Democracy for Realists: Why Elections Do Not Produce Responsive Government (2016), written by political scientists Christopher H. Achen and Larry M. Bartels, challenges the generally held assumptions about how democracy works in practice. Using years of research in political science, economics, and psychology, Achen (Princeton University) and Bartels (Vanderbilt University) suggest that voters are not the rational, policy-based decision makers commonly portrayed in democratic theory. Rather, they argue that outcomes of elections are determined by much broader forces—social identity, partisanship, and short-term perception—rather than the actual policy preferences of voters. Based on the behavioral school of political science, the book explores The Illusion of Electoral Mandates, Group Identity as the Foundation of Political Behavior, and Misattribution, Randomness, and the Limits of Democratic Accountability.

This guide uses the 2016 Princeton University Press e-book edition.

Summary

Achen and Bartels claim that conventional theories of democracy fail to accurately portray the functioning of political systems. At the heart of their argument is an examination of what they refer to as “the folk theory” of democracy—i.e., the general belief that citizens create policy preferences, select candidates that represent those preferences, and thus exert meaningful influence over governments. Thus, according to this point of view, elections act as a method of rendering public opinion into policy results. The authors assert that this portrayal is both empirically incorrect and based on faulty assumptions about human behavior.

First, Achen and Bartels explore the logical and empirical bases of this folk theory. Early chapters examine models of competitive elections which assume that voters have consistently formed opinions on policy. The authors demonstrate that these models collapse when political issues become multi-dimensional. They argue that because there is no way to determine a majority preference under these circumstances, collective preferences may be either unstable or incoherent. Therefore, the notion that elections routinely generate valid mandates is undermined.

Next, the authors examine empirical evidence related to voter behavior. Research indicates that most citizens do not have a high level of awareness regarding politics and do not express consistent ideologies. Rather, many voters use party identification, social cues, and limited amounts of information to form opinions. These results contradict the idea that voters thoroughly evaluate policy alternatives prior to casting votes. Furthermore, Achen and Bartels argue that even better-informed voters are subject to various types of cognitive distortions and social influences that inhibit rational deliberation.

Some researchers have responded to these problems through the development of a type of “retrospective” voting. Retrospective voting assumes that voters do not necessarily possess detailed knowledge of specific policy details; instead, voters reward successful incumbent officeholders and punish unsuccessful ones based on how well they performed in the past. Although Achen and Bartels agree that retrospective voting is indeed empirically evident in relation to economic conditions, they believe that it is considerably less reliable than typically suggested.

Using several case studies and statistical analyses, the authors argue that voters regularly misplace responsibility for events. Incumbent office holders may receive punishment for natural disasters or credit for economic improvements that were beyond their control. Moreover, voters habitually give excessive attention to current events. Since this short-term perspective leads to consideration of temporary fluctuations rather than long-term performance, retrospective voting creates only weak and sporadic incentives for effective governance.

The authors also challenge the idea that large-scale changes in government occur due to widespread agreement around ideological issues. In analyzing the impact of President Franklin D. Roosevelt’s (FDR) New Deal programs on American society during the 1930s, Achen and Bartels argue that Roosevelt’s successful bids for re-election resulted primarily from short-term economic improvement rather than from broad acceptance of his policy proposals. Over time, repeated short-term evaluations contributed to persistent partisan affiliations, although these developments did not necessitate consistent or knowledgeable policy decisions from voters.

Beyond critical analysis, the authors propose a new conceptual framework focused on social identity and group behaviors. They claim that individuals’ membership in socially defined groups based on race, religion, social status, and other categories greatly affects both their political orientations and perceptions of facts/events. Through the presentation of literature from sociologists and psychologists demonstrating that individuals tend to endorse views compatible with their group affiliations, and to perceive political information in a manner that supports those affiliations, the authors assert that people systematically misunderstand facts/conditions based upon their social group memberships.

The authors emphasize that partisanship plays an especially significant role in this conceptual framework. Achen and Bartels state that while party identification reflects elements of policy orientation, party identification serves as a primary social identity. When individuals affiliate themselves with a particular political party, they frequently revise their values/beliefs to align with those of their chosen party. This process may result in a number of systematic errors, as voters reinterpret facts to validate their pre-existing commitment to their party affiliation. Even politically involved/informed individuals are likely to employ partisan reasoning to maintain existing viewpoints.

Finally, in the last chapters, Achen and Bartels develop what they regard as a more realistic theory of democracy. They contend that political outcomes are best explained by competition among organized groups rather than as expressions of individual preferences. Political parties represent coalitions of these groups, and elections provide mechanisms for electing leaders, albeit without producing accurate responses to policy demands. While elections continue to perform important roles (e.g., transfer-of-power processes; rudimentary forms of accountability), they cannot possibly meet idealized conceptions of direct popular control.

In their Afterword, Achen and Bartels apply their framework to the 2016 United States Presidential Election. They argue that the election illustrates the role of social identity and party affiliation in the same manner as previous elections. Both Donald Trump and Hillary Clinton received considerable levels of support from clearly identifiable demographic groups. Despite Trump’s unusual candidacy, the election was very similar in terms of the effects of economic conditions and incumbent presence. 

Ultimately, Achen and Bartels conclude that perhaps the biggest obstacle confronting democracy today is not an institutional failure but a disconnect between democratic ideals and practical realities. The authors advocate for a more realistic understanding of how democracy operates, acknowledging both the inadequacy of citizen competence in forming policy-oriented opinions and the pivotal role played by social groups in defining individual political actions.