A Thermostatic Model of Congressional Elections

• Published date: 01 July 2024
• DOI: http://doi.org/10.1177/1532673X241253813
• Author: Matt Grossmann,Christopher Wlezien
• Date: 01 July 2024

Article

American Politics Research

2024, Vol. 52(4) 355–366

© The Author(s) 2024

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X241253813

journals.sagepub.com/home/apr

A Thermostatic Model of Congressional

Elections

Matt Grossmann

1

and Christopher Wlezien

2

Abstract

Are policymakers rewarded in elections when they succeed in moving public policy in their ideological direction? Or do they

face a thermostatic backlash, as citizens judge their policy moves as too hot or too cold? Our analysis of Congressional elect ion

outcomes since 1948 adds information on Congressional policy actions to traditional election models emphasizing the surge and

decline of presidential support and referendums based on presidential approval and theeconomy. We find that the electorate

reacts to the ideological direction of policy, voting against parties that push policy further to the left or the right in both midterm

and presidential years. Even after accounting for policy and traditional explanations, however, there remains a large midterm

penalty for the president’s party.

Keywords

policy, major laws, coattails, balancing, midterm loss, backlash

Politicians seek to move policy in their preferred ideological

direction and believe that elections give them a mandate to

govern (Azari, 2014). But that hardly means voters will re-

ward them. Consider research showing that public support for

policy change moves “thermostatically”against the direction

of policymaking, shifting leftward under Republican presi-

dents making conservative policy moves and rightward under

Democratic presidents making liberal policy moves (Erikson,

et al., 2002;Wlezien, 1995,2017). These shifts lead voters to

turn against the party of ambitious presidents that overshoot

public demand for policy change (Erikson, et al., 2002;

Wlezien, 2017). In presidential elections, the further policy

moves rightward or leftward, the more relative preferences

move in the opposite direction and the worse the candidate of

the incumbent party does (also see Bolstad, 2012), explaining

about half of the cost of ruling effect (Wlezien, 2017).

Policymaking may have similar consequences for Con-

gressional elections. Public judgments of Congress respond

to congressional lawmaking, as voters use policy output to

judge the ideological distance between themselves and

Congress (Algara, 2021). Voters tend to move against major

laws passed by Congress as they learn about details (and

infighting) from media coverage (Atkinson, 2017). Major

electoral losses for the majority party in Congress, such as in

the 2010 election, are the product of voters punishing leg-

islators for supporting policies that demonstrate they are

ideologically distant from voters (Nyhan et al., 2012). All of

these findings suggest that parties may suffer from moving

policy too far in their preferred direction. State-level evidence

also finds that voters counterbalance parties that succeed in

legislating major policy victories (Pacheco, 2013).

Despite these findings, research on Congressional elec-

tions has yet to investigate whether aggregate results reflect

the policies enacted by each Congress. The literature has

found many other factors that may influence Congressional

elections, including the approval of the president and the state

of the economy. But the most consistent finding is that these

factors do not explain midterm loss. In midterm elections,

presidents’parties lose seats and votes, even more on average

than they gain when voters surge toward their party in the

previous presidential election. Yet policy shifts may help

account for some of the midterm loss, as voters swing against

presidents and allied Congresses moving policy in their

ideological direction. Research finds that voters out of step

with the policy direction under a president seek to balance the

president’s party and to voice opposition in midterm elections

in the hopes that the president will respond by moderating

1

Department of Political Science, Michigan State University, East Lansing, MI,

USA

2

Department of Political Science, University of Texas at Austin, Austin, TX,

USA

Corresponding Author:

Matt Grossmann, Department of Political Science, Michigan State University,

303 S. Kedzie Hall, East Lansing, MI 48824, USA.

Email: matt@mattg.org