Civic Feedbacks: Linking Collective Action, Organizational Strategy, and Influence over Public Policy

Abstract

Democracy is premised on the ability of individuals, often working with others, to influence policies affecting them. However, existing theory cannot always explain why some organized efforts are more influential than others. We introduce the concept of civic feedbacks, arguing that the ways organizations engage individuals in collective action have feedback effects that shape the strategic position of organizations, the options available to leaders, and the likelihood of policy influence. The mechanisms through which civic feedbacks operate include the depth of accountability to the constituency, the network of elite relationships to which leaders subsequently have access, and their ongoing ability to recruit a committed and flexible constituency willing to engage new issues. Analyzing how these feedbacks redound to organizations over time enhances our ability to explain civic organizations’ differential rates of political influence. The concept of civic feedbacks returns organizations and organizational strategy to the center of the study of political influence.

A hallmark of any democratic regime is constituent influence over the policies that affect their lives. Constituents rarely affect policy alone, however. Their influence typically operates through vehicles of collective action—interest groups, civic associations, social movements, political parties, political campaigns, and networked communities (in this paper, we use the term “organization” to refer broadly to these vehicles of collective action). But not all collective efforts are equally successful. Why are some organizations better than others at achieving political influence, given a set of political opportunities and resources?

A venerable line of research explores the influence of collective action on public outcomes. Much of this work focuses on organizations’ available resources and the opportunity structures within which they operate. Yet there are many instances in which well-resourced efforts operating within favorable opportunity structures fail. Consider the effort to pass major climate legislation (specifically cap-and-trade policy) in the early Obama years. Despite unified Democratic control of government, an apparent ally across the aisle in Senator John McCain, a market-based policy design supported by deep-pocketed business interests, and a financially robust advertising campaign aimed at the public, the legislation failed (Skocpol 2013). Conversely, there are other cases in which grassroots efforts without many resources and inhospitable political opportunities succeed. For instance, poor farmworkers with few labor law protections organized and ultimately unionized with the leadership of Cesar Chavez in the 1960s and early 1970s, winning contracts and pay raises from major growers (Ganz 2009).

Existing theory, in other words, is incomplete. It underpredicts the potential for success among low-resource groups, or over-predicts the likelihood of success among the well-resourced. As a result, critics have called for better understanding of the pathways through which organizations can successfully leverage constituency-based action to influence the choices of political decision-makers (Anzia 2019; Pierson 2015). In particular, why do some organizations with substantial resources and seemingly conducive opportunity structures fail while others with few resources and dismal political prospects succeed?

We answer that call with a theory of participation and policy influence centered on organizational strategy and agency that we call civic feedbacks. We argue that civic feedbacks help us compare different forms of grassroots collective action to explain why some are more likely to succeed than others. What differentiates distinct forms of collective action? Organizations make many choices about how they want to engage participants in action. They can table outside grocery stores to ask passersby to sign petitions, or they can invite people to deeper forms of engagement, such as (virtual or in-person) meetings to discuss problems in their community. If we were just counting actions, one petition signature and attendance at one meeting each count as one action. Civic feedbacks, however, focus on the downstream consequences of those actions. Even if they generate the same amount of activity for the organization, the ultimate effects of each type of engagement differ. Those downstream consequences are civic feedbacks. Civic feedbacks explain why the different choices that organizations make about how to cultivate the participation of ordinary people can alter the strategic political position of those organizations, enhancing their ability to shift policy outcomes over time.

Civic feedbacks emerge through three mechanisms, based on three types of downstream consequences collective action can have for organizations. First, different forms of collective action vary in the development of constituent capacities: based on the ways they engage people in action, organizations can magnify individuals’ actions by enhancing their political consciousness, knowledge, commitments, strategic flexibility, and efficacy, and by connecting them with others with a shared sense of purpose. Second, forms of collective action vary in the extent to which they facilitate the recruitment and retention of others willing to engage new issues: embedding people in social contexts that cultivate their capacities fosters loyalty to the organization that facilitates ongoing recruitment—that goes beyond the usual suspects. Third, forms of collective action vary in whether they alter relevant networks of elite relationships: this rich cultivation of members with strengthened resources and commitment gives organizations an independent source of political standing, increasing their centrality as political players, and enhancing the network of elite relationship to which they have access. Through deep cultivation of the leadership and voice of their members—even those with few traditional resources—organizations can enhance their strategic position and strengthen their likelihood of achieving the policy outcomes they desire.

We define the concept of civic feedbacks and illustrate it with a case study of a grassroots organization that helped win a historic ballot measure in 2016 for early childhood education in Cincinnati, Ohio. In addition to showing how existing theory is inadequate for explaining observed outcomes in the Cincinnati case, we also demonstrate how civic feedbacks help explain a range of well-known historical cases. We conclude by discussing implications for future research.

Individual Participation, Organizations, and Policy Outcomes: Existing Models

A core function of democratic participation is exercising influence over political outcomes, often in concert with others. A large literature thus examines the extent to which constituency-based action influences policy (e.g., Andrews 2004; Baumgartner et al. 2009; Goss, Barnes, and Rose 2019; Hacker 1998; McAdam, McCarthy, and Zald 1996; Pierson 2004; Steinmo, Thelen, and Longstreth 1992). Figure 1 offers a stylized depiction of the interrelationships between individual participation, organizations, and policy outcomes, as articulated in this work and related literatures, along with our proposed amendment, civic feedbacks.

Figure 1 Civic feedbacks: A visualization of relationships among participation, organizational strategy, and policy

Note: The solid lines represent traditional relationships theorized in extant literature. The dotted lines represent the new civic feedbacks relationship.

Along Path A, a rich body of research has examined the relationship between organizations and political influence. This work shows how political opportunity structures, or the ex ante existence of indigenous mobilizing structures, network connections, media relationships, coalitions, material resources, and particular constituency traits, shape the ability of movements and interest groups to achieve their political goals (Meyer 2021; Amenta 2006; Amenta et al. 2010; Baumgartner et al. 2009; Goss 2009; Hansen 1991; McAdam, Tarrow, and Tilly 2001; Van Dyke and McCammon 2010; Staggenborg 1986; Wasow 2020; Gillion 2020). Many quantitative studies focus on the number of participants at a given event or the intensity of participation as a measure of the strength of collective action (Gillion 2013; Madestam et al. 2013).

Path B shows how these organizations affect individual participation (McCarthy and Zald 1977; Piven and Cloward 1977), by influencing people’s likelihood of civic activity throughout their lives (Corrigall-Brown 2012; Jennings 1987; McAdam 1989), their likelihood for activism within traditional (Goss 2009; Han 2014, 2016; McAlevey 2016; Rothenberg 1992) and digital-first (Karpf 2012) civic organizations, by fostering civic skills and social bonds (McAdam 1986; Milkman and Voss 2004; Munson 2009; Verba, Schlozman, and Brady 1995), or by activating participation through recruitment (Rosenstone and Hansen 1993).

The literature on democratic representation examines the relationship between participation and policy outcomes, by asking how responsive government is to the behavior of groups (Path C1) and individuals (Path C2) (Bartels 2008; Broockman and Kalla 2016; Canes-Wrone 2015; Gilens 2013). Finally, the policy feedbacks literature argues that policy outcomes themselves can affect mass political participation (Path D1) and organizations’ resources (Path D2) (Campbell 2003; Goss, Barnes, and Rose 2019; Mettler and Soss 2004; Pierson 1993).

As the existing theories depicted in figure 1 posit, civic organizations play two critical roles: they reach inward to shape the capacities and ways people participate (Path B) and they reach outward to shape the way participatory actions are converted into influence over outcomes (Path A). But existing approaches do not fully explain variation in organization success. One linkage missing from existing theory is the link from individual participation back to organizations along Path E. Civic feedbacks—a two-way process—refer to the iterations by which the nature of organizations’ development of individuals along Path B can feed back to alter organizations’ strategic alternatives on Path E.

Introducing Civic Feedbacks

Civic feedbacks is an iterative theory aimed at better understanding the mutually reinforcing relationship between the two core functions of civic associations. We argue that the strategic choices organizations make about how to engage constituencies (reaching inward) can be a bridge to their success in the public arena (reaching outward). We incorporate insights not just from political science and sociology, but also community organizing, urban politics, faith-based organizing, labor organizing, and elsewhere (Alinsky 1971; Ganz 2009; McAlevey 2016; Milkman and Voss 2004; Piven and Cloward 1977; Weldon 2011).

Civic feedbacks assume a repeat game, in which some forms of collective action expand the strategic choices available to the organization for later rounds of action. The concept of civic feedbacks borrows its iterative orientation from the policy feedbacks literature, in which policies change the strategic options of elite and mass actors, reshaping the political landscape for later rounds of policymaking (Campbell 2003; Mettler and Soss 2004; Pierson 1993). Civic feedbacks similarly assume that most organizations building collective action will have to engage in later rounds, because any kind of political change takes time. Organizations that develop civic feedbacks are constantly engaging their constituency in rich forms of action along Path B that alter the capacities of constituents by developing their political knowledge, commitments, efficacy, skills, relationships, and so on. Those altered capacities generate robust feedbacks along Path E, because an organization with a deeply engaged constituency base has more strategic tools in its toolbox—it has new relevance for and leverage over elites, and an increased ability to stand up for its interests, recruit more members, and pivot strategically as needed. Not all forms of engagement along Path B can have those feedback effects.

Civic feedbacks operate through three mechanisms. First, the development of constituent capacities shapes subsequent choices organizational leaders have regarding if and how to engage various political fights. Some forms of participation are better than others at developing people’s civic skills, political consciousness, motivation for ongoing participation, social relationships that sustain action, organizational loyalty, and so on. Organization leaders decide whether to engage particular political battles based on their assessments of the human resources they can mobilize in any given moment. The strength of available human resources depends on prior choices the organization made about what capacities to inculcate. We build on previous work (Han, McKenna, and Oyakawa 2021) to argue that three particular constituency characteristics expand the strategic choices an organization has: organizations with constituencies that are independent (e.g. not beholden to elite donors), flexible (in the goals and outcomes they are willing to pursue), and committed (to each other and the organization) will have more strategic choices compared to an organization with a more narrowly mobilized and episodic constituency.

Second, civic feedbacks shape organizations’ strategic choices by facilitating the recruitment of others willing to engage new issues. Political participation is inherently social and has spillover effects to others (Green and Gerber 2008; Rolfe 2012; Rosenstone and Hansen 1993; Sinclair 2012). An organization can engage constituents in action in ways that enhances these spillover effects and encourages participants to recruit others. Particularly when organizations engage with unexpected constituencies, such as encouraging left-leaning organizations to reach into traditionally conservative enclaves, or encouraging white-dominant groups to authentically engage race-class subjugated communities, these new participants can broaden the organization’s strategic options. Such expansions can also facilitate engagement with new issues, allaying concerns about organizational durability as initial goals are achieved and new ones must be embraced to keep the organization going.

Third, civic feedbacks can expand or contract the strategic choice set of organizations through the network of elite relationships they generate. Access to elite relationships is a key determinant of the power of advocacy organizations (Baumgartner et al. 2009; Hansen 1991). Such access is influenced by organizations’ ability to demonstrate a “recurrent competitive advantage” (Hansen 1991), which itself depends on the number of people they engage, the intensity of commitment they can consistently evince, or the geographic distribution or social position of their supporters. Influence arises from organizations’ ability to cultivate the kind of participation that matters to elites without getting co-opted, a frequent danger for movements (Strolovitch 2007).

In a repeat game in which organizations seeking to influence policy engage constituencies in collective action to pressure policymakers, civic feedback effects over time expand the strategic choices that organizations have. With deep cultivation of individuals along Path B, the Path E feedback loop builds a constituency with greater depth (through cultivation of capacities), breadth (through recruitment of others), and elite access (through expansion of elite networks). These mechanisms open up new strategic vistas for organizations. Particularly in uncertain political environments, such broadened strategic choice sets can provide organizations with the tools they need to respond to any given contingency. Civic feedbacks theory thus reconceives the bottom left of the figure 1 triangle as not just about organizations, but also about organizational strategy.

How are civic feedbacks different from existing models of collective action? First, the theory helps explain outcomes that existing theories cannot explain. Most theories of collective action are based on relatively static assessments of the resources and political opportunity structures an organization faces. Yet many outcomes do not correspond to the literature’s predictions (e.g., Skocpol 2013; Clemens 1997; Ganz 2009), and other research shows that the relationship between resources and influence is hardly linear or dispositive (Baumgartner et al. 2009; Bosi, Giugni, and Uba 2016). The concept of civic feedbacks makes a theoretical shift from viewing groups simply as aggregators of fixed resources (including individual actions) to examining them as dynamic strategic actors. Organizations make strategic choices about what kind of feedback effects they can cultivate to develop new, expanded resources for a repeat game. Civic feedbacks theory thus helps explain cases on Path A that existing theory does not. In particular, we show how groups, including those working with historically marginalized constituencies, can strategically reshape their potential power.

Second, civic feedbacks lay out the mechanisms by which groups can prospectively build their capacity for policy success. In this way, too, it moves past existing theory. The historical institutional literature on civic associations (Goss 2009, Skocpol 1992, 2003) emphasizes the importance of these organizations without probing the internal practices by which they build political power. The community organizing literature (Alinsky 1971; Ganz 2009) explores the micro-dynamics of organizing without showing how those practices build political power. Other work examining the relationship of strategy and strategic capacity as predictors of movement outcomes relies on retrospective, ex-post accounts (Ganz 2000, 2009; Mansbridge 1986; McAdam, Tarrow, and Tilly 2001; McCammon 2012). We extend that research to identify prospective, ex ante ways to predict which forms of collective action are most likely to lead to influence, and why.

Finally, civic feedbacks theory restores groups to the study of political participation and influence. Early studies of civic power (Lukes 1974; Gaventa 1980) and influence focused on such organizations (Bauer, Pool, and Dexter 1963; Key 1956; Truman 1951; Wilson 1973). However, the move toward behaviorism in the mid-twentieth century emphasized the role of individuals, treating collective action as the additive sum of individual actions. We argue that understanding the collective aspects of collective action necessitates also understanding the processes of articulation and amplification that underlie the way collective action becomes politically influential (De Leon, Desai, and Tuğal 2015). This theoretical turn not only enhances our ability to explain the relationship between political activity and political influence, but also restores the place of organizations in the study of political participation and policy outcomes.

Case Selection and Data Collection

To illustrate civic feedbacks, we analyze the successful fight for universal preschool in Cincinnati in 2016, and review several historical cases in light of this new theoretical approach. We identified the Cincinnati case through a multi-step process seeking to identify examples of otherwise marginal grassroots organizations exerting unexpected levels of influence. In other words, we sought “deviant” cases (Gerring 2008): outliers from existing theory that can be used to identify novel predictor variables and generate new explanations for the observed outcome. Such deviant cases bridge existing theory to new theory, generating new hypotheses for outcomes that old theory fails to explain.

To identify a pool of outlier cases that achieved political influence despite lacking ready access to the resources known to affect grassroots success, we drew on expert informants. We identified these informants through a prior research project, a representative landscape analysis of progressive civic engagement and collective action organizations in the United States (Han, McKenna, and Oyakawa 2021). We interviewed the twelve expert informants—philanthropists, organizational leaders, researchers, and others—in 2016, asking them to identify instances in which organizations had been able to translate the grassroots activity of their base into political influence. Twenty-one potential cases for study emerged, with ten cases coming up in three or more conversations.

We then explored these ten cases in depth, and identified one particular case—a fifteen-year campaign for universal preschool in Cincinnati, Ohio—as a “deviant” case for hypothesis generation. This case had within-case variation—two different entities within the broader coalition sought to win a policy initiative through collective action, but did so in different ways, with distinct feedback effects—enabling us to exploit longitudinal variation while holding the issue campaign, geography, institutional setting, and many other key characteristics constant. After developing the theory through the Cincinnati case, we then examine how the theory applies to other historical cases.

We collected a range of in-depth data on the case, including observations and interviews with twenty-nine stakeholders on all sides of the campaign. We also conducted a network survey of nineteen key coalition members from a range of public and private sector organizations who were identified as central actors in the campaign, ¹ asking respondents to identify, at two different time points, the people with whom they were exchanging information, sharing or aligning resources, strategizing, and negotiating conflict, giving us quantitative measures of shifts in the political dynamics of the campaign.

Most of our data was collected in 2017, after the campaign ended, and thus relies on retrospective recall by campaign participants and observers. Given potential biases, we sought, whenever possible, to triangulate interview and survey data with analysis of video and other primary and secondary documents produced in real time during the campaign, including interviews that had been conducted by other scholars, which they generously shared with us. In the end, we accumulated 1,495 minutes of interview tape, 681 pages of typed interview transcripts, 170 pages of typed field notes, and reviewed an additional 1,594 pages of primary and secondary source documentation, including three years’ worth of weekly reflections from the executive director of the anchor organization in our study, e-mail threads sent during the campaign from stakeholders, video recordings, campaign records (including expenditures, precinct-targeting, canvass walk lists and scripts, and board meeting minutes), newspaper articles, and policy reports.

A Case Study of Civic Feedbacks: Cincinnati Preschool Promise

In November 2016, Issue 44—a $48 million-dollar municipal levy that increased property taxes to support K–12 education and create a publicly funded preschool program in Cincinnati—passed with 62.2% support (even as Ohio voted for Donald Trump by a margin of eight percentage points). Despite raising taxes by $278 dollars per year for every $100,000 dollars of home value, the ballot measure won by 24 percentage points, the widest margin of victory for any new education levy in Cincinnati history.

The effort to develop a universal preschool program in Cincinnati had been ongoing for fifteen years with limited success. In the beginning, preschool proponents, led by business and philanthropic leaders, sought to engage collective action (along Path B in figure 1) without civic feedbacks (Path E). Although they enjoyed the abundant resources and political opportunities that would predict success according to existing theory, they were not able to achieve their goal. Only after a different congregation-based community organization cultivated a robust constituency along Path B was the campaign able to generate civic feedbacks along Path E, which enabled the successful ballot campaign. Unexpectedly, from the standpoint of existing theory, the congregation-based organization was a classic low-resource group: its members consisted of mostly low-income, Black residents. According to its I-990 tax filings, the organization’s total revenue in 2014 (when it began to work on preschool issues) was just over $100,000 and it ended the year with a deficit of about $27,500. In 2016, its total revenue was just under $300,000 and ended the year with a deficit of about $10,000. Despite these relatively limited financial resources, the group’s members prevailed in securing a universal preschool program on their terms, with a design that would benefit both the community’s poorest children and the predominantly Black preschool providers.

Phase I: Limited Civic Feedbacks in Cincinnati Preschool Promise’s Petition Campaign

Initially, the preschool campaign seemed destined for success. The campaign began when a police shooting of an unarmed Black man in Cincinnati in 2001 prompted a citywide discussion about combatting persistent and racialized poverty. Civic leaders settled on a plan to address what they saw as the root causes of the unrest with investments in early childhood education, among other programs. What later became known as the Cincinnati Preschool Promise (CPP) campaign emerged, with all the markers of a successful movement. The business community, led by prominent individuals such as a retired Proctor & Gamble CEO, partnered with the United Way of Greater Cincinnati and made a strong public commitment to reach 85% kindergarten readiness by 2020. The United Way and a business-led coalition raised $10 million in private money to fund preschool slots and research, increasing kindergarten readiness from 43% to 55% by 2008. But private fundraising stalled, with the campaign well short of its readiness goal. The campaign, now joined by a non-profit leader whose organization focused on improved educational outcomes, realized that achieving the 85% kindergarten readiness goal would require public funding.

CPP reached out to city council leaders asking for their support. Democrats who generally supported early childhood education constituted a supermajority on the city council, and they told CPP leaders to demonstrate that they had broad support to secure public funding. Thus, the preschool coalition pivoted to a Path B strategy designed to generate mass participation.

The CPP-led public campaign for preschool initiated a pledge drive in 2013 asking both local politicians and ordinary citizens to affirm their support for two years of preschool for every Cincinnati child. Launched with a professional logo and splashy press conference that observers said resembled a “presidential press gaggle,” the pledge drive collected 10,000 signatures, double the original goal. They gathered signatures by tabling outside grocery stores, attending community events and fairs, and trying to convince passersby to support a campaign for preschool.

But the campaign failed to achieve its policy goal. When CPP asked elected officials for public funding for preschool, city councilors—who themselves had signed the pledge to much fanfare—refused. Although CPP had the material and political conditions that would predict success along Path A—material resources; coalitional support within powerful business, elected, and civic networks; professional communications support; and a favorable opportunity structure (a Democratic-controlled city council)—their petition campaign had no feedback effects. The way they gathered pledges meant that they were not engaging any of the three mechanisms—they were not developing the capacity of the petition signers, the signatories did not have any spillover effects into recruiting others, and the petition did not lead to accountable elite relationships for CPP. When elected officials pushed back on the petition, CPP had no strategic tools in their toolbox to respond. They had not expanded their strategic choices. Path B, in other words, did not lead to any civic feedbacks along Path E, thus leaving CPP’s strategic position unchanged. The lack of civic feedbacks left the CPP with limited strategic choices.

Phase II: AMOS Develops Civic Feedbacks to Win the Ballot Initiative

In 2014, a new organization, AMOS, began to work on preschool issues in Cincinnati. At first, it was not part of the larger coalition that CPP represented. AMOS was a congregation-based community organizing organization with a new executive director, Troy Jackson. AMOS did not have a defined portfolio of issues; instead, it worked on issues that its constituency, a multi-racial community of faith across the city, identified as important. Upon arriving at AMOS, Jackson conducted more than 100 one-on-one conversations with AMOS community leaders and fielded a 2,000-respondent community survey. ² The meetings and survey revealed strong commitment to expanding preschool among AMOS constituencies to combat racialized childhood poverty.

Once they decided to take on preschool issues, AMOS could have simply mobilized its constituency for a citywide march or rally that demonstrated the breadth of existing support. Instead, they chose to invest in educating and organizing their constituency to demonstrate both breadth and depth along Path B. AMOS cultivated the capacities of leaders from the predominantly Black community which would benefit most from universal preschool, through a 150-person Leadership Assembly, and a 600-person public meeting in Fall 2014. Together, these meetings helped solidify a base of people within AMOS committed to working on preschool issues. Then AMOS created horizontal relationships and a sense of community amongst over 1,000 AMOS members though house meetings built around viewing a short video they created entitled, “Are We Crazy About Our Kids?” (Adelman 2014).

Even as AMOS began to collaborate with the CPP coalition, its leaders worked with their community members to articulate a set of policy principles that clearly stated AMOS’s interests distinctly from the interests of the broader coalition. They developed the People’s Platform, a statement of values in preschool policy that AMOS constituents wanted: minimum pay of $15 per hour, paid sick leave, and affordable health insurance for private preschool providers (who were likely to be drawn from the poorer, Black constituencies in Cincinnati); targeted resources for the poorest children and families to address racial disparities in the city; and the ability for parents to exercise voice in the design of the preschool program. The Platform was not just a piece of paper developed by AMOS staff. An examination of archival records, including meeting minutes, e-mails, and other communications reveal the deep levels of commitment, education, and understanding AMOS constituents had developed to articulate the platform. The platform was ratified in a public meeting of over 500 AMOS members in November 2015.

The feedback effects that resulted from the way AMOS cultivated its constituency became apparent when the principles of the People’s Platform were threatened. The threat emerged during a struggle over how to fund and structure the initiative that would eventually become Issue 44. AMOS preferred a new earnings tax and a standalone initiative for preschool, while the business and institutional power players in the CPP coalition preferred a more regressive property tax that would combine support for both preschool and the city’s K–12 schools. The decision to pursue a combined ballot initiative for an increased property tax was made in a March 2016 meeting that Jackson was unable to attend. Jackson was “livid” when he realized a deal had been cut in his absence, foreclosing consideration of the principles AMOS constituents had articulated in the People’s Platform and the funding mechanisms they supported.

At this moment of political challenge, AMOS had to decide how to respond to CPP’s decision to develop a plan that did not incorporate the principles its base had articulated. Jackson was facing intense pressure from coalition partners to acquiesce to the deal. Should AMOS give in? Or should they insist that the coalition to address the People’s Platform? In deciding how to respond, both Jackson and leaders in the CPP recognized the power of the constituency AMOS had built. The way Jackson had engaged people in the process of building the People’s Platform (Path B), in other words, had civic feedbacks (along Path E) that shaped its strategic position in this moment. Both Jackson and the CPP leaders recognized that AMOS had developed a constituency with the knowledge and commitment it needed to spring into action to call out the coalition if they disagreed with it (mechanism 1: development of constituent capacity). They knew this constituency had breadth throughout the city and reached constituencies CPP had not been able to reach before (mechanism 2: facilitating the recruitment of others into new issues). These feedback effects expanded Jackson’s strategic choices. Instead of acquiescing, he had the option to push back on the coalition.

Jackson refused to give in. Once Jackson refused, the coalition partners could have ignored him and tried to push forward with their plan. But they did not. Instead, they listened to his demands. Jackson insisted the CPP steering committee meet directly with AMOS community members to defend their choice of a combined levy, to react to the demands spelled out in the People’s Platform, and to give AMOS community members an opportunity to make their own choice about whether to stay in the coalition or not.

In two short weeks, AMOS was able to organize hundreds of its constituents to show up for a public meeting to hold CPP leaders accountable (evidence of mechanism 1 and 2: development of constituent capacity, and facilitating the recruitment of others). For two hours, AMOS’s constituents forced members of the city’s power elite—including the CEOs of some of the biggest companies in the city, such as the Children’s Hospital—to attend a meeting at one of the city’s oldest Black churches and to defend their choice of a combined levy and consider the planks of the People’s Platform (evidence of mechanism 3: altering the nature of elite relationships). Ultimately, the AMOS community voted to support the combined levy, but only after AMOS constituents succeeded in securing commitments for their platform to be included in the policy. AMOS and its constituents, despite being a low-resource group from the vantage of traditional theory, had leveraged its relationship with political elites to ensure its interests were reflected in the final ballot initiative, which prevented them from getting co-opted.

With buy-in from the AMOS constituents assured, Jackson and AMOS went on to lead the field campaign for Issue 44, the combined levy to fund preschool and Cincinnati Public Schools. They organized 750 volunteers and conducted a paid canvass to do the persuasion and mobilization work required to help the levy pass, financed in part by national funders attracted to the issue. AMOS knocked on over 50,000 doors, made tens of thousands of phone calls, and identified 10,000 new voter targets in favor of the levy. ³ It also built a groundswell of public support and developed new cadres of grassroots leaders, aiding in its overwhelming margin of victory.

Case Study Discussion

During the fifteen-year effort to bring universal preschool to Cincinnati, two different organizations within the city used two different approaches to building a grassroots constituency for preschool. Each effort had distinct downstream consequences for the strategic choices these organizations had. The contrast between the downstream effects of the pledge campaign in CPP’s first phase and AMOS’s base-building work in the second phase exemplifies the difference between organizations that do not generate positive civic feedbacks and those that do.

In the first instance, CPP used a pledge drive to try show the breadth of support for preschool. Even though it got twice as many signatures as it had promised, it gathered the signatures (along Path B) in a way that had no feedback effects (along Path E). Although CPP in Phase I had the trappings of an influential grassroots campaign—strong financial resources, widespread public support, well-networked grasstops leaders, and favorable political opportunity structures—it did not achieve its policy goals. When elected officials pushed back on CPP and refused to fund preschool, the CPP had no tools left in its toolbox.

In the second instance, by contrast, AMOS built its grassroots constituency through a series of house parties and community meetings that culminated in the People’s Platform. This effort had civic feedback effects that developed a deeply educated, engaged, and committed constituency (mechanism 1), built breadth throughout the city as people kept bringing other people into the campaign (mechanism 2), and gave AMOS access to key power players who were running CPP (mechanism 3). When faced with unexpected challenges to its power, such as the debate over the tax mechanism, AMOS had an expanded set of strategic options at its disposal because of the feedback effects. Instead of its leader just acquiescing to a deal that did not favor his constituency’s interests, they could push back as a collective.

Figure 2 provides evidence for the way the civic feedbacks that AMOS generated from its work reconfigured AMOS’s place in the power network. The figure reports the results from one of our network survey items asking respondents to identify the other members of the preschool campaign with whom they shared or aligned resources in 2013 (Phase I) and in 2016 (Phase II). As shown in the network graph on the left, in 2013, before AMOS was involved in the campaign, its leader and his allies were at the margins of the coalition of power brokers in the campaign. Jackson was mentioned by only four others in the network as someone with whom they shared or aligned resources, ranking tenth (of eighteen) on eigenvector values—an index that calculates the relative influence of a node in a network (Bonacich 2007). Two of the people who indicated sharing resources with Jackson in 2013 were clergy leaders associated with AMOS.

Figure 2 Cincinnati Preschool Promise resource sharing network, 2013 and 2016

Note: Fruchterman-Reingold layout, nodes weighted by eigenvector centrality

By the end of the campaign in 2016, however, the network graph on the right indicates that Jackson was at the center of those networks, brokering the flow of information and resources and negotiating strategy and conflict in the coalition (see Han, McKenna, and Oyakawa 2021 for more detail). In Phase II, the average eigenvalue of all nodes increased 47.4%, indicating more dyadic ties among actors as the coalition became more fully formed. Between Phases I and II, as shown in the change in size of the AMOS nodes, four of the six people whose eigenvalues increased the most were AMOS leaders, with Jackson’s centrality increasing the most of all: in Phase II Jackson had the highest eigenvector score on the resource-sharing measure, unmatched by anyone else in the network. After two years of organizing along Path B, the civic feedbacks along Path E shifted Jackson and AMOS’s positions in the power network, strategically positioned to make policy demands and to shape the language of the ballot initiative. Although the business interests had been, and remained, central, Jackson moved from being a marginal, small node to a large central one, positioned between the CPP and business “factions” in Phase II’s more coordinated network.

In addition to the network data, other evidence demonstrates the change in elite relationships (mechanism 3). CPS School Board minutes show that the political elites of the CPS Finance Committee described AMOS’ People’s Platform as “transformative” and the “bedrock of the Preschool Promise movement.” ⁴ Jackson’s choice to resist the coalition’s Steering Committee deal until elites answered to the organization’s platform put AMOS leaders in relationship with power brokers in Cincinnati in an unprecedented way. Although much of the underlying power dynamic of Cincinnati politics remains the same after the preschool effort, “We are now in the room when the big decisions get made,” Jackson noted. ⁵ After Issue 44 passed, city leaders continued to attend AMOS’ public meetings. ⁶ AMOS entered into close consultation with the United Way as it helped implement Issue 44, ensuring that agreements reached during the campaign were carried out.

Applying Civic Feedbacks to Other Cases

How might these findings apply to other cases? In table 1 we examine other historical and contemporary cases to investigate the generalizability of the theory. The table consists of a secondary analysis of key citations in the literature on salient social movements to determine whether there was clear, partial, limited, or no evidence of civic feedbacks. The table provides a brief accounting of outcomes at each movement’s peak period of influence and discusses the presence or absence of the three mechanisms of civic feedbacks.

Table 1 Applying civic feedbacks theory to additional cases of collective action

Sources: * Clemens 1997; Skocpol, Ganz, and Munson 2000

** Clemens 1997

† Schlozman 2015

† Green, Rozell, and Wilcox 2003

†† Heaney and Rojas. 2015, Tarrow 2005

§ Lacombe 2021

The table corroborates the civic feedbacks approach, namely that campaigns and movements that develop grassroots engagement without attention to how that engagement feeds back to develop constituency capacity, the recruitment of others, and the network of elite relationships are more likely to fail. By contrast, the historical cases that bear evidence of all three of the civic feedbacks mechanisms were, on balance, more successful.

Two organizations broadly considered to have been highly successful at their peak are the Grange and the National Rifle Association. Both showed clear evidence of dense, flexible, and relationally connected constituency bases with leaders in co-equal relationships of influence with powerbrokers in Washington and throughout the country. The Grange, the oldest agricultural movement in the United States, consisted primarily of poor farmers who collectively organized against grain-transport monopolies. As Skocpol, Ganz, and Munson (2000) observe of the “nationwide fraternity” that made up the Grange, there were robust linkages between thousands of local granges and regional leaders who “moved back and forth between Washington and [their] home regions” (534). Similarly, Lacombe (2021) writes of the NRA “new guard,” organizational leaders focused obsessively on developing “new ways to recruit members” even as they expanded their political operations and aligned with and infiltrated the Republican Party (20). The NRA’s potent combination of sustained mobilization capacity and partisan influence, Lacombe argues, helped turn it into one of the most powerful interest groups in America.

Middling cases of success include one of the main organizations of the early women’s movement (General Federation of Women’s Clubs, GFWC), the AFL-CIO in the New Deal period, and the Christian Coalition of America (CCA) in the late 1980s and early 1990s. In the AFL-CIO case, there was clear evidence of developing elite relationships but thinner evidence of the development of leadership capacities among rank-and-file members vis-à-vis movement targets (Schlozman 2015). Conversely, although the 850,000 GFWC affiliates across 16,000 clubs in 1955 were accorded “equal and independent terms of membership” (quoted in Clemens 1997, 196), they also eschewed and in other cases actively barred members from networking with elite power brokers. Instead, clubs focused on arts, literature, and politically neutral civic involvement. The decline of the CCA—once ranked the seventh most powerful interest group in America (Birnbaum 1997)—is an example of a social movement organization that experienced a boom and bust in member recruitment, declining to one paid employee and fewer than $25,000 in assets by 2017 (ProPublica 2018). These mixed cases highlight the fact that the mechanisms through which civic feedbacks operate are not fixed. They can develop, but later atrophy; and both their type and their level matter.

The cases with minimal or no evidence of positive civic feedbacks, which resulted in qualified failures, are the anti-Vietnam War movement and the anti-Iraq War movement. The anti-war movements failed to build a committed constituency and to realize mechanism 3 on elite relationships. Heaney and Rojas (2015) argue that the heavily Democratic post-9/11 protest movement in the United States counterintuitively demobilized when the Democrats won in 2008. Mere access to elites (in this case, presumed allies in the Obama administration and both houses of Congress) was not equivalent to influence over them and their foreign policy decisions. Further, the movement failed to achieve constituent flexibility along mechanism 2. Such flexibility not only signals a capacity to move nimbly across issues or rebound after a setback but also an ability to rethink strategy and escalate action if the political terrain shifts in the movement’s favor, which the post-9/11 antiwar movement failed to do after the 2008 election.

As these cases and our Cincinnati study illustrate, civic feedbacks theory draws attention to the qualitatively different ways in which organizations can engage constituents and to the downstream effects that that engagement has on ordinary people’s capacity to influence politics. When organizations develop constituencies with agency, they increase their own value to other political players and enjoy a broader strategic choice set—even when those constituencies begin with low levels of the traditional measures of resources and influence.

Future Research Agenda

We analyze a case that deviates from the outcomes predicted by existing theory to illuminate a new approach to the relationship between democratic participation and power. Probing civic feedbacks, we argue, can help account for outcomes that we have previously lacked the analytic frameworks to explain. We encourage a new research agenda that explores our hypothesized linkages between organizational strategy and individual political participation.

Testing the civic feedbacks theory. In Gerring’s (2008) typology, the Cincinnati preschool example represents a deviant case. Further research could examine additional cases of groups that deeply engage and develop the capacities of their constituents to see if they increase their strategic choice sets and expand their political options. The mechanisms probed here—development of constituency capacity, an ongoing ability to recruit and retain members willing to engage new issues, and a network of accountable elite relationships—could be tested as well. Or perhaps new mechanisms would emerge from new cases.

Determining scope conditions. To what types of groups does the civic feedbacks theory apply? If groups such as AMOS with minimal infrastructure are able to engage in the time-intensive process of building and engaging a committed base of constituents, then presumably grassroots groups with more money, time, members from dominant demographic groups, and preexisting elite connections could implement similar strategies. Yet higher-resource groups may be less likely to rely on civic feedbacks, opting to rely on other sources of influence to which they have access. Thus an open question is whether investing in civic feedbacks to expand strategic possibilities is a technique only needed by, or most likely to be used by, organizations with few traditional resources. Further research could help define the types of groups to which the civic feedbacks theory applies.

Sequencing. A third area for future research is the question of sequencing. Is a failed attempt relying on traditional resources a necessary stage for seeking an alternative path to policy success? Can a group such as AMOS only step in after more resourced and ostensibly more connected advocates flounder? ⁷ In addition, how do civic feedbacks change over time? What forces amplify or dampen feedbacks as these processes unfold? Further research should investigate how those forces evolve over time.

Political power operates in ways that are often hard to observe. We argue that civic feedbacks theory increases our ability to understand how it functions. In this dynamic view of influence and power over public policy outcomes, organizations can change their prospects for success by changing how they cultivate constituencies. We urge an appreciation for the depth and quality of organization, not mere counts of numbers, dollars, signatures, or participants. We uphold organizational strategy as a key variable to be examined in future research about the ways in which organizations can remake their political opportunities through their efforts to build and deploy their base. Finally, we call for the renewed study of organizations and organizational strategy as a factor in mass political participation, where it was once central in political science.