being a wave year for Democrats, only 30 s eats were classif ied as

toss-ups accord ing to the Cook Political Report, a nd the average

margin of victory across all d istricts was over 30%.1

© 2019 Washington University in St. L ouis

488 David Cottrell

Given that national election s are so close, why then are seats

so safe? While there a re a number of factors that potential ly ex-

plain the lack of elec toral competition acros s districts, one par-

ticularly conce rning explanation is t hat the lack of competition

is a byproduct of gerry manderi ng. Through gerr ymanderi ng,

parties work to pack and crack districts during the redistricti ng

process in an effort to gain an electoral advantage. This p acking

and cracking a lters the electoral landscape a nd inevitably leads to

changes in the level of elec toral competition in districts a cross the

country. These changes potentially result i n more uncompetitive

districts t han there otherw ise would be.

Yet empirical evidence that gerryma ndering resu lts in safer

seats in the House has b een mixed. Tufte (1973), for example, find s

that competition in the House de clined in the 1960s and attributes

at least some of the decli ne to a series of ger ryma nders that oc-

curred dur ing the reapportion ment revolution following Baker

v. Carr (369 U.S. 186, 1962). Others have similarly found a lin k

between the reduction in competition and ger rymandering (Ki ng

1989; Lyons and Galderisi 1995; McDonald 2004; Niemi and

Jackman 1991). However, the effect is usually isolated to st ates

where parties share control over the redistric ting process a nd

protect incumbe nts through biparti san gerry manderi ng. Yet not

all studies ag ree that gerr ymander ing leads to reduc ed electoral

competition. There is al so evidence that gerry mandering has little

to no effect on electoral c ompetition in the House (Abramowitz,

Alexander, and Gunni ng 2006; Glazer, Grofman, and Robbins

1987) or on second-order outcomes like polarization (McCar ty,

Poole, and Rosenthal 2009). There is even evidence th at gerry-

mandering works to enha nce electoral c ompetition rather than

reduce it (Gelman and Ki ng 1994).

One potential reason for the lack of c onsensus sur rounding

the effect of gerr ymander ing is that isolating the ef fect is hard to

do. Researchers must attempt to se parate the effect of ger ryman-

dering from that of various alt ernative explanations. There a re

two alternative explanations that are commonly use d to explain

the abundance of safe se ats. The predom inant explanation is that

safe seats are a resu lt of incumbency advantage (Cox and Katz

1996; Erikson 1971; Gelman and K ing 1990; Levitt and Wolfram

1997). This explanation suggests that se ats are safe not becaus e

they are designed to be that way, but because incumbents who

run for re-elec tion have a natural advantage over their c ompeti-

tion. They have name recogn ition, government resourc es, better