.J. Eric Oliver is professor of political science at the University of Chicago. Shang E. Ha is assistant professor of political science at Brooklyn College. Zachary Callen is assistant professor of political science at Allegheny College

LOCAL ELECTIONS AND THE POLITICS OF SMALL-SCALE DEMOCRACY

By J. Eric Oliver Shang E. Ha Zachary Callen

PRINCETON UNIVERSITY PRESS

Copyright © 2012 Princeton University Press
All right reserved.
ISBN: 978-0-691-14356-9

Contents

Acknowledgments…………………………………………………………………………ixIntroduction……………………………………………………………………………1Chapter 1 Size, Scope, and Bias: What Differentiates Local Electoral Politics?…………………12Chapter 2 Who Votes in Local Elections?……………………………………………………53Chapter 3 Who Runs for Local Office?………………………………………………………87Chapter 4 Systematic versus Idiosyncratic Factors in Local Elections………………………….116Chapter 5 What Influences Local Voters’ Electoral Choices?…………………………………..149Chapter 6 Rethinking Local Democracy………………………………………………………183References……………………………………………………………………………..209Index………………………………………………………………………………….215

Chapter One

Size, Scope, and Bias: What Differentiates Local Electoral Politics?

On April 7, 2009, voters in the Illinois towns of Lyons, Carpentersville, Palatine, and Bensenville did something unusual—they did not reelect their governing mayors. Reporting on these results, the Chicago Tribune speculated on why these candidates lost, a surprising event considering that most of their fellow incumbents in nearby towns won. For each place, the Tribune had a seemingly unique explanation: in Lyons, the serving mayor inadvertently made a series of sexist and racist remarks on tape; in Carpentersville, the village president had a long history of fights with the city council; in Palatine, the incumbent faced a challenge from a former star player for the Chicago Bears; and in Bensenville, a concerted effort from Chicago’s neighboring Daley political organization unseated the incumbent because of his opposition to an airport expansion project. But while these explanations sound plausible, there is no way of differentiating their factual basis from mere speculation. We have no idea if mayors are more likely to be unseated when they face former football stars or if making racist remarks is fatal to holding office, because we have no systematic explanations about how people vote in small-scale elections.

Why do we know so little about local elections? Part of the problem lies with our understanding of elections in general. Most political observers view elections the way Gertrude Stein saw roses—as categorically similar phenomena. In other words, voters use the same mental calculus regardless of whether they are voting for mayor, city clerk, dogcatcher, or president. But if this were the case, then not only would local elections hinge on the partisanship, economic performance, and candidate charisma that are so decisive in presidential elections, they also would be decided by the same factors and in the same way time and time again.

Yet, as the foregoing examples illustrate, this is clearly not the case. Elections are not all the same, and simple common sense would tell us that voting for president is much different than voting for a mayor, city clerk, or dogcatcher. In fact, unlike their national counterparts, local elections are usually nonpartisan, get little media attention, and seem to provide voters with little public information to differentiate candidates. Similarly, most local elections are often free of direct racial or ethnic appeals or the preoccupation with crime or unemployment that dominate the political agendas of most large cities. This is not to suggest that partisanship, issues, race, or any of the factors common in larger-scale elections will never be important in local contests. Sometimes they will and sometimes they won’t. But currently, we have no way of knowing when these factors matter and when they do not.

This, however, leads to an even bigger challenge to understanding local elections: how do we also differentiate among the tens of thousands of local governments in the United States? For not only are elections for mayor in places like Clarkston, Georgia, and Steubenville, Ohio, different than elections for president, they are different from each other as well. And just as we don’t know how, exactly, local elections are different from presidential ones, we also lack a mechanism for understanding how or why a local election in one place is likely to be different from that in another. And without such a framework, we ultimately have no way of distinguishing between the systematic forces that shape local politics from idiosyncratic events and parochial circumstance. In short, if we want to understand the dynamics of local democracy, we need to identify what separates the electoral politics of a municipality from that of a nation, state, county, special district, or any other form of democracy.

This chapter outlines a relatively simple way of doing this. Across the universe of democracies, three characteristics are the most powerful and widely applicable predictors of their electoral politics: size, scope, and bias. Once we know a democracy’s population (size), the magnitude of its constitutive powers (scope), and how uniformly it distributes its resources (bias), we can predict a great deal about who votes, who runs for office, and whether factors like incumbency, parties, ideology, issues, interest groups, and candidate charisma shape vote choices. In other words, we can best predict how people will vote in a particular election if we first understand what is distinctive about that democracy’s politics and, if we know its size, scope, and bias, we can predict what those electoral politics are like. To better understand this, let us separately examine the political dynamics of size, scope, and bias in more detail.

Size: the Power of Numbers

In 2010 voters in California and Vermont both elected governors, and while the formal powers of the two state offices are roughly similar, the contrast between the two elections could not have been greater. In California, Jerry Brown and Meg Whitman collectively spent over $200 million dollars on their campaigns, largely on television ads targeting key voters. Yet comparatively few Californians got to actually meet either candidate, let alone ask them any questions or express their views. For most Californians, Brown and Whitman were simply television personalities. In Vermont, Peter Shulman and Brian Dubie collectively spent under $2 million and ran far fewer television ads. Instead, they campaigned in the “old-fashioned” manner of going to town meetings, posting yard signs and billboards, meeting with prominent groups, holding public events, and even standing on street corners waving at cars passing by. Vermonters had many opportunities to personally interact with both candidates and make their assessments from first-hand experience.

The contrasting examples of California and Vermont highlight a long-standing concern among political philosophers: how big should a democracy be? Historically, philosophers have favored smaller population sizes. For example, Plato calculated that a republic should have no more than 5,000 citizens; Montesquieu thought democracies should not exceed 20,000 citizens. These calculations stemmed from normative concerns about the mechanisms of popular rule. Plato and Montesquieu believed that in order for citizens to have an effective and meaningful voice in public affairs, they needed to have direct access to their elected representatives and the governing processes. Democracies consisting of millions or tens of millions of people were unfathomable to premodern philosophers because the aggregation and coordination of so many different citizens’ preferences seemed impossible, particularly in an era before mass literacy and popular media. And even while contemporary technology allows for easier communication between citizens and their representatives, the concerns of the classic philosophers about democratic accountability and citizen involvement continue to hold today. In an era of media-packaged, television-centered campaigns, many commentators justifiably fear that politicians lose accountability to the concerns of the mass public (Dahl and Tufte 1973).

In reality, an increased population size is both a blessing and a curse for any democracy. On the positive side, larger size allows for democracies to accomplish grander projects and achieve greater economies of scale in their governing operations (Alessina and Spoalare 2003). Few villages, for example, can construct the zoos, parks, civic centers, airports, and other amenities that are common in most large cities, and most towns actually provide fewer services directly than do larger cities. Data from the 2002 Census of Governments indicate that places with populations of less than 5,000 provide fewer than five of twenty basic services on average compared to nearly nine services offered in places over 100,000 in population. These size effects work at nearly every level of government: Smaller towns are less likely to provide public housing, build hospitals, or have energy or waste treatment facilities than larger ones; Vermont could never sustain California’s world-class public university system and finds it proportionately more costly to run many public services; small countries do not have space programs, build super-colliders, or sustain large militaries. Simply by virtue of its size, a larger democracy is going to be more powerful, and usually more efficient, than a smaller one.

On the negative side, governmental accountability and access is lower in a large democracy than a small one. Partly, this is a simple problem of sheer logistics. A democracy’s ability to consider the viewpoints, opinions, and perspectives of all its citizens diminishes exponentially with every new member it adds. These coordination challenges are especially difficult with governing decisions where citizens are likely to have strong opinions and varying degrees of information. No large city, for example, could reasonably govern itself according to the practices of a New England town meeting—there simply wouldn’t be enough time for everyone to speak nor even a venue large enough to accommodate all citizens. As democracies become larger they must delegate more deliberative processes to experts, who, by necessity, will become increasingly distant from all citizens. In larger democracies, preferences must be expressed more through ratification than deliberation.

More importantly, larger democracies face bigger challenges to governing from their greater social heterogeneity. Large nations like the United States, India, Indonesia, and Brazil tend to have a higher amount of ethnic, regional, and class diversity, while smaller democracies like Denmark, Costa Rica, and Greece tend to be more homogeneous. Data from the World Values survey indicate that countries under a million in size have, on average, a Social Diversity Index Score (Okediji 2005) of .68 on a 0 to 1 scale; this compares to an average score of .83 for countries over 25 million in size. This trend also holds for American local governments, as illustrated in figure 1.1. Looking at just one possible measure of diversity, race, we see that American municipalities consistently grow more racially heterogeneous as they get larger. Places under 5,000 in size have an average Index of Qualitative Variation (IQV) score, a measure of racial homogeneity, of only .19; places over 100,000 have an average score of .58. The same trend occurs with nearly every other social category, including income, age, religion, and lifestyle choice—larger American cities tend to be more socially diverse than small ones.

For all these reasons, even as democracies grow larger and more capable, the connections between citizens and politicians grow more remote, mediated, and formally structured. Such effects also have electoral repercussions. Smaller democracies may afford opportunities for deliberation, interpersonal persuasion, and immediate accountability, but are also usually limited in the range of decisions they need to adjudicate. A New England town meeting may provide an opportunity for all residents to express their opinions and to deliberate town decisions, but these decisions are usually going to be limited to a rather narrow range of issues.

In larger democracies, the greater magnitude of political decisions, the increasing distance among citizens, and the greater diversity in their preferences mean that political processes will become more rationalized, organized, and systematic. With their larger projects and more heterogeneous populations, larger democracies will typically sprout more formally organized political groups. For instance, when a town grows big enough to support an airport, a new set of political constituencies is created among the landowners who wish to sell their properties to the airport authority, the local businesses who seek to profit from increased commerce, and the residents who may be adversely affected by noise and safety considerations. These political constituencies may manifest themselves as a political party or formal interest group such as an industry or environmental organization. Such organizations are less likely to emerge in smaller democracies not only because there are fewer issues of magnitude that could divide a more homogeneous membership base, but because aggregating member preferences is so much easier. A political party or formal interest group is simply unnecessary when a single candidate can directly contact all the voters and ask for their support. But when a political leader wants to rally thousands or millions of supporters, direct contact is simply unfeasible, and so an organization that mobilizes and coordinates political activists becomes a necessity.

The gubernatorial campaigns in Vermont and California provide typical examples of these tendencies. Candidates in Vermont spend most of their campaign funds on staffing, billboards, yard signs, and bumper stickers, and very little on television advertising. This is partly because Vermont is a small media market (and thus cheaper to advertise in) but also because the small size of the state allows for politics to have a more retail flavor. Voters in Vermont expect to meet their politicians in person, and a large percentage of Vermonters actually will have met the candidates. In California, the idea of a candidate trying to personally meet even a small fraction of her state’s 37 million residents is ludicrous. Political campaigns in California thus mean lots of media consultants, television advertisements, mass mailings, and well-calculated voter mobilization drives. Because so few of California’s voters have any direct contact with their candidates, party labels also become much more important as guides for orienting their voting behavior.

A democracy’s size not only influences the logic of political organization, but also who votes, who runs for office, and the levels of voter information about local campaigns. Past research has shown that residents of larger cities tend to vote less than residents of smaller places partly because they are less interested in local affairs or less familiar with local leaders (Oliver 2001). Similarly, we would also expect that candidates in larger places will be more ambitious in their political aspirations and professional in their political skills, simply because the investment of time, effort, and resources in running for office in a larger place is so much greater. And, as a democracy becomes larger and candidates rely more on mass media or reaching voters through particular groups, voter knowledge about candidates and campaigns may change. In smaller places, voters are more likely to be personally familiar with their office seekers; in larger places, they are less likely to know candidates, particularly if the contestants are not incumbents.

The likely impact of population size on municipal electoral politics is summarized in table 1.1. For simplicity’s sake, municipalities are arbitrarily categorized by four levels of population size. The estimates in the table also derive from the assumption that a certain portion of a town (roughly 25 percent) is ineligible to vote because its inhabitants are too young, are not citizens, or do not meet other requirements. If we also assume a 25 percent turnout rate, this means the number of votes that a winning candidate actually needs in order to attain local office can be quite small. For example, in a town of 4,000, only 3,000 residents may actually be eligible to vote. If only 25 percent of these turn out for a local election, a candidate needs only to secure just over half of the 750 votes (i.e., 376 votes) to win. With such a small number of votes required, the typical candidate can probably win election largely through personal contacts. Although most candidates would also seek to bolster their candidacy through yard signs, stickers, fliers, buttons, and the like, much of their support will come through more personal connections.

(Continues…)

Excerpted from LOCAL ELECTIONS AND THE POLITICS OF SMALL-SCALE DEMOCRACYby J. Eric Oliver Shang E. Ha Zachary Callen Copyright © 2012 by Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.