Nationalization in U.S. Senate and Gubernatorial Elections
| Author | Joel Sievert,Seth C. McKee |
| Published date | 01 September 2019 |
| Date | 01 September 2019 |
| DOI | http://doi.org/10.1177/1532673X18792694 |
| Subject Matter | Articles |
research-article2018
Article
American Politics Research
2019, Vol. 47(5) 1055 –1080
Nationalization in
© The Author(s) 2018
Article reuse guidelines:
U.S. Senate and
sagepub.com/journals-permissions
https://doi.org/10.1177/1532673X18792694
DOI: 10.1177/1532673X18792694
journals.sagepub.com/home/apr
Gubernatorial Elections
Joel Sievert1 and Seth C. McKee1
Abstract
Since the 1970s, American politics has taken an impressive turn away
from political dealignment and moved toward a more responsible party
system. As a result, elections have become more nationalized, a process
by which presidential and national politics exert greater influence over
down-ballot contests. We evaluate nationalization in electoral contests for
two high-profile offices—U.S. Senate and governor—that encompass the
same constituencies but constitute markedly different job descriptions and
mandated responsibilities. Our analysis utilizes both macro- and micro-
level data (election returns and surveys, respectively) to assess patterns
of nationalization in these elections. Although it is perhaps not surprising
that the linkage between presidential and senatorial contests has tightened,
it appears national political forces now also exert greater influence over
gubernatorial contests. Nonetheless, we do find regional variation in the
nationalization of senatorial and gubernatorial politics, which is more evident
in the latter office.
Keywords
nationalization, senators, governors, presidential vote, elections
1Texas Tech University, Luboock, TX, USA
Corresponding Author:
Joel Sievert, Texas Tech University, 113 Holden Hall P.O. Box 79409, Luboock, TX 79409,
USA.
Email: joel.sievert@ttu.edu
1056
American Politics Research 47(5)
Not very long ago, if co-partisan identifiers in different sections of the United
States were to meet (e.g., a New England Democrat and a southern Democrat),
they might view each other as second cousins twice removed. But these days,
we suspect a gathering of co-partisans from almost any place in the country
would seem like a reunion of close siblings. Nationalization has promoted a
conformity of political views such that sectional variation in partisan posi-
tions has substantially eroded. Anecdotally, just about any seasoned political
observer can now point to low-level electoral contests in which the leading
contenders stressed national issues and even parroted the talking points pro-
duced by the national parties. Indeed, we suspect if Alabama Governor
George Wallace were alive today he would acknowledge that there is a lot
more than a dime’s worth of difference between the two parties, and further-
more, whether he was defending the principles of the Democratic Party in
Birmingham, Alabama or Cambridge, Massachusetts, the playbook would
essentially be the same.
As we understand it, nationalization refers to an increasing linkage
between presidential voting patterns with subpresidential contests at the fed-
eral, state, and local level. As a consequence, the most visible and salient
election contest, that for the White House, sets the agenda for most other
American elections. For years, southern politics scholars recognized and
emphasized the role of presidential elections in leading to Republican elec-
toral success in lower offices (e.g., Black & Black, 1987), and this was
dubbed top–down advancement (see Aistrup, 1996). Nationalization is a
more expansive form of top–down advancement that is expected to permeate
all regions of the United States. While many recent studies assess nationaliza-
tion in congressional elections (Abramowitz & Webster, 2016; Jacobson,
2015a, 2015b), we examine nationalization in U.S. Senate and gubernatorial
elections. In doing so, we investigate the extent of nationalization across
offices that share an electoral constituency but have fundamentally different
political responsibilities.
Questions about the connection between presidential contests and Senate
and gubernatorial elections are not new. Indeed, there was once an active
debate over whether presidential elections and presidential performance
impacted high-profile nonfederal contests like those for governor. Simon
(1989) found a significant and robust relationship between presidential
approval and voting for governor, which was taken as evidence that presi-
dents cast influence over the entire American political system. By contrast,
Atkeson and Partin (1995) argued that there is substantial compartmentaliza-
tion in voter impressions of the responsibilities of senators and governors.
Their analysis found that gubernatorial vote choice is immune to the effects
of presidential approval, whereas Senate vote choice is tied to presidential
Sievert and McKee
1057
evaluations. Later work by Carsey and Wright (1998a, 1998b), however,
found that presidential evaluations affected voting behavior in both senatorial
and gubernatorial elections.1
We agree with Carsey and Wright (1998a) when they say, “For many citi-
zens, political judgments are general indictments or rewards of the party in
power, usually defined as the presidential party. Thus, we expect presidential
approval to influence all types of subpresidential voting behavior” (p. 995).
The evidence supports this statement, but we can go a step further in claiming
that not only does presidential approval exhibit wide-ranging and deep effects
on subpresidential voting behavior, but this has also become a long-term
development in American politics. Since the 1990s, there is little question
that the nationalization of American elections has continued apace. There is
no longer an argument as to whether nonfederal elections bear a connection
to presidential voting behavior. To us, the more relevant and understudied
question is now, “To what extent has nationalization impacted subpresiden-
tial elections like those for senator and governor?” In addition, we believe a
study of nationalization should examine its reach in sectional terms. Because
different regions of the United States vary in terms of partisan support and
partisan strength, it is sensible to gauge the level of nationalization in Senate
and gubernatorial elections not only throughout the country but also in its
major sections. Furthermore, nationalization should lead to electoral forces
working in similar ways across different states or electoral districts (Claggett,
Flanigan, & Zingale, 1984; Kawato, 1987). As such, an exploration of
regional variation provides another means of testing the extent and reach of
nationalization. In the analysis that follows, we employ Black and Black’s
(2007) five-region typology.2
Our assessment of nationalization in Senate and gubernatorial elections
proceeds as follows. First, we describe how the electoral dynamics of
American elections have changed since the 1950s to today. We then discuss
some of the legal and perceptual distinctions between a position in the U.S.
Senate versus a governorship. Next, we evaluate the extent of nationalization
in Senate and gubernatorial contests with election data going back to 1980.
We follow this analysis with an examination of nationalization at the indi-
vidual level with survey data. We conclude with a statement regarding the
far-reaching implications of growing nationalization in American politics.
The Changing Nature of American Politics
From the end of World War II and into the 1990s, there were a handful of
themes that described the nature of American electoral politics: candidate-
centered elections, dealignment, and divided government (Wattenberg,
1058
American Politics Research 47(5)
1998). Of course, these characterizations are closely related. If candidates
have relatively more sway over voters than the party label does, then it
makes sense that voters move toward independence (dealigning) when
Democrats and Republicans are not actively differentiating their political
brands. If the party label depreciates and, by comparison, the individual
candidate is elevated, then the incumbency advantage grows in importance
since many voters will defect from their party affiliation in order to reelect
an attentive representative (Cain, Ferejohn, & Fiorina, 1987; Fenno, 1978;
Fiorina, 1977; Mayhew, 1974), and this voting behavior in turn deters the
emergence of high-quality challengers (Cox & Katz, 1996). It follows that
a rise in candidate-centered politics will produce a decline in party-line vot-
ing (Jacobson, 2015a), which leads to a disconnect in preferences across
levels of office holding, manifested in ticket-splitting, which can foster
divided government.
With the passage of time, the aforementioned view of American politics
has largely faded. In the 1990s, several scholars noticed a waning of candi-
date-centered elections and a pronounced decline in ticket-splitting. In fact,
divided government often persists despite the palpable evidence of a resur-
gence in party-line voting3 (Bartels, 2000) that is attributable to the growing
ideological polarization of elected officials, and particularly members of
Congress (Fleisher & Bond, 2004; Hetherington, 2001). Black and Black
(2002) regard the election of President Reagan as a watershed moment
because the California Republican hastened the realignment of conservative
white southerners to the Grand Old Party (GOP) while dealigning white
southern moderates who used to be loyal Democrats. At the national level,
Reagan’s impact...
To continue reading
Request your trial