The Ideological Congruence Illusion: The Impact of Valence

• Author: Paul V. Warwick
• Date: 01 May 2016
• DOI: http://doi.org/10.1111/lsq.12120
• Published date: 01 May 2016

PAUL V. WARWICK

Simon Fraser University

The Ideological Congruence

Illusion: The Impact of Valence

Considerable evidence has been generated to establish that “ideological con-

gruence,” that is, a close matching of government policy positions with median left-

right voter opinion, generally prevails in liberal democracies. Based on a cross-national

analysis of election survey data, this article challenges that view and elaborates an alter-

native perspective. In this perspective, nonpolicy or valence considerations strongly

influence vote choices and electoral outcomes in a directional sense—sometimes favor-

ing the Left overall, sometimes the Right. Partly as a result, government positions

typically deviate substantially from median opinion, forming a pattern that is clearly

bilateralist or two-sided rather than center-concentrated.

One of the foremost features of liberal democracy is the ability of

citizens to choose among alternatives at election time. With choice, gov-

ernments whose policies or actions provoke public disapproval can be

removed from power peacefully and replaced by alternative govern-

ments, usually proposing alternative policies. The public’s ability to

implement change, at least periodically, should ensure that government

remains basically responsive to public opinion, a quality that, as Sabl

(2014) notes in a recent essay, is generally endorsed by normative

theorists of democracy.

While choice among governing alternatives is integral to respon-

siveness, there is a certain tension between the two traits. If voters can

punish nonresponsive governments by replacing them, then government

leaders who wish to remain in power are incentivized to adopt policies

and undertake actions that meet the public’s expectations. Taken to its

extreme, this should produce a close matching of government positions

or policy stances with some measure of aggregate public opinion, a prop-

erty that Powell (2009) labels “ideological congruence.”

1

But if a state of

ideological congruence is realized, that is, if governments generally

implement the collective preferences of the electorate as a whole, how

can meaningful policy choice be said to exist? Would not any differences

in declared policy positions seem pointless, if not deceitful, in such

LEGISLATIVE STUDIES QUARTERLY, 41, 2, May 2016 445

DOI: 10.1111/lsq.12120

V

C2016 Washington University in St. Louis

circumstances? Conversely, if rival parties or coalitions provide mean-

ingful choice in the sense that they make distinct policy commitments

that they intend to implement if elected, how can ideological congruence

be achieved?

The most straightforward solutiontothisdilemmawouldbeto

desist from interpreting responsiveness in terms of congruence. Soroka

and Wlezien’s (2010) well-known “thermostatic” model incorporates

this approach. In the model, governments tailor their policy stances to

meet public expectations but not to the point where congruence prevails

and electoral choice becomes meaningless in policy terms. Instead, as

the metaphor of a thermostat suggests, policy oscillates between two dis-

tinct alternatives (the temperature at which heat is turned on and the

temperature at which it is turned off), neither of which is likely to

correspond exactly with what the public ideally prefers.

The solution of discarding congruence as a criterion is not univer-

sally embraced, however. This appears to be partly a consequence of

confusion over its normative status. “Political theorists,” Sabl observes,

“mostly agree that there should be a certain baseline level of responsive-

ness. But it does not follow—and neither political theorists nor ordinary

commentators on democracy believe—that such responsiveness should

be maximized, that the optimal divergence between public opinion and

policy should be zero” (2014, 14, italic mine). In contrast, Sabl observes

that many of those doing empirical work on democracy assume precisely

the opposite—that normative theory endorses congruence—and, partly

for this reason, do so themselves. These include Powell (2000) in his

classic study of elections, Lijphart (1999) in his equally inf‌luential trea-

tise on democracy, and especially the two large-scale studies of elections

and governments by Budge, McDonald, and associates (Budge et al.

2012; McDonald and Budge 2005).

How is the conundrum addressed by these scholars? While Powell

and Lijphart adopt congruence as an indicator of democracy, they focus

on the degree to which it is realized in various systems, rather than its

presence in some absolute sense. A more ambitious tack is pursued by

Budge, McDonald, and associates, who offer a perspective that explicitly

combines the presence of both traits. Their extensive investigations reveal

that government policy in liberal democracies generally aligns in a one-

to-one fashion with the median voter’s position on the left-right dimen-

sion (Budge et al. 2012; McDonald and Budge 2005).

2

This congruence

or “median mandate,” as they term it, can be detected even when govern-

ment positions are estimated from party electoral manifestos, which

clearly do offer distinct policy alternatives (Budge and McDonald 2006),

so it is not a case that the positions staked out in election campaigns are

446 Paul V. Warwick