Constituency Size and Incumbent Safety: A Reexamination

• Author: Edward L. Lascher
• DOI: 10.1177/106591290505800207
• Published date: 01 June 2005
• Date: 01 June 2005
• Subject Matter: Articles

Constituency Size and Incumbent Safety:
A Reexamination
EDWARD L. LASCHER, JR., CALIFORNIA STATE UNIVERSITY, SACRAMENTO
Existing literature suggests that, if anything, larger constituencies make reelection more challenging for leg-
islative incumbents. Yet theoretical arguments commonly are unclear about why more populous constituen-
cies should disadvantage incumbents relative to challengers. Additionally, most empirical research has focused
on congressional races, none of which involve small populations. I argue that there is good reason to believe
that smaller size enhances challenger viability and thereby lowers incumbent safety. I also provide an empiri-
cal analysis of the relationship between constituency size and incumbent safety, focusing on individual con-
tests featuring an enormous variance in district population and other research advantages: races for positions
as California county supervisors. Controlling for both candidate and demographic variables, and examining
both likelihood of winning and incumbent vote percentage, I find that there is a modest but consistent posi-
tive relationship between constituency size and incumbent safety.
Various studies have considered the question of that vastly exceed in population the constituencies for vir-
whether constituency size affects the reelection
tually all American state and local electoral contests as well
prospects of incumbent American legislators. Some
as the constituencies of other legislatures using the single
scholars have found that larger constituencies produce more
district, first-past-the-post electoral system such as the
electoral losses for incumbents and/or lower victory margins
British House of Commons, Canadian Parliament, Canadian
(see especially Hibbing and Brandes 1983; see also
provincial legislatures, etc. Alaska and Wyoming (and to a
Abramowitz 1988; Carey, Niemi, and Powell 2000; Lee and
lesser extent many other states) are also quite dissimilar
Oppenheimer 1999, ch. 4,) Others have found a negligible
from other entities in terms of the unusual challenges they
or nonexistent relationship between constituency size and
pose for interacting with voters in states with land masses of
measures of incumbency safety (Black 1974; Krasn 1994:
more than 570,000 and 97,000 square miles, respectively.
ch. 3; Westlye 1991: ch. 7: Bledsoe 1993: 142-43). To my
Methodologists generally recommend examining the full
knowledge, no empirical study has argued that larger con-
range of an explanatory variable to enhance certainty about
stituency size enhances electoral safety.
causal inferences (King, Keohane, and Verba 1994: 215-16).
Yet there is reason to be suspicious about whether pre-
Accordingly, it is desirable to examine American subnational
vious research has captured the critical variance in the con-
elections to understand the impact of district size on incum-
stituency size variable. Scholars generally have focused on
bent safety. For example, as Carey, Niemi, and Powell
elections to the U.S. Senate, taking advantage of the “natu-
(2000: 684) indicate, state legislative districts vary in size
ral experiment” created by the American Founders (i.e., the
from a few thousand to hundreds of thousands of residents.
ability to compare electoral success rates for the same office
Yet aside from the work by Carey and his colleagues, little
across states of very different sizes). Such research has some-
systematic attention has been given to the effect of variance
times been bolstered by consideration of U.S. House races.
in district size at the state or local levels. Even in their study,
Of course, Senate and House races are most important in
district size is not a primary focus (such is also true of Bled-
terms of potential impact on public policy, and therefore
soe’s otherwise thorough study of the determinants of
intrinsically worthy of study. At the same time, in compari-
incumbent safety in city council elections, which only con-
son to the electorates for other American legislative races
siders the constituency size variable in passing; see Bledsoe
(and indeed, legislative electorates in most other democratic
1993: 142-43). An exception is Black’s 1974 study of San
countries), all of the congressional constituencies are quite
Francisco Bay Area city council elections, which offers a
large. Consider statewide congressional contests in Alaska
sophisticated theoretical analysis of the potential impact of
and Wyoming, with populations of 627,000 and 494,000 in
constituency size on a number of electoral variables, includ-
2000, respectively. While sparsely populated in comparison
ing incumbent safety. But the empirical portion of Black’s
to other states, Alaska and Wyoming have constituencies
analysis is thin, relying on a dichotomous division of cities
into “large” and “small” groups, in large part using percep-
tions of incumbent safety rather than actual results, and
NOTE: I wish to thank Dan Goldstein, Michael Hagen, John Hibbing, Jean
working with a sample with much less population variance
Hurst, Matt Newman, Rob Wassmer, and the anonymous reviewers
than would have been available if elections from across Cal-
for helpful information and suggestions. I am especially grateful to
ifornia were studied.
Tim Hodson for valuable input at various stages of this project.
In this study I reexamine the relationship between dis-
Political Research Quarterly, Vol. 58, No. 2 (June 2005): pp. 269-278
trict size and incumbent safety, focusing on races for county
269

---

270
POLITICAL RESEARCH QUARTERLY
boards of supervisors in California. Information on such
Court decisions in the 1960s related to apportionment of
contests constitutes a rich source of data virtually untapped
legislative seats, county districts must be roughly equal in
by political scientists. I argue that there appears to be a
population (the default arrangement is that district configu-
modest but consistent positive relationship between elec-
rations are established by the board members themselves).
torate size and incumbent safety for California county
Supervisors have similar legislative and administrative
supervisors. In contrast to much other work I also contend
responsibilities in all counties (i.e., overseeing and regulat-
that this empirical relationship is to be expected.
ing land use planning and municipal type services in unin-
corporated areas, approving the county budget, administer-
ADVANTAGES AND DISADVANTAGES OF FOCUSING
ing various required state programs, serving on various
ON COUNTY SUPERVISORIAL RACES
regional bodies, performing constituency services),
although the impact and scope of their decisions obviously
Since many readers are likely to be unfamiliar with the
vary greatly (Lascher 1993a). In short, county supervisors
characteristics of county elections, it is helpful to consider
across the state generally have similar jobs, compete for
in depth the pluses and minuses of concentrating on Cali-
positions on boards of equal size, run as independent polit-
fornia supervisorial elections for a study of this kind. Such
ical entrepreneurs in their own districts, serve four year
contests offer a number of prominent advantages. First,
terms, and usually do not face term limits. By contrast, state
there is a wide range and enormous variance in the values
legislatures differ substantially with respect to party com-
of the independent variable. California’s 58 counties range
petitiveness as well as institutional factors such as using
in size from tiny Alpine, nestled in the Sierra Nevada moun-
single or multi-member districts and term length (Carey,
tains and with a 2000 population of a little over 1,000
Niemi, and Powell 2000). This has benefits if one wishes to
people, to enormous Los Angeles with a population of
study the impact of a number of institutional features but
8.863 million, exceeding the size of all but seven American
complicates matters if one is concentrating on the con-
states other than California itself (for county population
stituency size variable.
data see California State Association of Counties/California
Third, county supervisorial positions are politically sig-
Institute for County Government 2001). Supervisorial dis-
nificant in the Golden State. Interestingly, political scientists
tricts in turn contain anywhere from a few hundred people
have deemed the work of California city councils suffi-
to over a million constituents, with a wide variety between
ciently important to be the subject of book length studies of
the two extremes. While the average supervisorial district
such topics as political representation and recruitment
consisted of about 103,000 people in 2000, the standard
(Eulau and Prewitt 1973: Prewitt 1970), even though coun-
deviation was fully 246,000 residents.
cil positions are generally lower in the political hierarchy
Second, many of the structural factors that might affect
than are county board positions (council members com-
incumbency reelection rates are effectively controlled by
monly run for seats on the county board while movement in
comparing across California counties. All county superviso-
the other direction is rare). Additionally, supervisorial posi-
rial elections in the state are officially nonpartisan (party
tions are often stepping stones to seats in the state legisla-
labels may not appear on the ballot), and existing literature
ture and elsewhere (Lascher 1993b).
suggests that in the vast majority of cases parties do not...