Interest Group Influence and the Two Faces of Power

• Date: 01 July 2019
• DOI: 10.1177/1532673X18786723
• Author: Leslie K. Finger
• Published date: 01 July 2019
• Subject Matter: Articles

https://doi.org/10.1177/1532673X18786723

American Politics Research

2019, Vol. 47(4) 852 –886

© The Author(s) 2018

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1532673X18786723

journals.sagepub.com/home/apr

Article

Interest Group

Influence and the

Two Faces of Power

Leslie K. Finger1

Abstract

This study revives the “two faces of power” conceptualization to consider

whether common interest group measures are valid proxies for interest group

strength. The “two faces of power” model proposes that power enables not

just influence on decisions, but also influence from controlling the agenda.

Using the case of state-level teachers’ unions, I test whether measures of a

group’s ability to set the agenda (the second face of power) have a stronger

relationship with policy outcomes and stakeholder perceptions of influence than

do measures of a group’s engagement with legislators (the first face of power).

I find support for this proposition. The second face, operationalized with union

membership rates, is associated with policy proposal, policy passage, and

stakeholder perceptions of influence, while open contestation, operationalized

with campaign contributions, is not. This suggests that operationalizations of

power based on surface-level measures such as campaign contributions may

not accurately capture interest group influence.

Keywords

interest group power, state politics, campaign contributions, lobbying,

education policy

It is unclear to what degree interest groups affect policy outcomes

(Baumgartner & Leech, 1998). Studies of group influence on policy making

in the United States produce mixed, and oftentimes null, results (Burstein &

1Harvard University, Cambridge, MA, USA

Corresponding Author:

Leslie K. Finger, Harvard University, 1737 Cambridge St., Cambridge, MA 02138, USA.

Email: lfinger@fas.harvard.edu

786723APRXXX10.1177/1532673X18786723American Politics ResearchFinger

research-article2018

Finger 853

Linton, 2002). This might be due to superficial conceptualizations of interest

group behavior. The community power debate of the 1950s and 1960s, which

proposed that groups can shape policy both directly, by fighting to shape

policy decisions, and indirectly, through agenda-setting, has been generally

ignored. Instead, scholars often use simple, surface-level measures of interest

group strength without rigorous attention to whether these are the best prox-

ies for interest groups’ ability to influence policy. Lowery (2013) writes that

it is not obvious given current approaches to studying influence that we can

recognize it in all of its complexities, perhaps sometimes mistaking evidence of

influence for something very different while ignoring deeper patterns of

influence when our surface indicators fail to detect it. (p. 8)

Current approaches to the study of interest groups suffer from theoretical

incoherence with regard to how interest groups seek to achieve influence and

a lack of consensus around how to operationalize these behaviors.

This article proposes that these inconclusive findings may be a result of

problems with the operationalization of interest group strength.1 It seeks to

take a step back and examine the interest group measures scholars commonly

use and ascertain whether they are valid. It does this by reviving the “two

faces of power” conceptualization to determine whether common measures

fully capture a group’s capacity to influence. The “two faces of power” model

of interest group strength suggests that interest groups exercise power two

ways: battling over policy proposals through lobbyists and campaign dona-

tions (the first face) or by setting the policy agenda before the battle even starts

(the second face). This distinction is often forgotten in studies of interest

groups and no scholar has sought to operationalize these two separate forms of

power, let alone examine whether they actually relate to policy outcomes. This

article seeks to undertake this task using the case of teachers’ unions.

Teachers’ unions have been consistently ranked by scholars as among the

most effective state interest groups (Thomas & Hrebenar, 2004), and they pos-

sess the characteristics envisioned to lead to influence—electoral involve-

ment, ample funds, lobbying presence, a large membership base, and so on.

Scholars have written that teachers’ unions gain these resources largely

through their institutional advantages, like the ability to bargain collectively

and automatic fee deductions (Moe, 2011), although recent work finds that

these institutions do not affect spending policy directly (Paglayan, forthcom-

ing). Because of their perceived strength and advantages, teachers’ unions rep-

resent an easy case in which interest groups should shape policy outcomes.29

Yet, as with other research on interest group influence, most studies do not

find teachers’ unions to be related to policy outcomes (e.g., Renzulli &

854 American Politics Research 47(4)

Roscigno, 2005; Shober, Manna, & Witte, 2006), and those that do find a rela-

tionship do so with uncertainty and for only some policy areas (Finger, 2018;

M. Hartney & Flavin, 2011). Nevertheless, because we would expect to be

able to empirically detect evidence of teachers’ union influence on policy,

teachers’ unions are a good case to examine whether measures of group power

are related to policy outcomes.

In order to test whether common measures of group strength enable policy

influence, I first operationalize the two faces of power. I show that campaign

contributions and membership rates are good proxies for teachers’ unions’

first and second faces of power, respectively. Next, I test whether these mea-

sures are associated with perceptions of influence and the proposal and pas-

sage of merit pay for teachers. I find that membership (second-face power) is

positively related to stakeholder perceptions of influence and policy out-

comes at all stages of the policy process. In contrast, campaign contributions

(first-face power) are unrelated to perceptions of influence or policy out-

comes at any stage of policy making. This may be because the use of first-

face power is sometimes defensive; interest groups may lobby or donate to

campaigns as a way to fight adverse policy proposals. Second-face measures,

which capture a group’s ability to set the agenda, are more consistently

related to influence on policy.

This study raises the importance of providing theoretical backing when

using interest group measures. Scholars have failed to consider whether their

interest group variables truly capture interest group strength, instead assum-

ing that activities that can be easily measured are enough. For many, this has

meant the use of campaign contributions or lobbyists, whereas for others, this

has meant the use of group membership rates (e.g., Daley & Garand, 2005;

Shober et al., 2006). This study shows that interest groups engage in a variety

of political activities and possess structural characteristics that may or may

not indicate strength. In particular, measures using visible behaviors may not

be able to distinguish between acts of strength and acts of weakness.

Therefore, scholars should be careful when operationalizing interest group

strength to consider whether the interest group behavior, characteristic, and/

or institutional advantage they are using adequately taps into a particular

group’s capacity to influence.

The Conceptualization of Interest Group Power

Where does interest group influence come from? The assumption among

scholars measuring interest groups is that interest groups affect policy as a

result of their direct interactions with legislators lubricated by campaign

contributions. The theory is that interest group lobbyists “purchase”