previous chapter
Exit And Public Voice In Representative Democracy
next chapter


2. Exit And Public Voice In
Representative Democracy

Under any economic, social, or political system, individuals, business firms, and organizations in general are subject to lapses from efficient, rational, law-abiding, virtuous, or otherwise functional behavior.

Albert O. Hirschman, Exit, Voice, and Loyalty

Democracy doesn't exist. It never has and never will. The distance between the democratic ideal and any actual government, past or present, is so vast that some theorists don't even use the term democracy to describe real political systems.[1] Given this gulf, one might question the value of examining alternative models of how democracy is supposed to work and, instead, ask to move directly to the practical matter of improving existing political systems. If the reader shares this skepticism, I beg indulgence, because it is difficult to assess and overhaul a system without first understanding how it is supposed to function in theory.

Examining different models of democracy is also worthwhile because this brief review reveals that every model of democracy gives at least some emphasis to electoral accountability and public expression. Competitive elections and the expression of political dissent are two ideas central to this book—the real threat of "exit" (in this case, rejecting an unresponsive public official) and the use of a public's "voice." Later in this chapter, I present a novel application of exit and voice to elections for public office. The present task is simply to demonstrate their general relevance to democratic theory. Readers may disagree about which democratic model America should or does, in fact, follow. If the central concepts examined in this book fit into diverse conceptions of the democratic ideal, then the insights in subsequent chapters should prove useful to a wide range of readers.



David Held provides a taxonomy of democratic models that range from classical Athenian democracy to modern capitalist democracy.[2] I shall examine many of these models to illustrate the roles of public voice and elections in each conception of the democratic process. Despite the dramatic differences in the designs of various democratic models, each is structured to ensure a strong public voice and elected representatives who are either accountable for their actions or relatively powerless.

Though quasi-democratic systems of local rule have existed at various times throughout ancient history, the Athenian model has left behind the most powerful legacy. This is partly because of the elaborate design of the Athenian political system and partly because written records of Athenian democratic political philosophy have survived. Athens's government mixed direct election of public officials with selection by lot and the regular rotation of public offices. Even those who were duly elected had short terms in office and, in most cases, were not eligible to run for reelection. This system did not ensure accountability, but it removed the problem of entrenched incumbency by imposing what are now called term limits. At least in its ideal conception, the Athenian system gave members of the public many opportunities to share their concerns with officials. Short terms, the wide distribution of public offices, and frequent meetings of powerful public assemblies were designed to prevent single individuals from having undue influence on policy decisions.[3]

The modern socialist model of democracy (not to be confused with actual systems such as the former Soviet Union) draws upon the Athenian tradition of face-to-face democracy. This ideal model uses many of the same institutions, but for large-scale systems, it replaces the direct assembly with a pyramidal council structure whereby the top-level planning body is connected through individual representatives back to thousands of lower-level, grassroots assemblies. In addition, this system places more emphasis on maintaining socioeconomic equality among citizens to ensure equal influence on policy. Public voice is the engine of this system, and the near-complete removal of representation by largescale elections makes the system dependent upon the quality of public ideas and their expression.[4]

Modern participatory democracy maintains direct avenues for public involvement in policymaking through referenda and initiatives. The

hallmark of participatory democratic models is the challenging and powerful role of the citizen. Citizens may not have much direct authority, but they are expected to be active in their local communities, political parties, and national policy debates. By eagerly and regularly engaging in public discussion, citizens may develop strong opinions and become skilled at the forceful expression of their values. Civic skills and habits will ensure that citizens have a clear voice and that they make their views known both directly and through reasoned candidate choices during regular elections of public officials.[5] Deliberative variants of the participatory model go a step further and encourage regular dialogue among citizens to bring their many views and voices together in search of an elusive moral consensus.[6] But even deliberative democrats recognize that the public's voice must be coupled with an effective electoral process to ensure that public officials remain accountable for the actions they take as authorities.[7]

Historically, relatively few large-scale political systems have modeled themselves on these three democratic ideals, partly because the models depend upon a highly motivated, skilled, and educated public. Many democratic theorists have held a more pessimistic view of the capabilities of the citizenry, and they have recommended a government that guards "the people" from themselves. In the Federalist Papers, James Madison advocated this "protective" model of democracy when he argued that "you must first enable the government to control the governed." In Madison's view, the public often forms self-destructive factions, so individual citizens should have only the power to vote for their representatives. Separate branches of government further insulate the political process from factionalism. Even though this model views the general public's voice as bothersome, Madisonian democracy still relies upon periodic elections to provide some measure of accountability.[8]

Madison's vision had a powerful influence on the design of the American political system, but modern American political theorists more commonly describe the U.S. system as resembling a pluralist model of democracy. The basic difference is that pluralists view factions not as dangerous but as "a structural source of stability and the central expression of democracy."[9] America's most influential student of democracy, Robert Dahl, developed the pluralist model in an attempt to describe the actual practice of American politics. Dahl argued that regular elections and political competition among diverse minorities and coalitions ensure a representative process. He later came to recognize

important differences between the pluralist ideal and the American political system, but his pluralist model remains largely unchanged. Like the Madisonian model, it relies on elections to ensure the accountability of public officials, but it also stresses the importance of public expression. If the people do not speak, pluralist government cannot craft policies that provide what the average citizen desires. By contrast, the Madisonian model views general public expression as white noise, and the more participatory models look for a "general will" among the public's many voices.

A final, capitalist model of democracy takes Madisonian and pluralist ideas to the logical extreme. The capitalist model of democracy assumes that individuals act in a way consistent with their own selfinterest in both economic and political life. Public officials are no exception to the rationality assumption, and they make political decisions that serve their interest in reelection. As a result, individuals and groups have political influence in proportion to their base of economic and political power. If these assumptions hold, then the political system should reach equilibrium so long as there are competitive elections to ensure proper political market performance. Even in this model, public voice still exists, albeit in the form of "market corrections" when officials misjudge the balance of power among competing interests. Regular, competitive elections are also pivotal, for without them public officials would begin to behave like unresponsive monopolies. The capitalist model of democracy is loathsome to many because of its unkind assumptions about human behavior and its reduction of public life to economic competition. Despite its alien form, however, even this purely economic model of democracy still has a place for minimalistic conceptions of voice and electoral action.[10]

Modern democratic theorists disagree about which of these models best describes the American political ideal and existing political institutions.[11] In practice, the American system is a hybrid of different models, with no one of them fully realized. Referenda and other ballot measures give citizens a chance to govern themselves directly, but the U.S. president is elected only indirectly through the Electoral College. Some egalitarian policies—from public education to campaign contribution limits—aim to equalize citizens' potential political influence, yet the most powerful economic interests wield considerable power in the political marketplace. Despite these complexities, it is possible to summarize both American political practices and ideals within a single abstract model, which the remainder of this chapter will develop.



When Albert Hirschman wrote Exit, Voice, and Loyalty, his primary goal was to help readers understand how both competition and the expression of dissatisfaction could improve the performance of failing companies, nonprofit organizations, and nations. Hirschman observed that firms and organizations often experience "repairable lapses" in performance, and if these go unnoticed, they can result in "permanent pockets of inefficiency and neglect."[12] A profit-seeking manager or wellintentioned president may not recognize the decline unless one of two things happens:

  1. Some customers stop buying the firm's products or some members leave the organization: this is the exit option. As a result, revenues drop, membership declines, and management is impelled to search for ways and means to correct whatever faults have led to exit.
  2. The firm's customers or the organization's members express their dissatisfaction directly to management or to some other authority to which management is subordinate or through general protest addressed to anyone who cares to listen: this is the voice option. As a result, management once again engages in a search for the causes and possible cures of customers' and members' dissatisfaction.[13]

Historically, people have used both exit and voice as response mechanisms. Many a firm has lost its customer base because it could not continue to provide the highest-quality product, and innumerable customers have tried to induce change through direct complaint. Many organizations have disappeared after their members found a better place to work, live, or play, but many members have worked for change from within an organization before giving up all hope and abandoning it.

Hirschman argues that whether people use the exit or voice option depends on their attitudes and circumstances. His model assumes that exit is the dominant reaction mode when better alternatives exist and switching products or organizations entails little cost. In a perfectly competitive market, consumers will always favor superior products and quickly exit any commercial relationship with firms that cease to produce the best goods available. Alternative products and organizations are not always present, however, and voice's role increases "as the opportunities for exit decline, up to the point where, with exit wholly unavailable, voice must carry the entire burden of alerting management to its failings." Exit is also more costly when it concerns "standardized durable consumer goods requiring large outlays" or, in organizational terms, when individuals have made a considerable investment in the organization.[14]

In these cases, exit loses its appeal, because desertion might require learning a new computer operating system or starting on the bottom rung in a new organization.

Before exercising their voices, however, people take into account a similar cost associated with expression. There are both the opportunity cost of forgoing the exit option and the "direct cost of voice which is incurred as buyers of a product or members of an organization spend time and money in the attempt to achieve changes in the policies and practices of the firm from which they buy or of the organization to which they belong."[15] Using an economic example, the cost of driving to a new grocery store with lower prices is lower than taking the time to compose and effectively deliver a complaint to one's previous store, not to mention the savings that are lost while waiting for the old store to respond by lowering its prices or improving the quality of its goods and services.

Because of these costs, the regular exercise of voice often depends upon a modicum of loyalty to a firm or organization. "As a rule," Hirschman argues, "loyalty holds exit at bay and activates voice." In the absence of loyalty, consumers will quickly switch from an inferior merchant, but if they like a particular store's produce, they may first complain to the grocer before shopping elsewhere for their leafy green vegetables. Such loyalty is not simply a matter of "faith." In comparison with acts of pure faith, Hirschman explains, "the most loyalist behavior retains an enormous dose of reasoned calculation." People will be more likely to exercise their voice and develop loyalty if they perceive that such actions have an impact over time. If loyal customers' voices go unheeded when they complain to the grocer about inferior produce, their loyalty will decline, leading to their eventual exit. Although not necessarily one, the loyalist "looks like, or turns out to be, a sucker" when his or her attempts to salvage a declining firm or organization fail.[16]

Even loyal customers or members may lean toward the exit option if they do not recognize a clear means of exercising their voice. "The propensity to resort to the voice option," Hirschman explains, "depends also on the general readiness of a population to complain and on the invention of such institutions and mechanisms as can communicate complaints cheaply and effectively."[17] In other words, an organization or firm will only receive feedback via voice if it has in place a straightforward complaint mechanism that its member-customers are accustomed to using. Exit is a relatively automatic process in this model, whereas voice

requires social systems to elicit and properly channel internal dissent.

In sum, Hirschman's model demonstrates that people can respond to declines in product or organizational quality through exit and voice. Dissatisfled people will use the exit option when there are clearly superior and available alternatives and they lose little investment by switching products or affiliations. Exit will also be the favored response when exercising voice involves high opportunity costs and direct costs. People will make their voices heard, though, if they have developed loyalty to the offending organization based upon its perceived responsiveness. Resort to voice will also be more likely if efficient complaint mechanisms are available and familiar.[18]

Hirschman uses this model to derive some surprising insights about both economic and political behavior. In the thirty years since the model was introduced, Hirschman and other scholars have found even more ways in which the model applies to local and national politics. Only some of these observations have relevance to my own use of the model in this book, but it is useful to review those insights to better understand the original purpose and power of the model.[19]

When one considers the interplay of exit, voice, and loyalty, it becomes apparent that some of the fundamental assumptions of neoclassical microeconomic theory are false. Just as democratic theorists have their ideals, so do economists. The difference is that political philosophers (normally) have the modesty to recognize their ideals as just that—ideal types that one dare not dream will take solid form. Many economists, by contrast, begin with the ideal model of a perfectly competitive economy and then presume that the mythical "invisible hand" sweeps markets free of inefficiencies. If consumers and entrepreneurs behave rationally in pursuit of their self-interest, competition will ensure the maximization of their interests, and the dysfunctional firms that fascinate Hirschman will cease to exist.

Economists and many other social scientists accept the general tenets of this idyllic model.[20] Hirschman's writings, however, explain why this theory often fails to explain complex economic realities. First, actual competitive economies can, paradoxically, prove less satisfying to consumers than those with only a handful of firms. As Hirschman explains,

A competitively produced new product might reveal only through use some of its faults and noxious side-effects…. Competition in this situation is a considerable convenience to the manufacturers because it keeps consumers from complaining; it diverts their energy to hunting for the inexistent improved

products…. The manufacturers have a common interest in the maintenance rather than in the abridgment of competition—and may conceivably resort to collusive behavior to that end.[21]

Simply put, a multiplicity of equally unsatisfactory options is better for a producer than a monopoly. Producers will still scratch and claw the competition to lure more temporary customers to buy their products, but they understand that so long as no one produces a superior product, none of them need change their manufacturing processes.

A dash of competition may also be welcomed by a monopoly that knowingly produces a substandard good or service. Hirschman argues that inefficient monopolies actually invite some degree of competition "as a release from effort and criticism" when their power "rests on location and when mobility differs strongly from one group of local customers to another. If … the mobile customers are those who are most sensitive to quality," their exit permits the monopolist "to persist in his comfortable mediocrity." The "lazy monopolist may actually have an interest in creating some limited opportunities for exit" because the most quality-conscious customers are also "likely to be most demanding and querulous, in case of any lowering of standards." By showing them an open door, the monopolist encourages their exit and thereby reduces the chance that they will voice harsh criticism that could diminish the confidence and satisfaction of other consumers.[22]

Just as the absence of exit makes a lazy monopolist nervous, healthy competitive firms should dread the silence of a market where consumers speak only with their checkbooks. If consumers quickly switch products in response to even minute shifts in quality or price, a declining firm will be "wiped out before it will have had time to find out what hit it, much less to do something about it." Among other factors, Hirschman argues, it is primarily loyalty that causes influential customers and organization members to "stay on longer than they would ordinarily, in the hope or, rather, reasoned expectation that improvement or reform can be achieved ‘from within.’ " This measure of "irrational" loyalty conflicts with the ideal of perfect competition, but it actually results in a superior outcome because it gives firms the opportunity to correct themselves before collapsing. The ultimate threat of exit ensures long-term efficiency, but the willingness to use voice prevents the waste of invested resources that results when an otherwise rational firm briefly missteps.[23]

These three insights into illusory competition, lazy monopolies, and the value of loyalty apply to the political realm as well as to the marketplace. Regarding his description of false competition among equally

inefficient firms, Hirschman notes that "competitive political systems have frequently been portrayed in just these terms. Radical critics of societies with stable party systems have often denounced the competition of the dominant parties as offering ‘no real choice.’ " Though Hirschman recognizes that noncompetitive political systems are hardly agents of rapid social change, he acknowledges that "the radical critique is correct in pointing out that competitive political systems have a considerable capacity to divert what might otherwise be a revolutionary ground swell into tame discontent with the governing party.[24]

Public agencies and institutions also sometimes have the same incentives to promote modest competition as do lazy monopolists in the private sector. Hirschman offers the example of public schools:

Suppose at some point, for whatever reason, the public schools deteriorate. Thereupon, increasing numbers of quality-education-conscious parents will send their children to private schools. This "exit" may occasion some impulse toward an improvement of the public schools; but here again this impulse is far less significant than the loss to the public schools of those member-customers who would be most motivated and determined to put up a fight against the deterioration if they did not have the alternative of the private schools.[25]

For political processes, as for markets, the optimal balance then is a healthy mix of competition and dissent. Just as a perfectly competitive economy is too quick to destroy firms that err, so can highly competitive multiparty politics promote underdeveloped political parties. With an abundance of parties to choose from, "members will usually find it tempting to go over to some other party in case of disagreement. Thus, they will not fight for ‘change from within.’ "[26]

These few observations are only some of the insights into political life that one can glean from Exit, Voice, and Loyalty. Hirschman's work provided other observations about the dynamics of two-party systems, ideological bias, and overreliance on exit and voice. Subsequent essays by Hirschman and other scholars have applied the theory fruitfully to urban neglect, party competition in Israel, Japanese party politics, and the 1989 revolution in the German Democratic Republic. In addition, one scholar has applied Hirschman's ideas to political reforms aimed at democratic empowerment to criticize reforms that fail to balance exit and voice.[27] The same concern is central to this book, but before I use Hirschman's model, I wish to make substantial changes in its details and level of analysis. The remainder of this chapter introduces these modifications.



Few scholars have extended the exit, voice, and loyalty model to the political process, beyond its role in shaping individual political organizations. An exception is Eva Sorensen, who recognizes the potential value of redefining "exit" in relation to a representative government:

If we take a closer look at political life, exit is limited by the obvious fact that it is difficult to exit from the nation-state. A "genuine exit" calls for a change of nationality. Nevertheless, exit plays a central role in political life in liberal democracies in which the primary means of empowering the citizens is voting. Through voting the individual is granted the power to exit one party to the advantage of another…. Voice channels … are available through membership of a political party and through participation in the public debate in the media.[28]

I go beyond the traditional exit, voice, and loyalty model in another sense. With only minor exceptions, the remainder of this book examines how citizens influence individual elected officials, as opposed to political parties, public agencies, or governments.[29] This shift from organizational behavior to the actions of public officials is substantial, because exit and voice are now understood to influence the behavior of individual representatives, rather than complex organizations. In practice, many elected officials operate as the heads of small organizations that include paid staff and informal advisers. In the interest of simple prose, however, I shall refer to these officials as individual actors who respond (or fail to respond) to the voice of the electorate.

If the main actors in this new model are elected officials, the central question is what causes representatives to experience what Hirschman might call "lapses from virtuous behavior." The democratic models reviewed earlier all agree that the first task of political representatives is to serve the public. Even the Madisonian model of democracy recognizes that the lay public should be the ultimate judge of whether its representatives are acting in the public's best interest. The variance in democratic theorists' confidence in the quality of public judgment is somewhat moot, as every democratic system uses elections of one kind or another to prevent representatives from becoming entrenched and unresponsive. In every democratic ideal, public officials must be held accountable for their actions as representatives of the citizenry. Whether one believes in a general will of the people or simply in an average interest among conflicting visions of the public good, it is important to

understand the circumstances under which elected officials represent the public's interests.

To influence the actions of their elected officials, voters can use exit, voice, or both response mechanisms. In this context, the terms exit and voice take on a different meaning, and in place of the former I shall speak of "electoral rejection." As I use the term throughout this book, attempted electoral rejection is voting for an alternative candidate when a public official seeks reelection. By backing an opponent, a voter seeks to discard the current official and replace him or her with a candidate perceived to better represent personal or collective interests. Just as a consumer switching products has no assurance that such a change will prove beneflcial, the voter takes a chance on a new official. Unlike the consumer changing shampoo brands, though, the voter cannot change officials autonomously; instead, the success of the individual's vote depends upon the votes of others. Only when the electorate is thought of as a body does electoral rejection carry with it the certainty of successful exit: thus, collective rejection is the successful election of an opponent who challenges an incumbent. I move from the individual to the collective level when I refer to electoral rejection, and I stress the uncertainty of the individual voter's attempt to reject an incumbent official or administration.

In part because of the limited influence of a single voter's electoral choices, the public's capacity for dissent is at least as important as its attempts at electoral rejection. As I use the term, voice is an attempt to influence the public decisions made by elected representatives. Hirschman's more narrow use of the term would equate voice only with the expression of dissent against actions that a citizen finds distasteful. I wish to use the term more broadly to include praise, neutral input on a specific issue, and requests for specific votes that are not sparked by the belief that the official will act otherwise. Any citizen's message to a public official may influence that official's actions in the future, whether or not the citizen intended to change the official's decisions.

Whether a given voter chooses to rely upon this voice or backs an opposing candidate

in an election depends upon the same considerations presented in the earlier discussion of firms and organizations. Some of these factors become more complicated, but at a high level of abstraction, Hirschman's economic model and the voting model I present are quite similar. Whether a dissatisfled constituent voices dissent or attempts electoral rejection depends primarily upon past experience using voice with an incumbent and the estimation of an opposing candidate's potential for superior representation. Simply put, if a voter perceives that the incumbent has responded well to dissent and the opponent is not much better, it is likely that the voter will remain loyal to the incumbent and rely upon voice to express any dissent. By contrast, if the incumbent is unresponsive and the opponent would better represent the voter's interests, it is more likely that the voter will attempt electoral rejection by voting for an opposing candidate.

Without taking any other variables into account, one can see clearly the costs and benefits of loyalty in this model. If voters in a given city council district have very little loyalty toward their incumbent councilor, they will reject the incumbent in the next election so long as an opponent seems to have greater potential to represent their interests. A sitting councilor from this district may rarely hear constituents' voices, because of their unwillingness to voice concerns, and this lack of communication may decrease the councilor's chances for reelection even further. Over time, because of its readiness to reject imperfect officials, this district would also forgo the benefits of long-term incumbency. By contrast, a district with extremely high constituent loyalty would suffer the opposite fate. This second district would reelect councilors to many terms in office, but overreliance on voice would undermine the electorate's credibility and permit the incumbent to act with impunity. The sitting councilor might hear many complaints, but if the district's loyalty was blind, dissent would manifest itself only as voice and never as a real electoral challenge. Between these extremes, a healthy public develops a modicum of loyalty toward responsive elected officials but conditions that loyalty on actual performance.[30]


Though it may not be obvious, this simple model relies upon two critical assumptions. This model assumes that voters know what government actions are in their own best interest, and it assumes that voters know a great deal about the intentions and actions of incumbents and challengers. In reality, voter self-awareness and political knowledge vary tremendously over time and across different social groups, so it is preferable to change these underlying assumptions into model variables. This makes the model more complex, but it also highlights the importance of two forms of public deliberation. For citizens to exercise their voice and vote effectively, it is necessary that they have sound insight into both their own interests and the virtues of competing candidates.


Because they depend upon clarification of interests and the articulation of policy judgments, the expression of an authentic public voice and well-informed attempts at electoral rejection are only likely to occur when they follow a period of sustained deliberation. I define public deliberation as discussion that involves judicious argument, critical listening, and earnest decision making. Following the writings of John Dewey, full deliberation includes a careful examination of a problem or issue, the identification of possible solutions, the establishment or reaffirmation of evaluative criteria, and the use of these criteria in identifying an optimal solution. Within a specific policy debate or in the context of an election, deliberation sometimes starts with a given set of solutions, but it always involves problem analysis, criteria specification, and evaluation.[31]

Lest this process sound too tepid, I should stress that the deliberative discussion of problems and solutions can include emotional appeals. As Jane Mansbridge insists, "We must avoid the traditional, frequently male, mandate to ‘keep emotions out of it.’ … Appeals for the common good require an emotional and cognitive probing of one's own feelings of empathy, admiration, revulsion, or horror." Mansbridge recognizes that "appeals to emotion can be dangerous," and she suggests that "emotional appeals must therefore stand up to reflection in tranquillity."[32] The point is simply that the expression of strong feelings has its place in deliberation.

Having defined public deliberation, I now wish to clarify the meaning of democratic deliberation, a term that further specifies the ideal relations among the people taking part in a discussion within a democratic political system.[33] To be democratic, deliberation must include diverse participants from the larger public, and it should use an egalitarian decision rule (e.g., consensus or majority rule) to resolve conflicts among participants. Some deliberative theorists, such as Joshua Cohen, argue that "ideal deliberation aims to arrive at a rationally motivated consensus," but striving toward consensus does not require the use of a strict consensus procedure, which gives each participant veto power over any collective decisions.[34]

In addition, participants in democratic deliberation must have equal and adequate opportunities to speak, and they must be able to comprehend what other participants say. Participants in democratic deliberation also have a responsibility to avoid manipulative discourse, provide other participants with any relevant knowledge they possess, and consider carefully what others say. Using Jürgen Habermas's terms,

participants in an "ideal speech situation" must have adequate opportunities to examine the meaning of one another's statements and challenge one another's "validity claims."[35] If all of these conditions are met, a discussion can be called both deliberative and democratic.

The basic purpose of deliberation is to make sound decisions. Though the philosopher John Rawls used the term sparingly in his 1971 treatise A Theory of Justice, deliberation is the central mechanism in his method of evaluating public policy. Rawls's influential moral theory argues that people can judge a policy by imagining that they are unaware of their actual social position. Behind this hypothetical "veil of ignorance," people can objectively evaluate the degree to which a policy protects the public's basic freedoms and ensures a decent quality of life for the least well-off.[36] Rawls argues that such a philosophical exercise is most fruitful when conducted in a deliberative process:

We normally assume that an ideally conducted discussion among many persons is more likely to arrive at the correct conclusion (by a vote if necessary) than the deliberations of any one of them by himself…. Discussion is a way of combining information and enlarging the range of arguments. At least in the course of time, the effects of common deliberation seem bound to improve matters.[37]

Rawls adheres to the commonsense notion that many minds are better than one at flnding an optimal solution to a problem. Though public skepticism about the value of committee decisions persists, Rawls's faith in group discussion appears to be well founded. Research in social psychology and small group communication has found that, on balance, group discussion generally results in higher-quality decisions than do methods that rely upon single individuals. Moreover, past research has shown that group discussion leads to better decisions than noninteractive methods of collective decision making. In sum, groups produce better average decisions than individuals, and groups that engage in deliberation outperform nominal groups that simply pool individual opinions without open-ended discussion.[38]

Deliberation is even more important for complex social and political problems. Unlike technical or scientiflc puzzles, public policy problems are inextricably interconnected, difficult to deflne and study, and impossible to remove completely. The evaluation of alternative solutions to public policy problems requires value judgments, as well as technical analysis.[39] The latter feature is most important: political deliberation is not only valuable as a means of grappling with complexity, it also serves democracy by helping citizens clarify the implications of their

basic values for public policy choices. The more enlightened a public's understanding of its own interests, the more likely it is to govern itself effectively.[40] Disentangling the beliefs and motivations underlying surface-level policy preferences is difficult, and deliberation can transform disagreement by reducing initial opinions to sets of contestable "presumptions," only some of which will stand up under scrutiny.[41]

Amy Gutmann and Dennis Thompson make a slightly different argument for deliberation. Though these authors recognize the virtue of discussion as a means of pooling information and developing citizens' interests, they emphasize its value as a legitimate method of addressing moral conflicts. "Deliberation," they argue, "is the most appropriate way for citizens collectively to resolve their moral disagreements not only about policies but also about the process by which policies should be adopted." Gutmann and Thompson criticize those who would leave moral issues for the courts to resolve or bar their entrance into the political realm altogether. In their view, political deliberation is the only fair and effective means of handling moral conflict. Deliberation requires both citizens and representatives to present reasons and justiflcations for their views and to consider alternative views. When successful, deliberation can confer legitimacy upon even majoritarian policy decisions, so long as those decisions take conflicting views into account. Even if it does not produce a solution acceptable to all, it may at least preserve mutual respect among the parties to an unresolved debate.[42]

Relating these advantages to the exit, voice, and loyalty model, deliberation can strengthen the public's voice, ease its ability to collectively reject unrepresentative officeholders, and inspire a modest loyalty to responsive elected officials. A strong public voice requires that citizens develop their interests in relation to policy and articulate those interests clearly. Public deliberation has precisely those qualities when it takes place in an open setting and is inclusive of a diverse sample of the larger public. At least those directly participating in deliberation are likely to sharpen their own understandings of public policy problems and the solutions that best reflect their basic values. If public officials learn the results of the deliberation, the process can also serve as a means of clearly expressing the participants' judgments to elected representatives.

Having developed their interests more fully, participants in a deliberative discussion are better able to evaluate candidates effectively. Once voters know how they view an issue, they are more likely to examine

candidates' positions and actions on that same issue. Unrepresentative public officials stand out under such scrutiny. Moreover, to the extent that open deliberation orients discussion toward the common good, such discussion increases the likelihood that voters will support policies that they believe will serve the general public's interests. If those beliefs develop simultaneously during collective deliberation, voters will be more likely to reach shared judgments, and that may, in turn, lead to similar candidate evaluations. If more evaluations are in sync, then voters should more often act in concert and collectively reject unrepresentative elected officials. Simply put, deliberation might lead voters not just to collective policy judgments but also to collective electoral action opposing (or supporting) incumbent officials, depending on the correspondence of those officials' actions to the voters' shared interests.

At the same time, an electorate that engages in deliberation with its public officials might develop a stronger loyalty toward those officials who demonstrate not only responsiveness but also respect for the deliberative process. Democratic citizens perceive deliberation as a legitimate means for arriving at public decisions, and officials who conflict with majority opinion but explain their views and consider voters' positions might withstand opposition during elections because of their willingness to deliberate. As suggested earlier, a measure of reasoned loyalty beneflts the representative system by keeping in officials who might misstep from time to time yet remain genuinely responsive to voters' concerns.


Deliberation's value extends beyond the reasoned judgment it permits, but to understand this additional virtue, it is necessary to make one final addition to the revised exit, voice, and loyalty model. The effective use of a citizen's voice and vote also depends upon a set of psychological variables necessary to sustain democratic political action. In chapter 1, I argued that healthy governments require public trust and that meaningful deliberation can bolster the citizenry's confldence in its political institutions. In addition, deliberation can change how citizens view themselves, their abilities, and their responsibilities. To this point, the revised model has assumed that citizens recognize the value of participation and have confldence in their abilities to speak and vote wisely. As with deliberation, it is worthwhile to remove this assumption and consider these psychological characteristics as variables in the model.


The effective use of political voice and vote depends upon the maintenance of democratic institutions that reinforce important public attitudes and beliefs. When authors write about the importance of a strong "civil society," some of what concerns them is the public's willingness and ability to use its opportunities for political expression and electoral action.[43] Hirschman calls this the "readiness" to exercise voice; citizens are ready to voice dissent or attempt electoral rejection when they have both the necessary skills and confldence in those abilities. In the case of public voice, citizens cannot express their concerns effectively without basic literacy and a modest public-speaking ability. In addition, citizens must know how to reach public officials, whether they seek a face-to-face audience or simply want to send a letter. Anyone reading this book surely has those modest skills, but it is important to remember that over 20 percent of adult U.S. citizens have only rudimentary literacy skills, and only a quarter of the citizenry know the names of their U.S. senators.[44]

Also, people often develop skills without gaining confldence. This phenomenon is so widespread among people of all backgrounds that psychologists have given it a prominent place in the literature on human behavior. Self-efficacy is the belief that one can competently perform an action, such as brushing one's teeth, defending oneself against an attacker, or writing a letter to the editor.[45]

Despite its general signiflcance, self-efficacy is not always an important influence on people's behavioral choices, and its relevance to political life can not be taken for granted. Research in political science, however, has found a clear and consistent impact for this variable. These studies have demonstrated a strong connection between a person's political self-efficacy and his or her willingness to vote and express dissent through both conventional means, such as letter-writing and demonstration, as well as more radical means, such as civil disobedience.[46] The readiness to use one's voice, then, depends not only upon objective skills but also upon one's subjective assessment of those same skills.

Complicating matters further is the difference between one's own perceived abilities and the skills one attributes to fellow citizens. Obviously, a person's confldence in his or her ability to express dissent is unshaken by any doubts as to other people's communication skills. But one's sense of group efficacy is important when one considers expressing dissent through a group or organization.[47] A good deal of political expression comes through collective entities, and if citizens have doubts about the competence of groups they have joined (or are considering

joining), they are less likely to try channeling dissent through those bodies. Why try making oneself heard through a citizen action group if one suspects that the group is incapable of drafting a coherent press release, let alone of reaching agreement on what its members wish to say? A modicum of group efficacy is necessary to maintain active participation in community associations, public forums, and interest groups designed to channel public voices to policymakers. In addition, without a related sense of group efficacy (i.e., confldence in the competence of the general electorate's ability to conduct an equally thorough candidate screening), a talented and self-assured voter might decide that an attempt at electoral rejection is not worth the bother.

Even with a strong sense of self- and group efficacy, citizens may still not exercise voice, believing that their actions, although competently performed, will have no impact. Cognitive psychologists refer to this belief as "outcome expectancy"—the expectation that one's actions will result in a desired outcome. Hirschman underscores the importance of this expectation by arguing that loyalty ultimately depends upon the expectation that one's voice will have influence.[48] Past research has shown that in politics, as well as other spheres of action, a person's outcome expectancy is a powerful predictor of whether he or she will take actions that require effort.[49] Only if citizens believe that public officials will act upon their advice will they go to the trouble of writing, calling, e-mailing, or directly addressing elected representatives.[50]

Political scientists call this belief "perceived system responsiveness." It bears some correspondence to the political trust discussed in chapter 1. With regard to voting, one may or may not trust that casting a ballot for a candidate is likely to influence the outcome of an election. As for voice, citizens vary in the degree to which they trust elected officials to give a meaningful response to one's expressed concerns. In these ways, low public trust is one of the important beliefs underlying sustained political action.[51]

If public expression and electoral rejection are so difficult, why do people take political action at all? To answer this question, I focus on the relatively simple act of voting. Voters unsure of the impact of their vote might still do so. Some voters might make a pessimistic assessment of their ability to make an accurate and influential voting decision yet vote nonetheless because they place such a high value on the influence that they might achieve. This consideration, which psychologists call "outcome valuation," is regularly paired with self-efficacy and outcome expectancy in research on behavioral choice. With regard to voting,

however, this belief hinges on the unspoken hope that one's single vote will be the one that "makes the difference," and save in the smallest local elections, that is closer to an act of faith than a rational calculation. Research on voting in the United States suggests that the more common motivator to vote despite one's sense of helplessness is the lingering sense of civic duty that many citizens feel. Whether one thinks of duty as a behavioral habit, a cultural reflex, or a heartfelt conviction, it appears that many people continue to vote "because that is what citi zens do.[52]

Once again, this set of political beliefs and orientations is important because it undergirds sustained citizen participation in public affairs. Democratic political institutions must "shape the psychology of citizens," because the system depends upon the involvement of those citizens. Deliberation, the expression of a strong public voice, and effective electoral action all require a citizenry with self-confldence, some confldence in system responsiveness, and genuine concern for the outcomes of political actions.[53]


And what of those who neither place a high value on their political influence nor glow with civic pride? If exit is daunting to this group of citizens, do they then turn to voice? Perhaps some idle cranks fall into that category, but many more citizens use neither exit nor voice. A subset of these know that their interests are not well represented, and another subset have not even made that initial calculation. It may come as a surprise, but Hirschman's original model had nothing to say about this residual group of individuals who have opportunities for exit and voice but do not use them. One group of scholars using the model to study urban politics quickly discovered this group of citizens. In their view, failure to respond when perceiving political problems amounts to system neglect. This term underscores the fact that the use of neither exit nor voice is still a behavioral choice—the choice to neglect the system through nonresponse.[54]

Civic neglect is easy to recognize. When neglect becomes a prevalent response mode, one can expect large percentages of a public to stop voting altogether. These nonvoters are unlikely to follow politics closely. They skip political sections of the newspaper, avoid political television programming, and even eschew policy- or campaign-oriented political conversation. When asked about their views, citizens neglecting

the system will demonstrate limited political knowledge and undeveloped opinions, but they will demonstrate strong distrust of government and elected officials.

When open and egalitarian political institutions are in place, civic neglect can still become widespread if citizens lack the sense of efficacy and motivation necessary to use those institutions effectively. Even when the public is brimming with confldence and civic responsibility, neglect might still result from a sober assessment of the difficulty of making the sound judgments that underlie effective voice and voting choices. In a typical political system, such "rational neglect" is likely, given the difficulty of arriving at deliberative judgments. The rational utility-maximizing denizens of economic models are not only presumed to doggedly pursue their self-interest, they are also assumed to know their interests. By contrast, in the model of representative democracy presented herein, a fundamental problem is the discernment and articulation of citizens' policy judgments. Chapters 3 and 5 demonstrate the rarity of adequate deliberative judgments, but here I wish to stress the profound difficulty of this problem even in theory.

Before a citizen can recognize the failure of an elected official to represent his or her interests, a citizen must first determine what those interests are.[55] When one looks beyond self-interested policy judgments to a citizen's estimation of what policies best serve the public interest, citizens must also make the effort to understand the needs and concerns of their fellow citizens—a challenge far greater than coming to know one's own interests. Once interests are developed, they must become more than tacit if they are to be of use in evaluating the performance of elected officials and the merits of political challengers. Much of what we come to know about ourselves and our social world is understood only unconsciously, and although we can generally use that knowledge in practice, it is difficult to apply it systematically unless it becomes conscious.[56]

Both exit and voice are also equally dependent on a related accomplishment—the successful evaluation of the incumbent's performance with regard to a citizen's interests or those of the general public. Now, the competent citizen must not only formulate interests but determine whether or not those have been (and will continue to be) served. Has the representative in question voted "properly" on relevant legislation? Has the representative made the voter's key issues top priorities while in office? Unless the voter can connect interests to performance, the voter will not know whether to reject an incumbent representative, let alone whether to exercise voice as a means of protest.


If the voter does decide that a public official has done a poor job as representative, yet another task lies ahead. As discussed more thoroughly in chapter 3, a voter must seek out alternative candidates as a means of electoral rejection and make similar judgments about the future performance of those candidates. This task is difficult because challengers usually have incommensurate political track records, if any at all. A voter must often make judgments based on hunches, analogies, and the misleading claims of the candidates themselves to gauge the likelihood that the opponent both shares the voter's views and will prove capable of pursuing those interests with the same vigor, skill, and resources as the incumbent.

Requiring interest formation and articulation, as well as incumbent and challenger evaluation, political exit is not the same as switching long-distance phone companies. The physical act of voting is not daunting to most citizens, but calculated acts of public expression and attempted electoral rejection are time-consuming and difficult. Only with the cognitive and emotional underpinnings described earlier can one expect citizens to avail themselves of opportunities for effective public expression and electoral action.[57]


As presented thus far, this model highlights what can go wrong in a representative system. In this final summary of the model, I demonstrate what happens when such a system operates properly. This description reveals a final, implicit element of the model. The point of government—even democratic government—is good public decision making. Just as citizens can only make sound judgments after deliberating, so public officials can only govern wisely after careful research and reflection.[58] Deliberative citizens might reach some clear judgments on particular issues and treat their representatives as sworn delegates, expected to do exactly as they have been instructed, but more often representatives must conduct their own deliberations and act as trustees of the public's best interest.[59]

From one perspective, the model of representative democracy that I have described is designed to promote just such deliberation by public officials. As shown in figure 1, citizens' choices about how to use voice and votes indirectly depend upon official deliberation. If officials reach a sound deliberative judgment about the public interest and then act


Figure 1. Model of Democratic Deliberation and Representation

[Full Size]
upon that judgment, they should fare well on election day so long as citizens reach the same judgment in their own deliberations.[60] After all, in the ideal model of representative democracy, when citizens evaluate policies and candidates, they take into account the actions of elected officials. This feedback loop ensures accountability for officials' actions, and that, in turn, creates an incentive for officials to deliberate carefully lest their judgments and subsequent actions fail to impress the deliberative electorate.

To summarize, voting and public voice have important roles in the democratic process, no matter how one deflnes democracy. In addition, a healthy representative political system requires ongoing policy deliberation and intensive candidate evaluation during campaigns. Both of these processes require active citizen deliberation about private and public interests. Without an underlying set of political beliefs (e.g., selfefficacy, public trust, and civic responsibility), deliberation and the twin mechanisms of voice and vote may fall into disuse, resulting in widespread civic neglect.

The model of democracy developed in this chapter demonstrates the necessary features of a healthy representative system. To promote the creation of deliberative public policy decisions by elected officials, a political system must have institutions that sustain ongoing citizen deliberation, nurture a resilient and motivated public, and provide clear opportunities for the influential expression of political voice and the real threat of electoral rejection. The next four chapters will assess the success of U.S. political institutions in maintaining a vigilant citizenry and stimulating productive political responses to lapses in public representation.

previous chapter
Exit And Public Voice In Representative Democracy
next chapter