_
$3.3»
.11....rf
4.94.
.9955”...
yuan. .4 d1...
4

¥
n 9.5)
.2131.

a.
k. I .5:
33s “5-1 5% x-
; [.4 z. 1 i
In“ I. r. .14.. km.
1,4 01 . 4 .
, tuna-«.53. f? a
, V . :9. 5:43.33. :51. sigh—tum
. .thvitfixd. :A n
. , . .19.}: 5...:
‘ .fsu . ‘ , . .
, 5,. i V .
5.... “Sudan. . ‘

v
‘ .‘
. .triinuuzs ‘

‘.
3.. t
.43.... f
.1: .1 I.
it. .2

Us:
(I 23.32. .3." n...
. .31! .l'... 3:30!

‘3 1

 

hv 47...,
l. 1 1..
In}... .1.

.23. x,
L1 .92.}? F ..
.

3.... ... .
{xvii

,. to k
x: :1
V I3?!
it!!!
u . I I.
.0531! s. A .
.. 11‘ an 1‘).
.:I.. 99“ , )'l

f, 1..
$3.

\‘I d...‘
21 . 3x: g , i
.!._S:A§1 Jr: ‘3‘-
: . 1.33 L... .
I!!.:‘b-hf.e , .
. ‘ 1. ,
‘ 3?... 3...
. ‘.\ .
.Bl I’..u||.;~ h...
. A . 1.. 55;: o

 

 

[ mess
I

 

700D

This is to certify that the
dissertation entitled

Decentralization, Economic Voting, and Accountability in a
New Democracy: The Case of Mexico

presented by

Paloma Bauer de la lsla

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Political Science

 

 

WW

 

Major Professor’s Signature

5/4/2006

 

Date

MSU is an Affirmative Action/Equal Opportunity Institution

.r

LIBRARY
Michigan State
University

 

 

 

- .-.--—.--a-.—---

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/C|RC/DateDue.indd-p.1

 

DECENTRALIZATION, ECONOMIC VOTING, AND ACCOUNTABILITY IN
A NEW DEMOCRACY: THE CASE OF MEXICO

By

Paloma Bauer de la Isla

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Political Science

2006

 

ABSTRACT

DECENTRALIZATION, ECONOMIC VOTING, AND ACCOUNTABILITY
IN A NEW DEMOCRACY: THE CASE OF MEXICO

By

Paloma Bauer de la Isla

The research question that motivates this dissertation is whether new democracies
develop functional mechanisms of accountability. Taking economic voting as an
indicator of electoral accountability and exploring the effects of decentralization on
incumbent support, I examine under which institutional conditions electorates in new
democracies might become economic voters. Employing both aggregate and individual
level data, the dissertation explores hypotheses about economic voting at the subnational
level in Mexico. The case of Mexico is particularly well-suited to study voters’ rationales
in a post electoral authoritarian context, where state-level institutions are struggling to
gain political and financial autonomy. The central argument is that the propensity of the
electorate to become sophisticated and effectively hold elected representatives
accountable is contingent upon state-level political and economic decentralization. I find
that state-level factors exercise a powerful effect on electoral outcomes. Specifically, I
demonstrate that state governments become more accountable to citizens with the
increase of decentralization - measured by state-level financial autonomy and electoral

competition.

enj
in)”
him
we

at:

Ln
for
int
3P
Sil
lllt
f0]

lri

\l’

 

\\

Cl,

 

 

ACKNOWLEDGMENTS

It gives me pleasure to thank those who helped me in countless ways both to
enjoy and to complete the writing of this dissertation. First and foremost, I am grateful to
my advisor Mark P. Jones. I was very lucky to have a chance to learn from and work with
him. I deeply appreciate his unconditional support, commitment, and confidence. These
were essential to the successful completion of this degree. I am grateful for his guidance
at every step, his common sense, and his friendship.

As a graduate student in the Department of Political Science at Michigan State
University, I had a chance to study with excellent teachers. I thank Dr. Michael Bratton
for a three-year research assistantship. A challenging experience, the assistantship
introduced me to African politics and allowed me to explore several political issues. I
appreciate Dr. Bratton’s constant willingness to help. I also appreciate Dr. Brian D.
Silver’s support throughout. His classes, at once challenging and fun, provided me with
the requisite theoretical basis essential for the pursuit of my own questions. I am grateful
for his advice and helpful suggestions. I have also been extremely fortunate to enjoy the
friendship and support of former professors at other institutions. I thank Todd A.
Eisenstadt of American University. I also thank Chris Reardon, Cliff Wirth, and Mark
Wrighton of the University of New Hampshire. I value my friendship with Professor
William Baker of the University of Maine and strive to emulate his example of
excellence in teaching.

This dissertation could not have been completed without the support and
generosity of Alejandro Moreno and others at the newspaper Grupo Reforma. I am

gratefirl for access to Reforma ’s pre-electoral surveys as these provide the basis for part

iii

of this study. I thank Eric Magar at ITAM (Autonomous Technological Institute of
Mexico) for giving me access to ITAM’s library resources while I was doing field work
in Mexico. Thanks are due to Carlo Angel Varela at Banamex (Banco Nacional de
México) for important compilations of electoral data. I appreciate Jorge Hemandez’s
extensive and careful help toward producing the macro-level data set. I am grateful for
generous support of my doctoral studies provided by from the Center for Latin American
and Caribbean Studies at Michigan State University.

Finally, I thank my friends and family for their company, laughter, and
encouragement. I am deeply indebted to Carina, my sister and best friend, whose
unconditional support helped me in so many ways to attain this goal. My debt to my

parents is without measure. I know my father shares with me this happy moment.

Cl

 

List of Tables

List of Figures

TABLE OF CONTENTS

Chapter 1. Introduction

1.1
1.2
1.3
1.4
1.5

Overview of the Argument
Contributions

Research Design and Data
Case Selection
Composition of Chapters

Chapter 2. Theory

2.1
2.2
2.3
2.4
2.5
2.6

Introduction

Dissertation’s Place within Democratization Literature
Economic Voting

Economic Voting in New Democracies

Accountability and Decentralization in New Democracies
Conclusion

Chapter 3. The Mexican Case

3.1
3.2

3.3

3.4

Introduction

The Hegemonic Party System
Political and Economic Centralization
Reform and Transition

Rise of the Opposition

Altemation and Divided Government
The Mexican Voter

Chapter 4. Aggregate-level Analysis

4.1
4.2
4.3
4.4
4.5
4.6

Introduction

Theoretical Considerations
Hypotheses

Data and Model

Analysis and Results
Conclusion

vii

viii

"OONQ-h-H

13
15
18
25
29
35

37
37
38
41
44
45
49
51

69
69
71
75
77
80
84

Chapter 5. Individual-level Analysis
5. 1 Introduction
5.2 Theoretical Considerations
5.3 Hypotheses
5.4 Data and Model
5.5 Analysis and Results
5.6 Conclusion

Chapter 6. Conclusion

Appendices
Appendix A. States and Election Years, Chapter 4
Appendix B. Question Wording and Coding, Chapter 5
Appendix C. List of Mexican States

Bibliography

vi

9O
9O
91
96
98
102
105

117
122
123
125
146

147

I‘
in) ~

 

 

1.]

3.1

3.2

3.3

3.4

4.1

4.2

4.3

5.1

5.2

5.3

5.4

5.5

5.6

5.7

LIST OF TABLES
Research Design
Ranney Index in Mexican States, 1972-1989
Ranney Index in Mexican States, 1990-2004
Holbrook-Van Dunk Index and Folded Ranney Index in Mexican States
Experience under Non-PR1 Rule, Mexican States, 1985-2004
Diverse Indicators of Financial Autonomy in Mexican States, 2004
Gubernatorial Elections across Mexican States, 1985-2004
Local Legislature Elections across Mexican States, 1985-2004
Main Characteristics of the States Included in the Analysis
Gubernatorial Approval, Selected Mexican States
Vote Choice in Gubernatorial Elections, Selected Mexican States
Vote Choice in Local Legislature Elections, Selected Mexican States
Gubernatorial Approval in Selected Mexican States, Merged Data Set

Vote Choice in Gubernatorial Elections in Selected Mexican States,
Merged Data Set

Vote Choice in Local Legislature’s Elections in Selected Mexican
States, Merged Data Set

vii

12

61

62

63

64

86

88

89

108

111

112

113

114

115

116

 

b.)

L’)

(J)

U!

 

 

 

1.1

3.1

3.2

3.3

3.4

3.5

3.6

3.7

3.8

3.9

4.1

5.1

5.2

LIST OF FIGURES
Overview of the Argument
State’s Revenues, 1970-2003
Number of Seats in the Chamber of Deputies, 1961-2003
Party Affiliation of Governors in the 32 States, 1980-2004
Experience under Non-PR1 Rule, Mexican States, 1985-2004
Holbrook-Van Dunk Index, Mexican States 1970-2004
Number of Seats in Local Congresses, 1974-2004
Distribution of Seats in Local Congresses, 1974-2004
Local Congresses where one Party has a Majority of seats, 1978-2003
Electoral Victories in MR Districts by Parties Other than PRI, 1974-2000
Total Votes for Incumbent Party, Gubernatorial Elections 1985-2004
State’s Own Revenues, Average Percentage 2000-2002

Mexican States, Own Revenues, Average 1989-2003

viii

11

59

60

60

65

66

67

67

68

87

109

110

 

Ill

dc

 

 

 

Chapter 1. Introduction

The broad research question that motivates this dissertation is one of
accountability in emerging democracies. Taking economic voting as an indicator of
electoral accountability and exploring the effects of decentralization on incumbent
support, I examine under which institutional conditions electorates in new democracies
might become economic voters. Specifically, by exploring the effects of political and
economic decentralization on incumbent support across the Mexican states, this study
seeks to assess whether and to what extent electoral accountability is taking place in a
new democracy.

Does a new democracy guarantee electoral accountability? That is, does a new
democracy develop those mechanisms of accountability necessary for voters to become,
in the immortal words of V0. Key (1964:568), a “rational god of vengeance and
reward”? Furthermore, do voters assign responsibility to the right level of government?
Are voters able to correctly distinguish between national and subnational performance
when voting at the state level?

Without doubt, accountability lies at the bottom of these questions. Being a
product of a number of institutional mechanisms, its importance for the success of any
representative democratic regime cannot be overstated. Accountability is, however, not
an immediate or automatic product of democracy, as democratic regimes can still exist
without the adequate working of checks and balances that guarantee accountability
(Bratton 1998). This is, I believe, an important shortcoming in some democratic regimes,

and a potential danger for their consolidation, quality, and even survival. Nations where

the essential democratic institutions are in place, but where accountability to its citizens is
not guaranteed, lack clearly a major component of democratic life and risk developing
serious problems of govemability. As argued by Rose, Mishler, and Haerpher (1998235),
“if a regime is to be fully democratic, the government of the day must be accountable to
the electorate.” By extension, a regime lacking accountability is in danger of losing its
citizens’ support and satisfaction vis-a-vis the system. Eventually, a dissatisfied and
angry civil society can end up alienating itself from the system through apathy,
indifference, or even through violent means. Thus, the dangers to the democratic
legitimacy, stability, and durability of the system are countless (Easton 1965).
Accountability is, however, a complex issue to explore. With this research, I hope
to shed light on this concept by looking at one specific component of democracy and one
essential mean of accountability: elections. Elections represent the most direct link
between citizens and the government, as well as the principal form of accountability. In
fact, for many citizens in the world, elections are the only mechanism to connect with
their government. The act of voting itself is critical in two major ways to ensure
representation, accountability, and responsiveness to the citizens by political leaders.
First, elections are the natural mechanism for citizens to hold their government
accountable for its past performance (Schedler 1999). By voting based on a rational
choice of whether to reward or punish the sitting government, voters are able to
experience first-hand an act of governmental accountability. Second, the way voters elect
their leaders greatly determines what kind of representation and the level of

responsiveness they will experience from their government.l When a democracy fails to

 

' See Przeworski, Stokes, and Manln (1999).

ensure its citizens that their political participation is meaningful,2 the durability and even
viability of the system is put at risk. In other words, elections are the most
straightforward mechanism of accountability in a democracy, and accountability is
needed for a democratic regime to succeed.

What is needed for elections to be an act of accountability? Following Dahl’s
(1971) emphasis on contestation and participation, scholars agree on a number of factors,
such as clean, periodic, and competitive elections; a competitive party system; and an
interested, informed, and involved electorate. But what does it take to make these factors
converge and make accountability work? Furthermore, assuming that the minimal
institutional standards are in place, are citizens in a new democracy “ready” for
democracy? Or, are citizens in such contexts less capable of being rational voters than
citizens in old democracies?

In this project, I argue that the decentralization of political and economic powers
at the subnational level can lead to make institutional mechanisms of accountability
work, such as elections. When a subnational government does not depend fully on the
central government for resources and political power, then elections acquire a form of
“local power” — that is, subnational politicians need their constituents’ electoral support
and therefore elections are meaningful for a majority of voters. Thus, elections acquire
different meanings, depending on the degree of decentralization attained in a given
region, which will vary across nations, and more importantly in this work, even within a

nation. In other words, those ideal results of decentralization that are argued to create

 

2 Specifically, by “meaningful” I mean here that voters know that with their vote (as a majority) they can
cast a government out of office.

 

 

h.

dc

cl

 

eventually a sophisticated electorate, such as transparency, availability of information,
and political knowledge, depend on the institutional context of a given state.

The underlying logic of this argument is that electoral accountability can only
exist, if those institutional arrangements designed to respond to citizens are functioning
appropriately. This study addresses the specific question of whether the expected
outcomes of political and economic decentralization reach the electoral arena. Ideally,
decentralization is supposed to promote democracy at the grassroots level, developing
hereby different mechanisms of accountability. Whether voters are able to use their

voting right as an act to hold the government accountable remains an open question.

1.1 Overview of the Argument

This study looks at an assumed and indirect outcome of decentralization, electoral
accountability, by means of exploring electoral choice at the subnational level in a new
democracy. Ideally, citizens living in a system with working accountability, make use of
elections as one way of evaluating their representatives.

Figure 1.1 shows a model of the argument that I develop in this dissertation. In
the first place, effective decentralization can only take place when politicians need to take
into account their constituencies for political support, and even survival. This happens
when politicians depend no longer on the central government (or hegemonic party), but
on local constituencies for their electoral support and political success. Effective
decentralization of political and economic powers is then assumed to bring the
government closer to the peeple, by delegating powers (political, fiscal, and/or

administrative) to lower levels of government. This presumably creates, on the one hand,

a cycle of greater clarity of responsibility in which there is greater flow of information.
On the other hand, it raises the levels of pluralism, electoral competition, and a state’s
financial autonomy vis-a-vis the central government. Altemation of power and divided
governments are common at the subnational level. Overall, subnational units are more
autonomous both politically and economically from the central government. This new
institutional environment is expected to create changes in the governing style of
subnational politicians, who depend increasingly on citizens’ evaluations to retain power.
The increased transparency of information and responsibility increases citizens’ political
knowledge and heightens their attention to public affairs. The result is a sophisticated
electorate, who in turn is more likely to make an electoral choice based on performance
evaluations. This, I argue, is a way of observing electoral accountability.

This model, as applied to the case of Mexico, explores the impacts of political and
economic decentralization on incumbent support across the Mexican states. Specifically,
the focus of this study is on the relationship between two ideal outcomes of a
decentralization process—namely, electoral competition and financial autonomy—and
incumbent support at the subnational level. The overall purpose is twofold: In the first
place, I aim to explore whether, and to what extent, decentralization impacts electoral
behavior. By using macro-level data, I find evidence that both political and economic
decentralization impact electoral behavior, specifically, economic voting, at different
degrees. This leads to the second overall goal, which is to fiu'ther explore with micro-
level data the existing variation across the states. The principal hypothesis leading this

work is that as the level of decentralization intensifies, voters become more sophisticated

and thus, hold the government accountable by voting based on a performance evaluation.

Overall, I find support for this hypothesis.

1.2 Contributions

This dissertation contributes to our general knowledge of the quality of new
democracies in the following three ways.
a) By exploring the subnational level of politics in a new democracy, where the transition
process towards democracy started at this level of government, rather than at the national,
like most other studies, this dissertation contributes to the understanding of different
stages and paces of the transition and deepening processes of democracy within one
nation. Mexico presents an excellent case for studying the evolution of democracy at the
state and local levels, which in some states started approximately 20 years before it fully
reached the national arena, while others have not experienced alternation of power until
today.
b) This study contributes to the broad field of economic voting, which remains placed
high on the political science research agenda. The political importance and effects of
economic factors sustain a lively scholarly debate that has plenty of open questions,
particularly in reference to fragile economic and democratic contexts, where the study of
economic voting is a particularly relevant exercise. Specifically, by exploring economic
voting at the subnational level in a formerly hegemonic party system, this study
contributes to evaluate the differing degrees of institutional capacity of subnational

governments.

c) The third contribution refers to decentralization, a public policy that has been
extremely popular in emerging democracies across the world. However, so far, most of
the literature has not spent much time on exploring the link between the expected
accountability mechanisms resulting from decentralization and electoral outcomes. This
dissertation contributes to the research program on decentralization by exploring the
potential implications of decentralization on democratization policies from the

electorate’s standpoint.

1.3 Research Design and Data

This dissertation employs a subnational research design to study electoral
behavior across the Mexican states. The research design, which I briefly will explain here
is summarized in Table 1.1. I make use of both subnational- and individual-level
analyses. First, look at the question of interest within a subnational level of analysis. For
this, I gathered an aggregate data set composed of a collection of macroeconomic, fiscal,
and electoral indicators spanning 34 years. This cross-sectional analysis allowed me to
examine different hypotheses across the 31 Mexican states and the Federal District
(Mexico City) over a significant period of time. The macroeconomic data is taken
primarily from Mexico’s Institute of Statistics (Instituto Nacional de Estadz’sticas,
Geografia e Informatica, INEGI). The electoral data and other socio-demographic
indicators used in the dissertation were obtained from the Federal Electoral Institute
(Institute Federal Electoral, IF E), Center for Research for Development (Centro de
Investigacién para el Desarrollo, CIDAC), Banamex (Banco Nacional de México), and

Mexico’s National Population Council (Consejo Nacional de Poblacién, CONAPO).

 

SUI

Rt

83

01

 

I;

 

 

Second, I look at the question of interest within an individual-level analysis using
survey-level data. The data used was collected by the Mexican newspaper group
Reforma, and it is based on pre-electoral surveys collected in a sample of states. The
sample of states reflects Mexico’s rich subnational variability, which raises the study’s
generalization power. The surveys, based on face-to-face interviews, contain identical or
highly comparable items, which allow both the comparability among the individual states

and the use of a merged data set.

1.4 Case Selection

In this dissertation, I use the case of Mexico to test hypotheses about the presence
of rational vote decisions in new democracies. Mexico makes an excellent test case for
four main reasons.
a) Given that Mexico represents the prototype of a new democracy, the findings can be
applied to a large number of contexts. By being a “prototype of a new democracy,” I
mean that the political context in Mexico is similar to that of other new democracies. For
instance, Mexican voters have been traditionally unable to distinguish between
subnational and national politics. As is the case in most of Latin America and other
regions of the world, the Mexican executive has counted with disproportionately much
more power than the other branches of government, which made national politics much
more visible to the eyes of the electorate. Within such a context, it is difficult for voters
to assign responsibility to the appropriate layer of government. In Mexico, subnational
politics developed in the shadow of a strong presidential system and a highly centralized

system of relationships between the center and subnational units—a context that is not

 

 

111

Ct

 

11;

Ill

{11

dt

 

(7

\\

uncommon in new democracies. Another trait of the Mexican system, which is shared by
many other developing democracies, is that within the processes of transition and
consolidation of democracy, both political institutions and civil society are struggling to
become more autonomous and develop more power and credibility on their own.

b) The rich variability across the Mexican states in terms of democratic experiences is
very well-suited to explore the questions of interest, such as why some contexts within a
nation are more prone to produce economic voters than others, and why some
governments develop different styles of governing that favor the use of accountability
mechanisms, while others still show traits of authoritarianism. In sum, Mexico presents
an ideal laboratory because of its variations in electoral competition levels and overall
degrees of decentralization.

c) Recent and extensive use of decentralizing policies across most of the developing
world call for checks on the performance of these measures. Mexico is especially well-
suited to evaluate decentralizing efforts, because it is formally a federal system (among
the three biggest in Latin America), which did not work as such during the one-party
regime. Therefore, this study is particularly appropriate to evaluate the effects of recent
efforts to decentralize power in a federal system.

(1) Last, but not least, Mexico is an important case of study because of its obvious
political, economic, and social importance within the region. Mexico shares with the US
an array of issues, such as geographic proximity, economic relations, immigration, and
security, among many others, which make it imperative to deepen our understanding of

Mexico’s political development.

1.5 Composition of Chapters

The dissertation is structured as follows. Chapter 2 consists of a review of the
literature related to this research. The chapter states the theoretical perspective on which
the arguments of this study are built. In Chapter 3, I present the case that will be used to
test hypotheses in this project, Mexico. The subsequent chapters empirically test the
argument of this research. Chapter 4 makes use of macro-level data to explore this
research question across Mexico’s 31 states and the Federal District between 1985 and
2004, while Chapter 5 uses micro-level data in a sample of states. Finally, Chapter 6
presents the conclusions reached in this study and expands on how they might contribute

to the comparative literature on electoral behavior.

10

Figure 1.1 Overview of the Argument

 

Political and Economic
DECENTRALIZATION

 

 

 

 

Pluralism; Electoral
competition; Altemation;

 

Divided governments; States
more autonomous from
central government;
State/local governments more

Greater clarity of
responsibility, transparency,
and availability of
information

 

 

 

 

 

attentive to citizens’ needs

 

 

Citizens with more political
knowledge and interest in politics;
Overall democratic character of
population grows

 

 

 

 

Greater political
sophistication among
electorate

1

Voter makes electoral choice
based on perforrnance-
evaluation of sitting
government; Economic Voting

 

 

 

 

 

 

 

 

 

ELECTORAL
ACCOUNTABILITY

 

 

 

ll

Table 1.1 Research Design

Level of Analysis
Subnational Individual
Data
Aggregate Survey

 

 

Dependent Variable:
Incumbent Support (percentage of votes

and seats in local congresses)

Independent Variables:

State-level economic performance, local
financial autonomy, electoral experience
under opposition, and political control
variables

Method:
Cross-sectional analysis

Sm:
Own data set, most indicators from IN EGI

 

 

 

 

Dependent Variables:
Individual approval and electoral support

for the incumbent party

Independent Variables:

Individual assessments of subnational and
individual economies, and individual
control variables

Method:
Individual-level survey analysis

Source:
Pre-electoral surveys, Reforma

 

 

Chapter 2. Theory

2.1 Introduction

In a democratic regime, elections are the most powerful instrument for citizens to
express their preferences vis-a-vis the government. In this sense, elections serve as the
most straightforward method to enforce accountability. However, does an electoral
democracy3 guarantee electoral accountability? That is, does an electoral democracy
develop those institutional mechanisms of accountability necessary for voters to hold
elected officials accountable for their performance? Assuming that elections meet all
requirements necessary to be classified as clean, open, periodic, and competitive, what
makes voters take reasoned electoral decisions? What leadsto citizen circumspection in a
new democratic regime? I argue that voters are more likely to make reasoned electoral
choices when they perceive that their vote has a real chance of impacting the election
either way, and that this can be achieved through appropriate decentralization of political
and economic powers.

Economic voting has been shown to be strongly dependent on a context of
institutional clarity of responsibility,4 which can vary both between and within nations
(Anderson 2000; Nadeau, Niemi, and Yoshinaka 2002; Pacek and Radcliff 1995; Powell
and Whitten 1993). I argue that decentralization determines partly the extent to which a
context of clarity of responsibility develops. Therefore, I want to explore whether, and to

what extent, decentralization indirectly leads to economic voting. The principal

 

3 According to Diamond, Hartlyn, Linz, and Lipset (l999:x), an electoral democracy is “a civilian,
constitutional system in which the legislative and chief executive offices are filled through regular,
competitive, multiparty elections with universal suffrage.”

4 A term developed by Powell and Whitten (1993) to explain the variability of economic voting in cross-
national studies in terms of the political context (or institutional arrangements) of a given nation.

13

hypothesis leading this work is that as the level of decentralization intensifies, voters
become more sophisticated5 and are more likely to use elections as a mechanism of
accountability. Furthermore, this might contribute to explain some of the uneven
evidence of the theory found both across developing democracies and within individual
nations.

This chapter reviews the literature of economic voting, accountability, and
decentralization within the broader literature on democratization. The objective of this
chapter is, in addition to defining the major concepts as applied in this study, to lay out
the theoretical framework on which this project is based on. The chapter is organized as
follows. Next, I draw on the extensive literature on democracy to place economic voting,
decentralization, and accountability in theoretical perspective. Second, economic voting
is reviewed in both established and new democracies. Third, I define the concept of
accountability as it is used in this project, and I discuss how decentralization is linked to
elections as a means of accountability. Finally, the last section concludes with a number
of testable propositions that link decentralization with electoral outcomes and

accountability, and which will be empirically tested in chapters 4 and 5.

 

5 By sophisticated here, I mean that voters are able to punish or reward the incumbent administration when
casting their vote at the subnational level. Throughout this dissertation, I equate sophistication with
economic voting, regardless of whether it is prospective/retrospective or pocketbook/sociotropic. The
general definition of economic voting followed here refers to the theory as a “reward-punish decision
calculus on the part of the voter. The voter who feels better off financially (or sees the nation as a whole
better off) will reward the incumbent Officeholder” (Atkeson and Partin 1995:99). According to this
definition, I assume economic voting to be a “theory about applied rational, i.e. reasoning behavior”
(Dorussen and Palmer 2002: l );

2.2 Dissertation’s Place within Democratization Literature

Since the “third wave” of democracy was unleashed in 1974, with the overthrow
of the Salazar/Caetano dictatorship in Portugal, the number of democratic systems across
the world increased in a dramatic way (Huntington 1991). Regardless of how one defines
democracy,6 it is clear to most observers that the worldwide trend has been towards more
democracy and freedom.7 Extensive scholarly work has both accompanied and followed
up the development of democratic systems across the world. At first, this field of study
was primarily concerned with questions on the conceptualization of democracy, such as
defining democracy and fitting the different variants of democratic regimes into
classifications of democratic regimes.8 Increasingly, more studies have turned to
questions on the substance (or results) of democratic regimes, and a recurrent question
refers to why some regimes succeed in their transition to democracy and eventual
consolidation, while others fail somewhere along the way.9 This scholarly concern is
clearly summed up by Diamond, Plattner, and Schedler in the following way:
“Democracy, like so many objects of desire, is a moving target... now that these polities
have reached democratic shores, often after years of intense struggle, they are discovering
that they cannot just lean back, relax, and enjoy the democratic sun. Instead, they find
themselves haunted by old demons that they had hoped to exorcise with democratic
rule...” (l999:1).

Broadly speaking, the literature on democracy and processes of democratization

has focused on the study of political institutions and political actors. Scholars of political

 

6 See Diamond (1999) for different conceptualizations.

7 According to Freedom House (www.freedomhouse.or2), the number of “free” countries increased from 40
(or 25%) in 1975 to 89 (or 46%) in 2005.

8 See, for example, Collier and Levitsky (1997); Diamond (1999); Linz and Stepan (1996); Schmitter and
Karl (1991).

9 For a recent study, see Diamond and Morlino (2005).

 

15

institutions have focused, for example, on comparing parliamentary versus presidential
systems; on the nature of representation as a result of electoral systems; and on the
overall institutionalization processes (e.g., Linz 1990; Linz and Stepan 1996; Mainwaring
1993; Jones 1995; O’Donnell 1994; O’Donnell, Schmitter, and Whitehead 1986; Shugart
and Carey 1992). Scholars studying political actors, on the other hand, explore the extent
to which political elites, parties, organizations, and citizens accept democracy as the
“only game in town,” and the overall role of civil society in a democracy (e.g., Booth and
Richard 1998; Putnam 1993).

To a great extent, a democracy’s success in consolidating10 is a function of the
balance reached between institutions and political actors. As Diamond and Chull Shin so
clearly explain: “Democratic consolidation will advance to the extent that the political
institutions of democracy are deepened and improved to become more open, responsive,
accountable, and respectful of the law and to the extent that democracy is seen by the
mass public to be delivering the political goods it promises: freedom, justice,
transparency, participation, and a predictable, stable, constitutional order” (2000:21;
italicized emphasis added).

Without doubt, representation lies at the heart of the link between citizens and
government. Scholars agree that in order to become consolidated, a democratic regime
must succeed in developing legitimacy among political elites and citizens (Diamond
1999). Citizens are engaged when they participate according to the “rules of the game” of
a democracy and when they know that their participation is meaningful to the system. As

explained by Diamond, “at the level of the mass public, consolidation is indicated when

 

’0 A regime is consolidated when a) all political actors are constrained by the rules of the game b) majority
of public opinion accepts legitimacy of democracy. Definition follows Linz and Stepan (1996) chapter 2,
and Diamond (1999), chapter 3.

l6

 

 

 

 

Ca)

 

the overwhelming majority of citizens believe that democracy is the best form of
government for their country at their time” (1999:68). However, as discussed by Bratton
and van de Walle (1997:11-12), “...for ordinary people — in Africa as elsewhere in the
world - the challenge of democracy concerns how to obtain accountability from officials
of the state.”

The quality of elections is one way of evaluating the extent to which political
accountability is being institutionalized. However, “elections alone are not a sufficient
condition of democracy” (Bratton and van de Walle 1997213). In this respect, Karl (1990)
warned against the dangers of falling into an “electoral fallacy”—that is, equating
elections with democracy. In fact, elections can still be held in regimes where systematic
abuses of political rights take place. On the other hand, it is also “not sufficient for
government to be responsive to citizen desires; citizens must also be able to hold
government accountable” (Gomez and Wilson 20062127). Hence, elections can only be
an indicator of a regime moving towards consolidation, when parties and political elites
allow elections to serve the purpose of accountability and informed voters use elections
in this manner. But, when does this happen?

Voters will hold the government accountable only if they believe that the rules of
the game are accepted and enforced, and that there are self-enforcing restraints on the
government. That is, an electoral democracy meets partially the first requirement about
the balance needed in the system to consolidate—that of working political institutions.
Elections, in a liberal democracy,” remain the most straightforward mechanism of

political participation, but beyond meeting the minimum requirements, they serve in a

 

” See Diamond (1999) on the distinction between liberal and electoral democracy.

l7

broader meaning of representation, because political rights and civil liberties are
guaranteed. In such a context, the democratic quality of citizens increases.

A growing body of theoretical work and empirical evidence suggests that a
vigorous civil society is instrumental, if not vital, to the consolidation of new
democracies.12 Scholars have argued the manner in which civil society organizations and
movements provide outlets of participation, and how such networks foster trust and
cooperation—all of which leads to democratic consolidation. The institutionalization of
elections, which link the government and civil society, lies at the heart of democratic
consolidation. Ideally, beyond being open, free, competitive, and regular,

institutionalized elections are used by voters to express preferences and evaluations.

2.3 Economic Voting

One way to measure whether elections are serving as a mechanism of
accountability between elected officials and citizens is by exploring how much voters
actually reward or punish their governments when it comes to voting or expressing
approval or disapproval towards the incumbent government. Indeed, this straightforward
relationship between governmental performance and citizens’ evaluations has been
developed as a barometer of institutional performance as seen through the eyes of the
citizenry. As argued by Dorussen and Palmer, “economic voting is an instrumental act.
Ideally, the electorate expresses at the ballot its preferences about proper management of
the economy...Politicians who fail to heed the policy goals of the public, or who show a

lack of competence in ‘managing’ the economy, are turned out of office. Economic

 

‘2 On civil society and democratic consolidation, see Diamond and Plattner (1994); Foley and Edwards
(1997); Gyimah-Boadi (1998); Putnam (1993); Schmitter (1997).

18

voting thus constitutes a mechanism of democratic accountability” (2002:1-2). Drawing
primarily on the experience of developed democracies, a large number of studies have
demonstrated that governmental leaders and political parties are either punished or
rewarded for their economic performance at the time of elections. The responsibility
hypothesis, which in its simplified form states that “voters hold the government
responsible for economic events” has driven the economic voting literature (Lewis-Beck
and Paldam 20002114). This straightforward relationship has been extensively
researched, producing hundreds of articles and books during the last decade.13 In fact, the
field of economic voting remains one of the favorites among scholars to study voting
models, partly because of the data availability and reliability, but also because efforts to
test the theory have produced many other open questions. Overall, the theory lends itself
to study electoral behavior from a wide array of angles.

Economic voting refers to the basic following idea, which was articulated in 1978
by Tufte (65): “When you think economics, think elections; When you think elections,
think economics.” In fact, this field began being articulated a few years earlier, when
Kramer (1971) concluded that the fate of an incumbent president’s party depended
largely on the state of the economy. This hypothesis was further developed on the basis
of rational behavior theories, which basically argue that voters follow their self-interest
and are capable of thinking rationally when casting a vote. That is, according to Downs’
Economic Theory of Democracy, voters act rationally in politics and seek to maximize
their “utility income from governmental activity” at the time they make an electoral

choice (1957237). Thus, according to the theory of economic voting, voters who base

 

'3 For a complete review of the economic voting literature, see Lewis-Beck and Stegmaier (2000).

19

their vote choice primarily on their evaluations of the sitting government, are making a
rational choice.

Scholars of economic voting have been immersed in different debates over the
exact dynamics underlying the voting calculus. For instance, at the aggregate level,
questions revolve around which macroeconomic indicators best explain the relationship
between economics and electoral outcomes. Other questions deal with the dimension
voters use to evaluate their government (pocketbook versus sociotropic), or by the
timeframe they consider (prospective versus retrospective). And yet another question
refers to the layer of government that voters look at when they evaluate (national versus
subnational). Despite these debates, however, there is an overall consensus in the
literature supporting the link between economics and elections. In short, most scholars
agree that the “economy matters” on election day. Next, I briefly review the evidence
found in the American context and in other Western developed democracies.

At the aggregate level, scholars who have explored the impact of macroeconomic
performance on presidential approval or electoral outcomes, (e.g., Kramer 1971; Hibbs,
Rivers, and Vasilatos 1982; Lewis-Beck and Rice 1992; Markus 1988; Norpoth 1984;
Tufte 1978) argue that incumbent support is a function of its economic performance.
Most studies using macro-level data usually link aggregate percentage support for the
incumbent government with different macroeconomic indicators, such as unemployment,
inflation, and economic growth rates. Although there is some disagreement in the
literature as to which macroeconomic indicator has the most explanatory power,
macroeconomic performance has been consistently found to be part of the voting

calculus.

20

At the individual-level, researchers have explored the link between individual
perceptions of the economy as a whole or their personal financial situation and vote
choice or presidential approval (e.g., F iorina 1978, 1981; Key 1966; Kinder and Kiewiet
1979, 1981; Kiewiet 1981). Typically, micro-level studies include, in addition to the
standard indicators of individual assessments of the economy, a number of
socioeconomic and political control variables, such as income and education levels, and
partisan identification. Within this line of literature, there are also a number of
disagreements and unresolved questions but, despite them, scholars have found overall
considerable evidence in support for the economic voting argument.

The classic question of how citizens hold the incumbent government accountable
for the performance of the economy has been answered in two main ways. On the one
hand, the classic economic voting hypothesis presumes that voters assess their personal
financial situation, or “pocketbook,” to make a vote choice. In fact, the pocketbook
voting hypothesis lies at the core of the original theory underlying the economic voting
literature. The idea that individuals “vote their pocketbooks” (Tufte 1978) originated in
Downs’ (1957) research, who argued that voters make a rational decision when they vote
by either rewarding or punishing the government, according to their personal experience
and calculus on election day.

However, results from the pocketbook hypothesis at the individual-level have
been mixed or in some cases inconclusive (Fiorina 1978; Klorman 1978; Lewis-Beck
1988). These results have led other scholars to argue that individuals are more likely to be
sociotropic rather than pocketbook voters. According to the sociotropic approach, voters

tend to put more emphasis on the broader state of the economy in the country, rather than

21

 

 

dc
re
“'1'.
ant
Ma
prn

pri:

 

focusing on their own personal financial well-being. For instance, Kinder and Kiewiet
(1978, 1981) focused on the relationship between individual assessments of the general
state of the economy and individual vote choice. Overall, they find that personal
experiences are “politically unimportant” (19792522). That is, according to these authors,
personal (economic) problems are important to citizens, but they do not expect the
government to be responsible for solving these. On the other hand, Kinder and Kiewiet
find that citizens do identify the broader, collective state of the economy as politically
relevant.

A different debate within the literature deals with the timeframe that voters use
when they evaluate their governments. On the one hand, the classic economic voting
hypothesis, following arguments developed by Downs (1957) and V.O. Key (1964),
presumes that voters are self-interested and retrospective in nature. On the other hand,
other scholars have put forward the argument that when individuals make a voting
decision, they look at the prospects of the economy. That is, according to this line of
research, voters evaluate the economy prospectively—they seek the option they perceive
will benefit them the most in the future (Chappell and Keech 1985; MackKuen, Erikson,
and Stimson 1992). The well-known terms of “banker” and “peasant” introduced by
MacKuen, Erikson, and Stimson (1992), best illustrate the distinction between a
prospective and a retrospective voter. According with this analogy, while a peasant pays
primarily attention on his or her personal short-term circumstances, a banker focuses on
the long-terrn horizon, being more sophisticated and objective in his or her decision-
making process. With respect to the US, the authors conclude that “the electorate

responds with the sophistication of the banker” (19922606). Again, despite existing

22

 

 

 

rh

Il7c‘

ret~

disagreements, both approaches have found sufficient support in the literature (Lewis-
Beck and Paldam 2000:114).

Finally, yet another disagreement found in the economic voting literature, and one
that is highly relevant to the question of democratic accountability, refers to whether the
electorate is capable of assigning responsibility for economic performance to the
appropriate level of government. That is, do electoral results at the subnational level of
government respond to national economic and political trends rather than local
conditions? Are the performance failures of subnational officials correctly punished by
voters? The results of studies looking at both possibilities are mixed as well.

According to the referendum-voting model, which lies at the heart of the
economic voting theory, electoral outcomes are a result of the incumbent presidential
administration’s performance. That is, the higher the approval ratings of the president
(and/or the higher the levels of popular satisfaction with the state of the national
economy), the higher the likelihood that the president’s party (or he himself) will be
reelected. Consequently, state-level political leaders are not evaluated on the basis of
their own performance, but rather on that of the president’s party. It is expected that, if
the president’s enjoys positive evaluations, his party will benefit at the state-level,
regardless of the specific performance of subnational politicians.

Significant macro-level research applied to the American context suggests that
voters see the President as solely responsible for national policy outcomes (e.g., Chubb
1988; Erickson 1989; Holbrook-Provow 1987; Peltzman 1987; Piereson 1975; Simon,
Ostrom, and Marra 1991). On the other hand, scholars have found support for the

argument that voters do look at state-level economic performance when they make a

23

 

0\

C21

Pa

1'65

511.

for

the

SCI

 

my,

 

subnational electoral decision (e.g., Atkeson and Partin 1995; Lowry, Alt, and Ferree
1998). Micro-level data exploring this question has found mixed evidence as well. While
some scholars find that voting in subnational elections reflects evaluations of national
performance (Simon 1989; Stein 1990), other researchers have found that voters do
distinguish between subnational and national policy responsibilities (e.g., Atkeson and
Partin 1995; Carsey and Wright 1998; Howell and Vanderleeuw 1990; Partin 1995). Still
other studies suggest that voters use a combination of national and state-level evaluations
(King 2001; Svoboda 1995).

As shown, despite the different debates taking place in the literature, there is an
overarching agreement among scholars that identifies economics as part of the decision
calculus. However, researchers who have reviewed the literature (Lewis-Beck and
Paldam 2000; Nannestad and Paldam 1994) agree that the theory presents a problem of
“spatial and temporal instability.” In other words, the problem with the theory is that
results so far exhibit variability across countries, and even within countries over time.
Still other scholars argue that the economic voting theory does a poor job in accounting
for individual-level differences (Duch 2001; Gomez and Wilson 2001, 2006). Moreover,
the instability problems of the theory are exacerbated when it travels to new democratic
settings, where severe limitations to mechanisms of governmental accountability exist
(for example, extreme centralization of powers, clientelist practices, or low levels of
institutionalization of party systems), and the conventional economic voting model
cannot be as widely generalized among the population as it is done in advanced
democracies. The next section reviews the state of the economic voting literature in new

democracies.

24

 

h:

Cl

511
th’.
\‘01
pril
rel;

EVE

der

pre

 

all

no

 

do

eCl

 

2.4 Economic Voting in New Democracies

Are voters in new democracies as capable as those in established democracies to
hold their governments accountable? Until recently, most studies had examined economic
voting in the American context and Western European democracies. This is surprising on
the one hand, given the obvious electoral importance that economic issues should have in
those volatile contexts of the developing world. In fact, elections represent for many
citizens in developing democracies the only opportunity to express their policy
preferences. On the other hand, however, it is understandable to some extent that such
studies did not focus on new democracies until recently, considering the common belief
that electorates in these contexts are not “highly informed, policy-oriented economic
voters” (Dorussen and Palmer 200222). Electoral behavior in new democracies has been
primarily explained by factors such as ideological propensity, presence of patron-client
relationships, or socio-demographic variables, rather than by voters’ performance
evaluations. Voters in these contexts are easily assumed to lack the needed conditions to
be critical voters, such as access to a free press. The applicability of the theory in new
democracies is also constrained by the wide variation in political and economic contexts
present in new democracies, in contrast with those of advanced democracies. Data
availability and reliability have been a constraint as well in exploring economic voting in
fledging democracies. But, as I detail in the next section, scholars of democratic
development have increasingly applied the model to study citizen responses to the
economy in new democracies. However, most of the evidence found so far is mixed or
questions altogether the applicability of the model to nascent democracies. Next, I review

a number of studies that have found clear evidence in support for the economic voting

25

 

 

 

 

ICI

P0

prr

tht'

 

be

 

lllt

01\

R11

 

theory, and then I mention studies that have found weak or no support at all for the
theory.

One of the early studies looking at the link between the economy and elections
outside the context of established democracies was Remmer’s (1991) analysis of 21
presidential elections in Latin America. Looking at a time period dominated by the harsh
regional economic crises of the 19805, Remmer finds an important relationship between
economic conditions and vote choice. Also in Latin America, Seligson and Gomez (1989)
find evidence in Costa Rica’s 1986 general election. More recently, Gélineau (2002)
finds support for the theory looking at Argentina, Brazil, and Venezuela during the 19805
and 19903. Specifically, he finds that economic voting is contingent on the economic
(e.g., inflation, unemployment) and political (e.g., reelection, degree of decentralization)
contexts. In a cross-national study, looking at elections taking place between 1988 and
1994 in 38 countries, Wilkin, Haller, and Norpoth (1997) found strong support for the
retrospective voting argument. These authors argue that, despite the wide variation in
political contexts across the nations included in their study, “voters around the world
prove capable of practicing a standard form of economic voting. They do so by holding
the major incumbent party accountable for recent economic performance” (301).

The return of former communist parties in several post-communist regimes has
been a phenomenon that scholars have tried to explain with the economic voting theory.
A number of studies have found support for the argument that citizens’ evaluations of
their nation’s economic situation play an important causal role on the electoral fortunes
of incumbent governments (e.g., Duch 1993; Markowski and Toka 1998; Mishler and

Rose 1996). For example, Markowski and Toka (1998) found that the electoral recoveries

26

in the early 19903 of former communist parties in Hungary and Poland can be explained
by the referendum model of economic voting. According to the findings of these authors,
citizens in these countries punished the sitting governments for economic downturns by
returning old communist parties back to power.

Other studies have found mixed support for the economic voting model. In a
study looking at the 1998 presidential election in Venezuela, Weyland (2003) finds that,
on the one hand, economic evaluations appear to have had a significant effect on
President Hugo Chavez’s victory, but he also finds that the candidate’s personal
characteristics had important effects on voters’ decision to support him. Another study
that finds weak support for the economic voting hypothesis looks at Zambia, where the
authors find much stronger evidence in support of non-ecOnomic factors, such ethnic
group membership, traditional party affiliations, and urban/rural location (Posner and
Simon 2002).

A number of other works on new democracies question altogether the
appropriateness of the conventional economic voting model. In a study looking at
incumbent support and popular support for economic reform policies in Poland,
Przeworski (1996) finds that, in contrast with the expectations of the economic voting
model, higher inflation increased support for the incumbent. Similarly, looking at the
1993 parliamentary elections in Poland, Powers and Cox (1997) find that voting behavior
could hardly be explained by citizens’ economic evaluations. The authors find that other
issues related to social legacies of the communist regime were more salient to voters. , in
a study seeking to explain the replacement of incumbent promarket, prodemocracy

governments with ex-communist parties in Lithuania, Hungary, and Bulgaria, Harper

27

(2000) finds weak support for the economic voting hypothesis. Exploring the case of
Mexico, Dominguez and McCann (1996) also find that retrospective and prospective
assessments of the nation’s and one’s own economic conditions were only marginally
related to voter intentions. They concluded that partisan, institutional, and candidate
assessments provided the stronger bases for the divisions in Mexican mass public
opinion.

Overall, the mixed findings found in the literature of economic voting in new
democratic settings suggest that voters in a transition context may not automatically
respond to economic outcomes according to the conventional model, that is, by voting for
the incumbent when times are good and vice versa. It appears the model must recognize
that there is variability in the strength of economic voting resulting from different
institutional contexts and heterogeneous populations that respond differently to economic
outcomes (2001).

Taking sides with this argument, I argue that the standard economic voting model
cannot be applied in the same manner across developing democracies as it is in
established democracies, because new democratic settings lack often the necessary
contextual conditions that enable voters to act as economic voters. Furthermore, I argue
that the tendency of an electorate to cast an economic vote varies also within a nation,
because the levels of democratization can vary from region to region. I want here to refer
to an analogy based the following statement by Norpoth (19962317) “there are signs that
the inclination to [engage in economic voting] is hard-wired into the brain of citizens in
democracies.” I do not debate the claim that citizens in general might be predisposed to

be economic voters. However, I take issue with the claim that they are actually engaged

28

economic voters in any given democratic context. I argue that in a democracy, the
“switch” has to be “turned on” in order to let citizens act as economic voters. And some
democracies might not have enough “voltage” to produce economic voters. In line with
this argument, I defend the claim that institutional and contextual arrangements are key
determinants of economic voting. Institutions, like “flow regulators,” are in charge of
distributing the “energy” (be it in the form of increased information flows, trust from
citizens towards the government, or governmental responsiveness, for example) which
might explain the heterogeneity of economic voting in consolidating democracies.
Building on this argument, I propose that decentralization can be regarded as a potential

“generator” of economic voting in new democracies.

2.5 Accountability and Decentralization in New Democracies

As mentioned before, the relationship between citizens and the government is at
the heart of the challenges faced by modern representative democracies. Indeed, as
argued by Diamond (19992145), “creating the formal institutions of democratic local
government does not guarantee that elected officials will be open and responsive or that
citizens will take the opportunity to get involved. The critical link between formal
institutions of local democracy and more vibrant citizen involvement is civil society.”
Diamond shows that “the grth of civil society and the decentralization of political
power are parallel processes, closely intertwined in the development of democracy”
(19992145). One policy that has been enacted to address this problem in old and new
democracies alike is the decentralization of powers from the center to the periphery.

Decentralization’s broad purpose is to create an institutional context in which civil

29

society can participate more and government can respond more. In other words, one of
decentralization’s indirect effects should be to facilitate accountability” from political
leaders to citizens. One way for political elites, parties, and organizations to show their
commitment to democracy is by fostering the deepening of democracy at the local level.
“Democratic decentralization requires by definition the transfer of control to lower levels
of elected government—the devolution of real authority over some aspects of the policy
agenda” (Diamond 1999: 149).

A popular and complex concept, which encompasses many different aspects,
decentralization demands proper conceptualization. This study follows Rondinelli and
Nellis’ (1986; cited in Rondinelli 1989281) definition, which states that decentralization is
“the transfer of responsibility for planning, management, and the raising and allocation of
resources from the central government and its agencies to field units of government
agencies, subordinate units or levels of government, semi-autonomous public authorities,
or non-govemmental private or voluntary organizations.” Decentralization is categorized
into different dimensions,15 among the most common are: political, fiscal, and
administrative (Montero and Samuels 200423). They all share the overall goal of “the
dispersion of functions and power from the center to the periphery” (Rodriguez 199729).

Political decentralization refers to “direct elections of state/provincial and/or local

political offices” (Montero and Samuels 200425), and implies greater electoral

 

’4 Here I follow Schmitter’s (2004:47) definition of political accountability as a “relationship between two
sets of persons or (more often) organizations in which the former agree to keep the latter informed, to offer
them explanations for decisions made, and to submit to any predetermined sanctions that they may impose.
The latter, meanwhile, are subject to the command of the former, must provide required information,
explain obedience or disobedience to the commands thereof, and accept the consequences for things done
or left undone.”

'5 Decentralization is categorized by various scholars in somewhat different ways. Rodriguez (1997), for
example, refers to four types of decentralization: political, spatial, administrative, and market
decentralization/privatization.

30

competition at the subnational level. When political powers are decentralized, it is
assumed that both citizen participation and representative government will increase
(Rodriguez 199729). By having more autonomy from the center, subnational politicians
should depend also directly on their constituents’ electoral support — this is one reason
why electoral competition increases. However, and as Montero and Samuels (200427)
warn, in practical terms, political decentralization by itself does not mean that the
subnational unit is fully “decentralized.” If the state/province is politically decentralized,
but lacks resources or policy autonomy, open and competititve elections become fairly
meaningless. Thus, political decentralization is highly interrelated with the other two
dimensions: Fiscal decentralization, which refers to the degree of autonomy subnational
units have over their resources (both revenues and expenditures); and administrative (or
policy) decentralization, which is defined as “the relative authority or responsibility that
state/provincial and local governments have to set goals, muster resources, and
administer and implement public policy” (Montero and Samuels 200427). Only the
interplay between political power and resources can help overcome the common
“overburdened and underresourced” problem (Diamond 19992140).

Since the 19708, decentralization has been placed high on the political agendas of
many nations across the world. Each nation’s individual motivations vary, but the basic
argument is that decentralization increases governmental efficiency of service provisions
and responsiveness at the local level (Rondinelli 1989). An overarching argument leading
decentralization efforts in new democracies is that by delegating powers to lower levels
of government, the central government can be prevented from preserving its authoritarian

(central) grip of power. This is especially true in post-colonial Africa, post-authoritarian

3l

Latin America, and post-communist Eastern Europe. These regions shared the
commonality of tight central powers that eventually grew tainted by corruption,
inefficiency, and unresponsiveness. To dilute this centralist tradition, decentralization of
state power and democratization of government at the local level was adopted in many
nations. Pressures and/or incentives to decentralize came from various sources, such as
neoliberal reforms, international lending agencies, democratization processes,
sociostructural causes, and political-institutional forces (Montero and Samuels 2004).

A number of observers of democracy have argued that decentralization processes
have been key to democratic opening and furthermore, for deepening democracy. This
argument rests on the assumption that the size of a political unit is directly linked with the

success of its democratic system (Diamond with Tsalik 1999:117-160). That is, the

smaller a country (or political unit), the greater its likelihood to sustain its democracy.

This argument makes the case for devolution of power to lower levels of government
through institutional arrangements such as federalism and decentralization.

According to its supporters, decentralization translates into the empowering of

local-level actors, which develops grassroots democracy. Following Diamond (19992121-

122), local democracy sustains democracy in five major ways. First, decentralization

enhances citizen’s educated involvement in the democratic process. A strong local

government facilitates citizens’ participation in the decision-making and policymaking

processes. Citizens learn hereby firsthand about the workings of the government and are

increasingly aware of the value of democracy for the community as a whole. This

argument is strongly supported in Putnam’s study of Italy (1993), in which a strong

“civic community” is found to be determinant in sustaining democracy. Second,

32

 

decentralization is thought to increase accountability and responsiveness from the
government to the citizens. This argument follows the idea that the more informed and
engaged citizens are in the governmental process, the more likely they are to ask their
political leaders to address their needs. Overall, it has been found that citizens feel closer
to lower levels of government and therefore more confident and satisfied in their capacity
to influence local government in contrast with a distant national government (Diamond
1999:125-126). Third, local leaders are supposed to be more representative of the
community than national politicians. Fourth, the dynamics of a local government
facilitate the practice of checks and balances vis-a-vis the center. The idea is that it would
be far more difficult for the central government to abuse its power, if local governments —
via citizens— develop local oversight systems. Fifth and lastly, decentralization of
political power means that opposition parties have a real chance of winning office. Since
parties in power have a real probability of being defeated, opposition parties are more
likely to accept defeat. This clearly enhances governmental legitimacy both from the
parties’ and voters’ perspectives.

According to this view, decentralized governments are expected to increase both
the institutional clarity in the political context and citizens’ overall trust in the system.
However, a number of empirical studies exploring this link have found inconclusive or

mixed results. While some scholars (Putnam 1993; Seong 2000) have found evidence
supporting the positive relationship between decentralized local governments and
citizens’ satisfaction with the system, other researchers have questioned whether
decentralization actually results in “good” local government. As explained by Diamond

(1 99921 32-138), decentralization presents some potential problems. First, it can lead to

33

the preservation or creation of “authoritarian enclaves.” Recent studies on subnational
politics have demonstrated that decentralization can empower “bad” local governments as
much as “good” local governments (Cornelius 1999; Fox 1994). For example,
subnational populist leaders can mismanage decentralized resources in order to retain or
extend their oligarchical grip of power, which leads to corruption and political patronage
(Prud’homme 1995; Diamond 1999; Stepan 2000). Second, “small size may also create
strong pressures for conformity and intolerance of difference” (Diamond 19992134).
Third, revenue decentralization can create greater disparities across regions. For instance,
rural areas are put at a disadvantage vis-a-vis urban centers, because they do not have as
many resources to generate revenues by themselves. Looking at the effects of
redistributive policies within fiscal decentralization, Prud’homme (1995) has argued that
far from reaching equity across regions, decentralization policies are likely to exacerbate
disparities, if they are not accompanied with proper administrative checks and balances.
Similarly, other studies warn against the transfer of policy authority without adequate
financial support (e.g., Bahl and Linn 1994). Fourth, decentralization can lead to “waste,
redundancy, and confusion” (Diamond 19992136). For instance, services might be
duplicated if the lines of responsibility are blurred. Also, if subnational actors are given
extreme autonomy without major responsibility, the danger of gridlock is always there.
Diamond (19992137) refers to Brazil as an example of this trap, arguing that “it seems
beyond question that the level of political decentralization in Brazil constitutes a major
obstacle to democratic consolidation.”

This review suggests that the effectiveness of decentralization policies is

contingent upon different factors that vary across and within nations. These factors could

34

 

be “packaged” into the following two points. a) Governments (both incumbents and
opposition) need to depend on constituents’ electoral support for their survival, that is,
decentralization policies have a strategic power; and b) decentralization policies have to

be accompanied with adequate funding and enforcement mechanisms.

2.6 Conclusion

Drawing upon the empirical and theoretical debates reviewed above, I explore in
this dissertation hypotheses about economic voting at the subnational level of
government in Mexico. The broader purpose of this analysis is to explore the relevance of
decentralization measures in a new democracy on subnational electoral behavior. The
normative purpose is to find out whether and to what extent electoral accountability is
working within a new democracy.

Extending the theory of economic voting to the Mexican context presents
interesting theoretical challenges. On the one hand, given Mexico’s political and
economic contexts, the presence of economic voting appears questionable. To begin with,
the centralized nature of the 71-year PRI regime did not foster the development of a
responsive relationship between the government and citizens. Further, the strong
presidential system leaves the extent to which voters can distinguish between
presidential, congressional, and moreover, subnational policy realms open to debate. The
low level of institutionalization of the party system (Mainwaring 1999) and the

clientelistic nature of politics (Fox 1994) suggest also important limitations to
governmental accountability. Finally, the study of economic voting in the Mexican

context is a challenging one, because there is no reelection of any kind. Reelection is

35

 

generally assumed to be a key element for accountability to work (Manin, Przeworski,
and Stokes 1999248).

Despite these theoretical challenges, this dissertation aims to shed light on this
understudied but critical question of democratic accountability. Thus, the central
argument of this study is that Mexican citizens behave as economic voters, but that their
propensity to do varies with the degree of decentralization of the state they live in. Thus,
voters in more decentralized states will be more likely to be economic voters. This
happens, I argue, because more decentralized states will have the needed contextual
conditions (institutions) that allow voters to use elections as an accountability mechanism
(vertical accountability), and make politicians responsive. State-level politicians, in turn,
will be more likely to develop greater autonomy from the central government and take
their constituents into account, because political leaders derive their power primarily
from subnational electoral support. Following Garman, Haggard, and Willis (2001), I
argue that the success of decentralization depends largely on the political incentives of
national-, state- and local-level authorities. Only if they derive their political power from
local constituencies and they have the needed financial viability to fulfill their

responsibilities, then decentralization is likely to function, increasing the levels of clarity

of responsibility and, eventually, producing accountability.

36

Chapter 3. The Mexican Case

3.1 Introduction

This chapter provides the contextual framework for analyzing the Mexican case.
It concentrates on the transformation of the contemporary Mexican political system. In
doing so, I review the political trends that took place in Mexico’s recent history.
Specifically, I discuss the highly centralized one-party regime established by the PRI, in
which mechanisms of accountability were seriously curtailed. I also highlight the gradual
development of a series of reforms that led to the slow political liberalization of the
Mexican political system. Among the major political changes that the various sets of
reforms unleashed, I focus on the decentralization of political and economic powers,

which so dramatically reshaped the world of subnational elections, as well as relations

between the center and the periphery.

The chapter is divided into three parts. The next section draws upon history to
describe the institutionalization processes that defined Mexico’s (un)democratic
evolution and present “difficult democracy.”l6 This allows exploring the nature of the
post-revolutionary regime, in particular the institutionalization process that worked so
well to achieve and maintain political stability, as well as steady economic development
at least until the 19705, but also to hinder the development of a competitive party system.
This section looks at the resulting intergovernmental relations within Mexico’s powerful
presidential system, in which the figure of the president grew legitimate but not

accountable to the citizens (Levy and Bruhn 20012132). The third section reviews

 

’6 A term coined by Vargas Llosa (2000), after Mexico formally transited to democracy in 2000. Earlier,
Vargas Llosa had referred to the Mexican system as the “perfect dictatorship.”

37

Mexico’s political liberalization process, which led to the gradual empowerment of the
opposition and, eventually, to the culmination of the democratic transition in 2000. I
emphasize the high levels of variation in terms of democratic development that exist
across the Mexican states. Finally, I review the literature looking at the Mexican voter,

setting the stage for the empirical analyses in chapters 4 and 5.

3.2 The Hegemonic Party System

Until recently, Mexico’s political system was best known for its “longest one-
party rule.” The PR] achieved during its 71-year rule to establish itself in such a powerful
position that, for most Mexicans, the party was equivalent to the Mexican state. In this
section, two institutional legacies are discussed: the PRI’s one-party rule, and the

centralized inter-governmental system.

Without a doubt, the Mexican Revolution (1910-1917) had lasting consequences
that shaped post-revolutionary politics in Mexico. Violence did not end with the end of
the revolution.17 In 1917, the country was in need of some kind of peaceful mechanism to
create political order and bring unity. To fill in this vacuum, a political party was
fonned—a party that would encompass all power factions into a strong central authority,
which would in turn control the distribution of power among selected powerholders. As
Mexican political historian Lorenzo Meyer so clearly stated, since its formation, the goal

of the party was not electoral politics but to “administer power without sharing it”

(1986231).

 

’7 See Meyer and Aguilar Camin (1986) for a detailed account of the immediate post-revolutionary period.

38

Following the founding 1929 of the Institutional Revolutionary Party (Partido

Revolucionario lnstitucional, PR1),18 the government reached out to incorporate different

interest groups into its structure.19 Institutionalization was the regime’s principal tool to

centralize power, and a huge strategic success, as Levy and Bruhn argue: “...the genius

of institutionalization lay in how the regime attained such extensive control without more

blatant and brutal repression” (2001259). Among the most important groups incorporated

formally and informally in Mexico’s interest group structure were the military, the

Catholic Church, business, peasantry, organized labor, intellectuals, and the media.

Indeed, “By incorporating the organs of mobilization, the PNR controlled participation”

(Klesner 1988). Corporatism, defined as “a formal relationship between selected groups

or institutions and the government or state” (Camp 2003212), was the PRI’s device to

channel the most influential groups’ demands (the labor, agrarian, and popular sectors),
enabling the government to control public demands. Thus, through a series of contract-

like relationships, in which the regime provided resources and/or support to advance
political careers in exchange for electoral support. The PRI managed to develop an
almost incontestable control. This dynamic had serious consequences for the further
development of institutions that were' designed to guarantee competitive and democratic
participation, but in practice were seriously weakened. Hence, the resulting regime was a
stable one, because it maintained a tight base of support from a wide array of interest
groups, but it was, I argue, also a regime that did not produce a culture of accountability

in its citizens.

 

'8 The PR] had two antecedents: in 1929, the party was named National Revolutionary Party (Partido
Nacional Revolucionario, PNR), then it was shortly renamed the Party of the Mexican Revolution (Partido
de la Revolucion Mexicana, PRM). It finally acquired its current name in 1946. For a study looking at the
origins and transformation of the party, see Garrido (1989).

'9 For a detailed discussion on the role of groups in the Mexican political system, see Camp (19962125-

1 SO).

39

Accountability took only place in counted sets of relationships; among them
perhaps the most important occurred between the president,20 who personified the PRI’s
power, and the “revolutionary family” or leadership elite, which stemmed directly from
the FRI-structure.21 This system was based for the most part on bargaining relations that

sustained the PRI’s ruling elite across the nation. In contrast with other nations, there

was little overt repression; control was primarily exercised by either accepted political

bargain or cooption. Therefore, even though in formal terms a system of checks and

balances existed in Mexico, in practice it was overridden by an overpowering presidency,

which was based on a tight unified government, high party discipline, and by having the

president as the head of the party (Weldon 1997). Thus, political accountability existed

only to regulate these often corrupted power relationships that excluded almost entirely
citizens’ political participation.

The processes of political recruitment and nomination became crucial in the

formation and maintenance of this political elite (Camp 1980, 1996). At the very top of

the pyramid, the president recruited candidates based on loyalty ties, and distributed them

among the most prestigious positions within the PR1 and the government in general. This

rewarding system extended to the other levels of government; upper level politicians such

as governors in turn secured political loyalty by recruiting lower level functionaries.

772115, a successful political career depended upon establishing a patron-client relationship

with a successful political superior, because “hegemony meant that political ambition and

 

2° Mexico’s stron g presidentialism has been extensively studied. See, for example, Martinez Asssad (1992)
and Weldon (1 997).

2’ The “revolutionary family” comprised former presidents, powerful regional and national leaders, as well
as leaders of Mexico’s major interest groups. These interest groups cross-cut the entire political spectrum
from lefi to right, but still fit within the ideologically flexible “catch all” revolutionary family. The
structure of this group was maintained through loyal ties to the president, fear of political defeat, and self-
interest to retain power, prestige, and wealth. The resulting consensus, headed by the president, drove
Mexico’s political and economic development.

40

the attainment of office were not viable outside the party” (Diaz -Cayeros 1997). In fact,
this dynamic resulted in a hierarchical network of political cliques22 through which
politicians negotiated positions and guaranteed PRI’s hegemony at all levels of
government. This clientelist23 system extended from the president all the way to the lower

ranks of local government, and excluded the voter from any decision-making process.

Political and Economic Centralization
It has been frequently observed that Mexico “has a particularly centralist system
of government, even by Latin American standards” (Nickson 19952200). During the PRI-
regime, relations between the different levels of government were determined by the
highly centralized nature of Mexico’s political system. Even research on subnational
politics was largely understudied, and it was not until the 19905 that the explorarion of
subnational politics took life (e.g., Cornelius 1999; Lujambio 2000; Rodriguez 1997;

Rodriguez and Ward 1995).

Politically, governors and other state-level officials represented extensions of the
president’s power. Just as described before, governors owed their political loyalty to the
president, who directly appointed them. Therefore, the tacit agreement was that governors
would keep social and political calm in their states. They achieved this by appointing or
nominating other state-level officials, such as municipal presidents, making sure hereby
that their own collaborators would in turn govern in line with the PRI. However, if a

governor was not capable of regaining electoral support for the PR1 for some reason, then

 

22 In Mexico, these teams or groups are known as camarillas, defined as “a sort of nested series of
hierarchical groups under the aegis of a national or regional political leader” (Ward 19952138).

23 Clientelism is defined here as “a relationship based on political subordination in exchange for material
rewards” (Fox 1994: 1 53).

41

he was likely to be replaced. In fact, the rate at which presidents removed governors is a
good indicator of the degree of intrastate conflict in Mexico (Eisenstadt 2004).

Not surprisingly, these “governing” and “electoral” styles were frequently (and
increasingly) a source of conflict within states and municipalities. Governors and
municipal presidents chosen by the president or by the incumbent governor were often
unpopular with the general public and members of the local government (Cornelius and
Craig 1991226). Accountability mechanisms just could not develop within this closed
system. The value of regions, elections and that of voters were seriously harmed.
Mexican writer Carlos Monsivais succinctly captured this reality:

“The operation that stripped the provinces of their possibilities for

proportional growth and fair treatment had several angles. Politically, the

Partido Revolucionario Institucional (Institutional Revolutionary Party,

PRI) confiscated autonomy thanks to the wiles of corporativism, the

fusion of personal economic interests and administrative rights, the brutal

and systematic repression of dissidence, depoliticization (in many ways

the systematic exclusion of citizens from the public sphere), and

misinformation via the hijacking of interpretations and gross reduction in

access to news” (19922248).

Economically, post-revolutionary Mexico evolved in a parallel manner to the
political intergovernmental arrangement during this period: extremely centralized. Just as
the PRI-regime managed to control subnational leaders by political means, it managed as
well to centralize economic power. The adoption of import substitution industrialization

(ISI) was adopted as the official model of development in the early 19405, and it further
concentrated resources in the federal government. The ISI served the PRI to limit, in a
legitimate way, the authority of subnational leaders over financial matters, tightening

hereby control of the center over the periphery. The pursuit of a “national goal”

(industrialization) justified placing local interests to a second place. However, this was

42

only the official discourse. In fact, for politicians, the primary goal was to secure their
own political careers. Again, the PRI secured a federal arrangement that limited local
fiscal power by ensuring that local politicians could retain office, or by offering attractive
careers.

Along with ISI, a nationwide tax system established in 1948 served the PR1 as
another tool to centralize economic power. This fiscal agreement was reached through the
introduction of the national sales tax.24 Mexican politicians faced a tradeoff between
keeping control over financial resources and tax authority or reducing their electoral and
career risks. Eventually, local politicians decided for the latter (Diaz-Cayeros 1997). The
current intergovernmental fiscal arrangement of Mexico was established in 1980, with the
creation of the National System of Fiscal Coordination. States “voluntarily” joined the
system, giving up several taxing powers in exchange for revenue-sharing grants from the
federal government. Thus, the federal government consolidated its fiscal centralization by
becoming the only level of government allowed to levy taxes a wide array of issues.
States still retained theoretically the capacity to collect certain taxes, however, with time,
the federal government came to monopolize all sources of revenue and states grew
increasingly dependent on tax revenue shares (participaciones), which increasingly came
to constitute the most important source of local government finance (Figure 3.1).

In sum, the intergovernmental political arrangement developed by the PR1 was

based on a tight centralization of political and economic powers. Politically, regional
politicians accepted this arrangement because it was the only way to secure their political

ambitions, and because they required federal financial support to maintain their own

 

24 For an in-depth study of Mexico’s fiscal system during the first half of the 20'” century, see Diaz-Cayeros
(1 997).

43

electoral alignment. In this context, the electorate was excluded for the most part from the
political realm. Economically, the states grew increasingly (and to different degrees)
dependent on the federal govemment. Public revenues were used by the PRI as an
ideological weapon. Clearly, this arrangement hindered states’ capacity to manage their
own resources and be responsive to local constituencies, because there was no need to do

SO.

3.3 Reform and Transition

Since the 19705, the PRI was forced to gradually introduce a number of political
and electoral reforms. At first, these reforms were driven by the PRI’s political interest to
maintain its centralized control. Unintendedly, however, it permitted opposition parties
to compete effectively, which eventually favored greater autonomy for subnational
governments and unleashed the democratic transition process.

Political liberalization in Mexico dates from the political opening that occurred in
the early 19705 in response to increasing demands for greater democracy and
representation unleashed by the student movement of 1968. In an effort to both regain its
credibility and sustain its legitimacy, the PRI government increasingly experimented with
reallocating power by introducing proportional representation seats in the Congress; by
giving more autonomy to local governments; and by recognizing selected electoral
victories of opposition parties at both the state and the local levels. As Smith (1979)
emphasized earlier, the question was one of “limited pluralism;” the limits needed to be
extended, without sacrificing control over legitimacy. The reallocation of power in no

way meant any intention to loosen central control.

44

The PRI, however, miscalculated. As a result of the opposition parties acquiring
sufficient strength in the Chamber of Deputies (lower house) to have a say in the policy
process, major changes followed in electoral matter (see Figure 3.2). In 1989, the Federal
Electoral Institute (Instituto Federal Electoral, IF E) was created as a completely
autonomous institute in order to conduct the entire electoral process in a professional
manner. The establishment of the IF E is indeed a major breakthrough in the process of
Mexican democratization because it finally broke the direct link between the official
party and the institution that administers elections. The previous institution for
monitoring elections, the Federal Electoral Commission, was run directly by the interior
secretary as part of the executive branch. The Federal Electoral Tribunal (Tribunal
Federal Electoral, TRIFE) was also created in 1989 as an autonomous body capable of
challenging IFE decisions.25 The 1993 and 1994 electoral reforms extended the autonomy

and jurisdiction of the IFE, and further decreased PRI’s overrepresentation in Congress.

Measuring the Rise of the Opposition

The opposition in Mexico has been traditionally divided primarily into two parties
representing both sides of the political spectrum: on the right, the National Action Party
(Partido Accio’n Nacional, PAN) and the Party of the Democratic Revolution (Partido de
la Revolucion Democrdtica, PRD) on the left.26 Until the 19805, opposition political
parties were closely controlled or outlawed altogether. In fact, until 1979 only those
parties that did not threaten the PRI-regime (the so-called “satellite” parties) were

permitted to take part in elections.

 

2’ See Eisenstadt (2004) on the development of Mexico’s electoral justice.
2" For in-depth analyses of the PAN and PRD, see Mizrahi (2003) and Bruhn (1997), respectively.

45

To measure and distinguish among the different degrees of electoral competition
taking place across the Mexican states place from 1988 onwards, I applied two indexes
developed within the American context: the Ranney Index and the Holbrook-Van Dunk
Index. The Ranney Index is an indicator of the degree of two-party competition in the US
states. This index was developed in Ranney’s (1976) Parties in State Politics originally
published in 1965. The original measure is actually a measure of the strength of the
Democratic Party in state government. It takes into account the proportion of seats won in
the state House and Senate elections, the Democratic percentage in the gubernatorial
election, and the percentage of the time the govemorship and state legislature were
controlled by the Democratic Party. These factors are averaged together over a specified
period of time, yielding a measure that ranges from 0 (complete Republican domination)
to 1 (complete Democratic domination). The midpoint, of course, indicates evenly
divided control, or “perfect competition”.

I adapted the index to the case of Mexico by using three basic dimensions:

a) Proportion of success: the percentage of votes won by the PRI in gubernatorial
elections and the percentage of seats won by the parties in the (state/local) legislature

b) Duration of success: the length of time the parties controlled the govemorship and the
length of time the PRI controlled the legislature

c) Frequency of divided control: the proportion of time the govemorship and the
legislature has been divided

The resulting index is a measure of control of government, with a score of 0
indicating complete control by the opposition and a score of 1 indicating absolute PRI

control. The values of the Ranney party control index were calculated for two periods;

46

1972-1989, and 1990-2004. 1989 is taken as the limit of a one-party phase, taking the
first opposition victory of a govemorship as the threshold point. The results are presented
in Tables 3.1 and 3.2. Several patterns can be identified. During the first period, no state
qualified in either of the last three categories: two-party, modified one-party opposition,
and one-party Opposition. Only Nuevo Leon reached the modified-one party PRI
category. The results reflect the political scenario of a hegemonic party system: the PRI
had an indisputable advantage over the opposition. The scenario looks quite different
during the second period. No state falls into the one-party PRI category, while 18 states
qualify in the modified one-party PRI category and 12 in the two-party group, which can
be considered the ideal. Only Baja California falls in the modified one-party opposition
category. All two-party states had experienced opposition rule.

The utility of the Ranney index as a measure of party competition is realized by
folding the scale so the two noncompetitive extremes are brought together at the low end
and the midpoint becomes the high point: 1 — (Ranney - .5). The resulting index ranges
from .5 (no competition) to 1 (perfect competition). The Folded Ranney Index is
presented in Table 3.3.

Different authors have detected a number of limitations present in the Ranney
index; a) The index is based exclusively on state offices and does not reflect the strength
of the parties at other levels; b) the index gives more weight to some state offices than to
others; the way the index is constructed, the state legislature is given much more weight
than the govemorship; and c) this measure of interparty competition is a “snapshot of a
moving object” and does not capture change. The Hobrook-Van Dunk Index was

designed to address these limitations. It is based on district-level outcomes in state

47

 

lt’fl

 

legislative elections. When applying the index to the Mexican case, I used municipal-
elections data instead.

Complete absence of competition is indicated by a score of O on the district-level
measure. This would be a case where all candidates are unopposed. As the scale increases
from 0, it indicates greater competition. Table 3.3 presents the values for both the
Holbrook-Van Dunk index and the Folded Ranney index. The states are ordered
according to their value on the municipal-level measure. For both indicators, higher
values present more competition. The Holbrook-Van Dunk index seems to measure
competition much more directly than the (Folded) Ranney index. The strongest case for
face validity can be made for the municipal-level indicator. My reasoning is as follows:
the Ranney index is based largely on aggregate party strength in government, not on
actual election results. The only component of the Ranney index that takes into account
any individual election results is the gubernatorial election component. Given that only
one gubernatorial election is held every six years, it is possible that the results of this
election may not represent the true degree of competition in the state. In short, the
Ranney index is really a measure of party control of state government, not the
competitiveness of state elections. On the other hand, the Holbrook-Van Dunk index is
based entirely on actual outcomes in state legislative municipalities. Another difference is
that the Ranney index is specifically oriented to two-party competition, whereas the
municipal-level measure is based on overall electoral competition, regardless of the
number of parties.

There is some overlap between the two indices, suggesting that southern states

tend to be the least competitive and northern and center states tend to be the most

48

competitive. Also, both indexes (Figures 3.4 and 3.5) illustrate the enormous changes in
proportions of competition across time. Both indexes coincide on the two states with the
least degree of competition in the recent period: Hidalgo and Quintana Roo. According to
the Folded Ranney Index, Baja California, Baja California Sur, Chihuahua, Guanajuato,
and Michoacan have the most degrees of competition between 1990 and 2002, while in
the Holbrook-Van Dunk index the states of Aguascalientes, Jalisco, Nuevo Leon,
Querétaro, and Zacatecas appear also among the most competitive ones. The indexes
certainly exhibit the great increase in competition from one period to the next. And they
also show the prevailing unevenness of competition across states. Campeche, Hidalgo,

Nayarit, Oaxaca, and Quintana R00 are still remarkably behind.

Altemation and Divided Government

Mexico’s unicameral state legislatures were profoundly transformed during this
time as well. Just as was the case in the federal Congress, electoral reforms opened the
doors of local congresses to the opposition with the installment of proportional
representation (PR) seats (Figure 3.6). Shifting steadily towards pluralism, Figure 3.7
illustrates the dramatic transformations taking place in local-level politics: from complete
control of seats in local congresses until 1974, the PRI counted with just below 50
percent of total of seats by 2004. Figures 3.8 and 3.9 show similar trends through
different indicators of the growing presence of the opposition at the local level of
government. For instance, Figure 3.8 illustrates how one party (the PRI) had a majority

127

contro in all 32 local legislatures prior to 1988, excluding third parties from the

 

27 Majority here means at least two-thirds of total seats.

49

decision-making process. By 1993, the PRI had control over 23 local legislatures, and by
1999 only over two.

The new significance of local legislatures in Mexico has been extensively
explored by Ward (1995), Ward and Rodriguez (1999), and Lujambio (2000) who argue
that non-PRI governments at the local level have indeed presented different styles of
governing. For instance, when looking at early PAN governments in Chihuahua and
Ciudad Juarez, Ward (19952141) finds that “the panista administrations of 1983-96 were
characterized by operations that resembled business as much as public bureaucracies.
They sought to emphasize fiscal responsibility and avoided debt financing of public
works. . .There was greater transparency in municipal financing; budgets and expenditures
were posted and published regularly (an innovation that the PRI subsequently
maintained.” In a different study that looks at the six Mexican states of Aguascalientes,
Baja California, Chihuahua, Guanajuato, Puebla, and Oaxaca, Ward and Rodriguez
(1999) find evidence that local congresses are indeed exercising greater checks and
balances vis-a-vis the executive branch. In general terms, they find that contexts with
greater pluralism, electoral competition, and alternation of power tend to produce more
responsive non-PRI state executives to local legislatures. Within this context, both
government performance and voters began to matter.

Along this line, divided government became a novelty in Mexico. The first
divided government took place in Baja California in 1989. Between 1989 and 1994, six
more states experienced divided government, while between 1994 and 2000, fifteen more
cases were added to the list. At least in their early stages, divided governments have

served as an important arena of oversight at the local level, as parties have to explain, for

50

 

de
3131
ide.
obs.
of d

193

example, their decision-making on legislative matters to their constituents (Ward and
Rodriguez 1999; Lujambio 2000). It became clear that checks were beginning to be

institutionalized through the local congress (Ward and Rodriguez 19992700).

3.4 The Mexican Voter

So far, I have emphasized the effects of the PRI-regime on the institutions, and
have shown that accountability mechanisms started only to work when elections were no
longer only a ritual. That is, when politician’s “political survival” and ambition (Ames
1987) did no longer depend on one party, but on citizens’ electoral preferences. A crucial
question to address is, within this authoritarian context, what kind of electorate
developed? By reviewing the major trends of the literature on the Mexican voter, my
intention is to show that the Mexican electorate was highly constrained by the
authoritarian structures, and only slowly reached a level of politization that allowed
voters to effectively fight for their political preferences. In this section, I organize the
literature on Mexican voting behavior by making use the classification applied to the US
literature on electoral behavior by Dennis (1991).

The initial phase of electoral behavior studies in Mexico corresponds to a
descriptive approach. The basic question is “how did an individual vote?” and it hardly
approaches the question of “why?” Gonzalez Casanova (1965) is among the first authors
identified within the descriptive approach. In his work, Gonzalez Casanova made some
observations about the PRI’s electoral dominance. Following him, a significant number
of descriptive studies were developed, dealing with specific federal (e.g., Gonzalez Graf

1988; Martinez Assad 1992); state-level elections (e.g., Molinar 1986; Aziz Nassif 1994,

51

2000); and broader analyses of electoral geography (e.g., G6mez-Tagle 1989, 1992).
Gomez-Tagle has been considered the most representative author of this approach with
her work La dificultad de perder (1989), in which she follows the decline of the PRI’s
hegemony from 1961 to 1988. Her work does not include explicit models of electoral
behavior, but her analysis makes the assumption that the interaction between big actors,
such as the state and social groups explain this behavior.

More recently, descriptive studies have been advanced by more refined
measurement and description instruments, such as indexes of competition (e.g., Pacheco
1986). The use of public opinion polls has also developed the state of this line of research
by providing individual-level data. Political magazines, such as Este Pais (since 1991),
Nexos (since 1989), Etcétera (since 1993), V02 y Voto (since 1993), and the newspaper
group Reforma-E1 Norte (since 1994) have made extensive and intensive use of public
opinion polls, producing both basic reports in newspapers and magazines of a large
number of polls at the national and state levels, as well as launching important public
opinion-based scholarly work (e. g., Dominguez and Lawson 2004; Moreno 2003).

The sociological approach has dominated voting behavior studies in Mexico. The
most important variable used by a majority of these studies to explain the electoral bases
of political parties has been the urban-rural divide (Pacheco 1986). Other authors
(Molinar and Weldon 1990; Klesner 1995) considered other socioeconomic variables
(such as marginalization or literacy rates) and regional variables to describe electoral
behavior. One of the most representative works within this tradition is Zavala’s (1991)
analysis of the 1988 presidential election. He finds that variables, such as the number of

miles of major highways, the ratio of individuals in the lower economic strata, and the

52

ratio of Catholics and professionals within the population, were some of the explaining
factors of the PRI vote in 1988. He asserts that the PRI in 1988 owes its victory to the
rural population, to self-employed workers, labor unions, the unemployed, and the
illiterate (1991).

In Mexico, the development of the psychological approach followed to some
extent the Michigan School, but introduced a very important modification: the concept of
party identification was substituted by the one of political culture as axis of the
explaining causal models of electoral behavior. This was due to the hegemony of a single
party and the non-competitive character of the elections, which basically took away any
meaning to the party identification concept. In its place, the psychological approach in
Mexico sought to explain voting behavior, especially the electoral choice of Mexican
voters based on a complex group of affective and psychological attitudes and
predispositions encompassed in the concept of political culture.

The first research work making use of the political culture concept is the
landmark study The Civic Culture (Almond and Verba 1963). In regards to the Mexican
case, the authors found that only a minority of citizens possessed a participant culture,
while the great majority exhibits a parochial or subject political culture. They explain this
political culture pattern as being consistent with the authoritarian tendencies that they
observed in the political institutions at that time. That is, according to the authors, the
Mexican authoritarianism was based on strong socio-psychological basis: the
predominance of individuals with a parochial or subject political culture.

Without doubt, The Civic Culture is one of the most influential works in studies

on Mexican politics. In fact, during the following decade, over a dozen of other works

53

used the database that Almond and Verba gathered For example, Segovia (1975), in his
work La Politizacion del Nina Mexicano, found also evidence supporting the theory that
political culture was indeed responsible for the political structure of the country because,
within an authoritarian regime, political socialization is inevitably strongly influenced by
the state.

Other scholars developed the psychological approach further by leaving aside the
concept of political culture and including party identification. For instance, Moreno
(1993) applies directly the Michigan Model to the Mexican case. By retaking the concept
of party identification, he finds that in the 1991 mid-tenn election the most important
variable of the voter’s choice was the voter’s previous partisan sympathy.

Within the rational approach, an early study of this kind was undertaken by Ames
(1970). Based on the assumption that governmental policies affect people’s electoral
choices in an unequal manner, depending on the magnitude of the individual’s needs,
Ames sought to explain electoral participation and vote direction from 1952 to 1967 by
using aggregated state-level data. By measuring both direct (number of rank and file in
each state and the benefits that the government provided) and indirect factors (such as
urbanization, proximity to the US border, and the structure of the opposition), Ames
intended to find a correlation between public investment and PRI electoral victories.
However, he found that the indirect factors have a greater explanation power than the
direct ones.

Among the many studies looking at the electoral impacts of provision of public
goods, Molinar and Weldon (1994) analyze the impact of the controversial social

development program Solidarity (Programa Nacional de la Solidaridad, PRONASOL).

54

The authors looked at the electoral impact of this program and clearly found that electoral
and political conditions had a stronger say than poverty per se in the selection criteria.
They found that public investment was greater in those states in which the PRI got a clear
majority of votes in the 1988 federal elections. In these states, this study shows a clear
rational relationship between voters and the PRI. The outcomes were different in states
with higher opposition presence: in states where Cardenas with the CD (Corriente
Democra'tica) 1988 and the PRD 1991 were stronger, public investment was higher than
in states where the PAN was stronger. Molinar and Weldon conclude that these
differences are due to the different electoral bases of the opposition; while the PRD’s
base was similar to the PRI’s, the PAN’s basis was substantially different. It made
therefore more sense for the PRI-govemment to invest more in those PRD-strong states.
Baennan (1994) makes use as well of economic arguments to explain the left vote
in presidential elections between 1946 and 1988. His model uses aggregated data,
including balance trade, inflation rates, and income per capita. His hypothesis states that
as economic conditions worsen, Mexican voters vote left to punish the PRI, hoping that
the left would improve the economic situation. His results show that Mexican voters
respond indeed to such macroeconomic indicators by punishing or rewarding the PRI.
Among the first works using individual-level data, Kaufman (1971) and Coleman
(1975) are important to mention. Kaufman (1971) looked at the level of material
satisfaction and its relationship with political participation and support of political
institutions. He found a positive relationship between both material and political

satisfaction. On the other hand, Coleman (1975) found that PAN support is not

55

determined by a certain social class or religious group, but by voters who vote against the
PRI by voting for the PAN — a strategic vote.

More recently, Dominguez and McCann (1996) represented for a long time the
most important work of the rational approach. By analyzing two pre-electoral polls of
1988 and 1991 they introduced a two-step (or anti-PRI) model, which provided a rational
explanation of vote choice. According to the model, voters first asked themselves
whether or not they want to support the PRI. Second, if a voter decided not to support the
PRI, s/he would either abstain or vote for the opposition, regardless of whether the
opposing party with the highest probability of defeating the PRI coincided with the
voter’s ideological preferences. Thus, for a majority of voters, elections were nothing
more than a symbolic plebiscite or a strategic move.

Buendia (1995) questions Dominguez and McCann (1995) model by stressing the
importance of retrospective and prospective economic factors, presidential image, and
risk aversion to explain the 1994 presidential vote. His evidence comes from a
nationwide poll. Buendia finds that Mexican retrospective voting existed, but that it was
mainly a sociotropic vote and not a sociotropic one. He also finds strong evidence about
the effect of risk aversion hurting both opposition parties.

Moreno and Yanner (1995) present a comprehensive model of the 1994 election
based on a preelectoral nationwide poll and an exit poll. They find consistent evidence
that partisan identification was a strong predictor of the vote. Contradicting Dominguez
and McCann’s (1995) and Buendia’s (1995) claims, they see structural factors still
closely related to the direction of the vote. Their evidence regarding retrospective and, in

general, rational voting is not consistent across the two data sets. In the exit poll they find

56

strong evidence of retrospective and issue voting, but in the preelectoral poll they only
find some importance of specifically political issues.

Scholars have often tried to answer the following puzzling question: “Why was it
so natural for Mexican people to complain about their government officials and then vote
for them, even when democratic and institutional opportunities for a real change exist?”
(Cinta 19992174). For instance, Cinta (1999) and Magaloni (1994) look at this “fear of
change” and test the role of uncertainty in Mexico’s electoral arena. They find that, for a
long time, Mexican voters tended to be risk-averse. Effectively, the PRI was the party
that received the worst evaluations at the national level, and nevertheless, it represented
the option that received the highest number of votes. The PRI benefited from the record
of its past performance, a record that while not fully satisfactory to everybody, at least
guaranteed certainty about the govemment’s future performance.

In general, researchers of Mexican electoral behavior have agreed on three basic
assumptions. First, the dominant cleavage in Mexican politics has revolved around the
country’s political opposition against a crumbling authoritarian regime and its electoral
arm, the PRI (Dominguez and McCann 1995; Moreno 2003). The pro-govemment side
includes the principal beneficiaries of clientelism and one-party rule in Mexico:
government employees, owners of large businesses, members of the PRI’s state-
corporatist organizations, and peasants in villages traditionally favored by the
government. On the opposition side lie the professionals, small- and medium-sized
businessmen, independent unions and peasant organizations, and other groups that have

traditionally been abused by Mexico’s authoritarian system. In general, opposition voters

57

tend to be better educated, more engaged in politics, more urban, and younger (Moreno
2003).

Second, analysts have noted a pronounced and persistent regional cleavage that
divides the country roughly in half (Klesner 1995). In the North, West, and the state of
Yucatan—areas where the Catholic Church remains influential and local business elites
have long opposed centralized control from Mexico City—the country’s conservative
PAN constitutes the main opposition force. In the poorer, more rural South, the leftist
Party of the Democratic Revolution remains the principal challenge to PRI rule. Mexico
thus comprises a pair of essentially two-party systems—PRD-PRI in the South, PAN-PRI
in the North. Only in the capital and a handful of provincial areas does competition
regularly take on a multiparty character.

Third, scholars generally agree on the existence of a “floating” or strategic
opposition vote among either the PAN or the PRD (Dominguez and McCann 1996). In
the presidential elections of 1988 and 1994, for instance, the total opposition vote
remained roughly the same, at least according to official figures. But the distribution of
that vote among opposition parties varied dramatically. In 1988, the opposition sided
primarily with leftist candidate Cardenas, whereas in 1994 it sided with Diego Fernandez
de Cevallos of the PAN. This finding mirrors electoral swings in certain states, such as
San Luis Potosi and Baja California, where the anti-PRI voters appeared to strategically
switch between different opposition parties (Rodriguez and Ward 1995). The existence of
this floating opposition vote tends to reinforce Mexico’s regional divide, and it suggests

an element of rationality among Mexican voters.

58

.GmE “858

_1 ll.ll ill l l l lllllllllllll llllllllllllllllllllll 11' iii 1 llllll
z, 903:9.» .maoumu. 1:1 mozcgwm :30 II n,

filllltlltllll ll.ll

53>

e e z 1 z z z z 1 z z z z 1 z z z
0 0 0 0 0 no 0 0 0 0 0 0 0 0 0 0 0
Am % do do A0 ac % a. a. a. a. as o« a. e. e. 0/.

 

 

     

.. . 898m»

 

 

000600. F”

 

 

 

. . _ 8908.;

1')le dXW W

 

 

m o8_ooo.~»
8068.3

llllillll ill l l lyllll llll lll .lll.l l l l l l ll.l.llllrll ill. l.

«mafia ESE: 83.3.: £555: has. E can:

i
m
z
u
L
m
I
I
-...
_.

 

 

 

 

59

Figure 3.2 Number of Seats in the Chamber of Deputies, 1961-2003

 

 

 

 

 

500
400 2 ,_f pr, 1 ,,
L-0-Total seats
300 g l
g l—i—Relative Majority
a)
- t
200 {qr-Proportional
Representation
100 T “'T' T T T
0
N
b
'9

 

 

 

 

Sources: IFE and www.diputados.gob.mx

Figure 3.3 Party Affiliation of Governors in the 32 States, 1980-2004

 

301
1
25!

2o

15 ‘ +Opposition

t .__— .___.___._—..

10 ,.

 

0 ~ , —-v—r I —>7 fru—‘v—‘r

9 N '5 <9 '\ 9 N ‘b

 

 

 

Source: CIDAC www.cidacorgmx

 

60

Table 3.1 Ranney Index” in Mexican States, 1972-1989

 

 

One-Party PRI Modified One-Party PRI

Campeche Puebla Nuevo Leon

0.9737 0.9079 0.8482
Tabasco Tlaxcala

0.9668 0.9054
Hidalgo Baja California Sur

0.9626 0.9052
Guerrero Estado de Mexico

0.9595 0.9046
Zacatecas Yucatan

0.9564 0.9007
Querétaro Tamaulipas

0.9551 0.8986
San Luis Potosi Sinaloa

0.9434 0.8973
M ichoacan Chihuahua

0.9354 0.8936
Chiapas Baja California Sur

0.9343 0.8918
Morelos Durango

0.9321 0.8898
Quintana Roo Nayarit

0.9317 0.8883
Veracruz Sonora

0.9283 0.8757
Colima Aguascalientes

0.9266 0.8707
Coahuila Jalisco

0.9189 0.8545
Oaxaca Guanajuato

0.9103 0.8539

 

 

2’ The index is composed of 3 elements: a) the average percentage of the popular vote won by PR]

gubernatorial candidates; b) the average percentage of the seats in the local congresses held by the PRI; and
c) the percentage of all terms for governor and local congress in which the PRI had control. For each state,
all 3 percentages were averaged together and carried to four decimal places.

The index has a possible range of 0 (total opposition success) to 1 (total PRI success), with .5 representing
absolutely two-party competition. The resulting clusters suggest the following categories (Ranney 60):
.8500 or higher: one-party PR1

.6500 to .8499: modified one-party PRI

.3500 to .6499: two-party

.1500 to .3499: modified one-party opposition

.0000 to .1499: one-party opposition

61

Table 3.2 Ranney Index in Mexican States, 1990-2004

 

 

Modified One-Party PRI Two-Party Modified One-Party Opposition
Campeche Aguascalientes Baja California
0.7091 0.5475 0.2654
Coahuila Baja California Sur
0.7793 0.4956
Colima Chihuahua
0.7273 0.5201
Chiapas Guanajuato
0.8284 0.4488
Durango Jalisco
0.6745 0.4501
Guerrero Estado de Mexico
0.7901 0.6163
Hidalgo Michoacan
0.8151 0.5551
Nayarit Morelos
0.6695 0.6275
Oaxaca Nuevo Leén
0.7572 0.5091
Puebla Querétaro
0.7954 0.5563
Quintana Roo Tlaxcala
0.8176 0.5991
San Luis Potosi Zacatecas
0.7847 0.5555
Sinaloa
0.7143
Sonora
0.7181
Tabasco
0.7758
Tamaulipas
0.7678
Veracruz
0.7445
Yucatan
0.6424

Sources: CIDAC: http://www.cidac.org.mx/espanioLpolitica 2000.htm

62

Table 3.3 Holbrook- Van Dunk Index” and Folded Ranney Index” in Mexican

 

 

States
State Municipal-Level Competition Folded Ranney
I 970-1 989 1990-2002 1 970-1 989 I 990-2002

Aguascalientes 24.88 55.38 0.6293 0.9525
Baja California 49.99 84.34 0.6082 0.7654
Baja California Sur 19.43 72.67 0.5948 0.9956
Campeche 4.01 45.26 0.5263 0.7091
Coahuila 21.79 55.13 0.5811 0.7207
Colima 44.23 59.18 0.5734 0.7727
Chiapas 7.98 63.54 0.5657 0.8284
Chihuahua 47.19 73 .37 0.6064 0.9799
Durango 27.24 70.56 0.6102 0.8255
Guanajuato 29.72 79.46 0.6461 0.9488
Guerrero 10.42 57.53 0.5405 0.7099
Hidalgo 3.00 37.56 0.5374 0.6849
Jalisco 33.92 72.65 0.6455 0.9501
México 18.02 59.58 0.5954 0.8837
Michoacan 12.02 83.34 0.5646 0.9449
Morelos 9.7 57.87 0.5681 0.8725
Nayarit 15.14 42.04 0.61 17 0.8305
Nuevo Leén 18.52 69.75 0.6518 0.9909
Oaxaca 10.23 46.38 0.5897 0.7428
Puebla 18.00 52.29 0.5921 0.7046
Querétaro 20.97 56.41 0.5451 0.9437
Quintana Roo 4.61 45.53 0.5683 0.6824
San Luis Potosi 22.26 61.5 0.5566 0.7153
Sinaloa 25.56 61.97 0.6027 0.7857
Sonora 29.27 69.29 0.6243 0.7821
Tabasco 8.33 58.69 0.5332 0.7242
Tamaulipas 24.07 60.89 0.6014 0.7322
Tlaxcala 14.46 58.98 0.5946 0.9009
Veracruz 13.45 59.46 0.5717 0.7555
Yucatan 15.8 70.3 0.5933 0.8576
Zacatecas 10.31 66.6 0.5436 0.9445

 

Source: CIDAC: http://www.cidac.org.mx/espaniol nolitica 2000.htm

 

 

29 The Holbrook-Van Dunk formula for statewide values of the district-level measure is:

100 — [(average %vote for winners + average margin of victory + %uncontested seats + %safe seats)/4]. A
safe seat has a winning percentage of 55% or more. 1 used municipal council elections data. Since there are
no uncontested seats in Mexico, I used only the 3 other elements of the formula, dividing by 3. As the scale
increases from 0, it indicates greater degree of competition.

3° The formula for the folded scale is 1 — (Ranney - .5). The resulting index ranges from .5 (no competition)
to 1 (perfect competition).

63

Table 3.4 Experience under Non-PRI Rule, Mexican States, 1985-2004

 

 

Year of

Altemation, Year of Altemation,
State Governor Local Legislature
A68 1998 1995
BC 1989 1989
BCS 1999 1999
CAMP
COAH
COL
CHIS 2000
CHIH 1992, 1998 1992, 1995
DGO
GTO 1995 1997
GRO
HGO
JAL 1995 1995, 2003
MEX 2000, 2003
MICH 2001 2001
MOR 2000 2000
NAY 1999 1999, 2002
NL 1997, 2003 1997, 2003
OAX
PUE
QRO 1997 1997
QROO
SLP 2003 2003
SIN
SON 2000, 2003
TAB 2003
TAM
TLAX 1998, 2004
VER
YUC 2001 2001, 2004
ZAC 1998 2001
DP 1997 1997, 2000, 2003
Source: IFE

64

Figures 3.4 Folded Ranney Index, Mexican States, I970-2004

1

 

 

1

 

0.9 -
0.8 ~
0.7 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1

 

0
g,
Y9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$0

$090300?

V
“Co

I

I

.9965“ 90° 0&0 04° {9° 39’ 3*‘963‘
90
e

 

 

I 1 970-1 989
El 1990-2004

 

 

 

 

0.9

 

0.8 -—
0.7 -—
0.6 -—
0.5 --
0.4 T
0.3 --
0.2 -~
0.1 -

 

 

0

1 r

.—

_l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
Til—T

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l

I

I

Q- C (a

l

 

I 1 970—1989
Cl 1990-2004

 

Source: Table 4.

65

 

 

Figure 3.5 Municipal-Level Competition (Holbrook- Van Dunk Index), Mexican States, 1 970-2004

 

 

 

I 1970-1989
I 1990—2004

 

 

 

 

-_-_ ‘WW
——- 2...... W
—— WW'
- :.-'-...~..---.. . . n: . .
1_t--- WW“ ~
-h"':uu. I~I.TT‘"‘“‘W-—Wn .xmumrv‘". ‘-

. WWW

—W
} —l__—
--___

"’ ‘wmi‘ims-*

’m'“ "9‘

 

 

 

A WV'MW.‘?I.VMWV J. z ,

 

 

90

 

80

I I l I I I I m
Q Q Q G 6 Q Q C
l‘ \O m V M N c-

e a {a o o ~2- 4 o s e <1- c

 

 

66

Source: Table 4.

Figure 3.6 Number of Seats in Local Congresses, 1974-2004

 

 

 

 

 

 

1200

1000

800 i—o—TotalSeafts 7
Fan—Relative Ma'on't '

600 t 1 Y1
' 1

400 ‘-A-Proportional
. _ Renegades): '

200

0 ,
1974 1989 1998 2001 2004

 

 

 

Sources: Lujambio (2000: 55); CIDAC:www.cidac.org.mx/dat(Ls[congresos

Figure 3. 7 Distribution of Seats in Local Congresses mercent), 1974-2004

 

100

 

 

90
80
70
604
502
40.
30-.
20-

10
0

if, Opposition’

 

 

 

 

1974 1989 1998 2001 2004

 

 

 

Sources: Lujambio (2000: 57); and CIDAC: www.cidac.org.mx/d_atos/congresos

67

Figure 3.8 Local Congresses where one party has a majority” of seats, 1978-2004

 

35

 

30-

25-

20-

15‘

10-

5..

 

 

 

0 I I I r T I T U I I I I I U I

«egagexwahebneeesgu
eeeessssssesssss

W

 

 

 

Source: Lujambio (2000); CIDAC http://www.cidac.org.mx

Figure 3.9 Electoral Victories in MR Districts by Parties Other than PRI, 1974-2000

 

70

 

 

 

 

 

I Other Parties
I PAN

 

Percent

 

 

 

 

 

 

 

 

1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000

 

 

 

Source: Lujambio (2000)

 

3 ' Two-thirds of total seats.

68

 

he

M

1111

pr

81

 

 

 

Chapter 4. Aggregate-Level Analysis

4.1 Introduction

As seen earlier, Mexico’s protracted democratic transition started at the
subnational level, however, at very different paces across states. While some state
governments have experienced a relatively significant history of democratic transition,
other states have not had any alter nation experience. Scholars have referred to the
dangers that such “subauthoritarian authoritarian enclaves” (Cornelius 199923; Diamond
1999) pose for the democratic quality of the nation as a whole. Without doubt, Mexican
states vary in their degree of democratization (Hiskey 2005; Hiskey and Bowler 2005;
Solt 2001, 2003), and some have certainly not developed fully in democratic terms yet.

The present chapter explores the effects of macro-level data on support for the
incumbent party in gubernatorial and local legislature elections in Mexico. The analysis
is based on an aggregate-level data set collected for all 31 Mexican states and the Federal
District covering the period between 1970 and 2004. The main observation period,
however, is between 1985 and 2004, the period during which subnational politics in
Mexico became competitive. During this period of time, as seen in chapter 3, Mexico
underwent a series of economic and political reforms that put the democratic transition
process into motion in a number of states, and eventually ended the PRI’s hegemonic rule
at the national level.

By looking at the effects of political and economic decentralization on voting
behavior, the purpose of this chapter is twofold: First, it seeks to find out whether the

straightforward relationship between economic performance and electoral results found

69

in the American and Western European contexts holds at the subnational level in Mexico.
This goal represents the first step within the broader picture of the dissertation. The
second purpose is to add a measurement (effects of decentralization) to the classic
economic voting model. The overall goal is to see whether or not economic voting is
taking place in the Mexican states, and measure the impact of economic and political
decentralization on electoral outcomes.

The results of this chapter support the leading argument of this dissertation in five
principal ways. First, the classic economic voting hypothesis is supported by the results
of the gubernatorial elections models. The results show that a state’s economic
performance (measured by a state’s annual Gross State Product or GSP growth) has
explanatory power when voters make an electoral decision. Second, financial
decentralization—as measured by a state’s locally-generated revenues—has significant
explanatory power on incumbent electoral support in gubernatorial elections. That is, a
state’s level of financial decentralization (measured by its financial autonomy) vis-a-vis
the federal government is positively related with electoral support for the incumbent.
Third, decentralization of political power—measured by a state’s experience under a non-
PRI administration—appears to be an important indicator of behavior at the polls.
Pluralism and electoral competition—assumed to exist in contexts where the former
hegemonic party is no longer dominant—appear to produce those institutional elements
that lead to economic voting. In line with this argument and fourth, the results suggest
that voters distinguish between their state’s political institutions and those at the national
level. This is very significant because it points to some degree of functioning institutional

clarity in the country. Finally, this chapter finds that when accounting for state-specific

70

factors, the results confirm a high degree of uneven electoral behavior across the Mexican
states. Specifically, the results show that the assumed effects of decentralization do not

reach unifome all states. This issue is further explored in chapter 5.

4. 2 Theoretical Considerations

As reviewed in chapter 2, the classic economic voting argument dos not apply in
the same manner to new democratic settings as it does in old democracies. For voters to
be able to assign correctly responsibility to the respective government branch, a certain
amount of clarity of responsibility is required (Powell and Whitten 1993), which can only
grow under certain political conditions. Such conditions, as I argued in chapter 3, did not
develop during the one-party rule in Mexico between 19291and the 19805. Democratic
developments in the last twenty years reached the Mexican states on an uneven basis.
Therefore, it is expected that some states have higher probability of meeting the
conditions necessary for voters to be able to reward or punish their state government,
while citizens in other states do not. In this analysis, economic voters are broadly defined
as pocketbook or sociotropic voters. Both depend ultimately on how well the state is
performing in the economy. Thus, if a state is performing well, its economic voters will
support the incumbent party at the polls.

What explains variation in electoral choice and therefore, in levels of
accountability across the Mexican states? Given the broad variation across the states on a
wide number of factors, the hypotheses explored in this chapter are based on the

following three assumptions.

71

a) Regional Economic Performance

Mexico is a nation rich in regional contrasts, and variations in economic
development across its regions explains to a great extent this heterogeneity. States differ
greatly according to a large number of factors, such as the ratio of its population working
in the agricultural sector, ethnic composition, and overall marginalization levels. Higher
rates of economic growth have been found to be correlated with higher levels of electoral
competition, and therefore, with more effective government.

b) Regional Autonomy vis-a-vis the Center

As seen in chapter 3, Mexico’s states grew increasingly dependent on Mexico
City for resources and political instructions. As noted in a study exploring the nature of
regions in Mexico, and particularly the capacity of the central government to exercise
power at the local level: “The more dependent a locality becomes on outside inputs and a
more sophisticated economic infrastructure, the more dependent it is likely to become on
external political centers in order to get things done, such as roads built or irrigation
installed.” (Roberts 19922236).

Economically, state governments control in average less than a quarter of the
revenues of the federation, in contrast to the three-quarters controlled by the national
government. However, states greatly differ in their levels of financial autonomy vis-a-vis
the federal government. For instance, Tables 4.1 a and d present the states that, according
to these indicators, are the most autonomous vis-a-vis the federal government: the
Federal District, Nuevo Leon, Chihuahua, Baja California, and Jalisco. Conversely,
Tables 4.1 b and c contain the states that depend the most from federal revenues, and

those that produce the least amount of own revenues from taxes. Among the states that

72

appear to be the least financially autonomous are Chiapas, Guerrero, Nayarit, Oaxaca,
and Tabasco. Recent studies on local politics have shown that opposition parties in power
have tended to demand more autonomy from the center. For example, in a study
evaluating Chihuahua’s first government of alternation (1992-1998), Aziz Nassif (2000)
finds that the PAN sought to distinguish itself from the PRI by an overall reorganization
of financial matters, trying to bring about transparency and increasing the state’s own
revenues.32 It can be argued that the motivation behind this is that an opposition
government in Mexico, once it finally wins the needed electoral support to gain office, it
wants to retain that support at home and for this, money is needed. Therefore, opposition
governments have tended to be more efficient in their public spending (Ward 1995).
Some scholars have found that democratization. leads to decentralization
processes. For instance, Beer (2004) finds that “as electoral competition has taken root at
the state and municipal level in many areas across Mexico, democratically elected
mayors and governors have begun to demand more autonomy from the central
government and more control over local fiscal policy.” According to this view, it is
expected that increased subnational competition generates pressures for fuller political
and fiscal decentralization. However, just as political power underwent a gradual (and
uneven) process of liberalization, decentralization, with the aim to “centralize through
decentralization” under the PRI (Rodriguez 1997), took also an uneven and slow shape

across the states.

 

’2 In fact, the panista administration (1992-1998) in Chihuahua managed to duplicate its own revenues
(Aziz 2000255). However, it has been argued that the government was blamed for raising taxes, and that
this contributed to the PRI’s recovery of the govemorship in 1998.

73

c) Local Political Environment

Mexican states greatly differ in their political environment, which is determined
by various factors. First, as I showed in chapter 3, states differ in their levels of electoral
competition. States where the opposition is the weakest or almost non-existent, have been
found to be fertile ground for impunity and continuing authoritarian control under the
PRI (Cornelius 1999211). In fact, some states achieved truly competitive elections33 by
the mid-19905, while others have not done so yet. Second, and related to this last point, a
state’s political environment differs from another, if it has experienced alternation of
power. The alternation of political parties in office is considered to be crucial for the
process of democratization. In a strict sense, those that have not experienced alternation
are still considered authoritarian regimes (Przeworski, Alvarez, Cheibub, and Limongi
2000). As I have argued before, if a state experiences alternation of power, the
government system gains in legitimacy both within parties and citizens. As Diamond has
argued: “Opposition victory at the subnational level also provides a good test of the
party’s platform and its performance in government. A party that is out of power can
easily criticize the governing party and make untenable claims about what it would
achieve if it were in office. However, if the opposition holds power at the local level,
voters have an opportunity to assess its performance” (Diamond 19992131). For instance,
Rodriguez and Ward (1994; Guillén Lopez 1995; Espinoza Valle 1999), looking at Baja
California, concluded that the alternation in the govemor’s office led to an improved
separation of powers among the three branches of government, to greater municipal

autonomy, and better administered elections. The third and last factor refers to divided

 

3’ I mean here that the incumbent faces a realistic probability of being defeated—or that alternation has
taken place.

74

government. In contrast with the argument found in the literature on the US system,
which frequently refers negatively to the consequences of divided government (gridlock),
it has been argued that divided government in Mexican local legislatures is a positive
phenomenon for state-level democratization. For example, Lujambio (2000267-72) argues
that the open debates and negotiation taking place in divided governments serve indeed
as educative experiences for both political parties and citizens. In sum, observers of
subnational politics in Mexico, have argued that states that have competitive elections,
have experienced an opposition government (and perhaps a divided government) exhibit
different styles of governing that have led to the use of elections as mechanisms of

accountability.

4.3 Hypotheses

Hypothesis 1: GSP Growth Rate
H1 proposes that states that perform better economically will show higher levels of

support for the incumbent party, and vice versa.

This hypothesis follows the classic theory of economic voting, which argues that
if an individual citizen is satisfied with past economic performance, s/he will approve of
the president, but if not, s/he will register disapproval. Extending this individual-level
mechanism to the aggregate-level, it is expected that a similar relationship will take place
between macroeconomic indicators and incumbent support at the polls. It is expected that
citizens will form positive (negative) assessments of their government’s economic
performance, if the economy is doing well (poorly), and consequently support (reject) the

sitting government.

75

Hypothesis 2: Locally-Generated Revenues
H2 proposes that states with greater financial autonomy vis-a-vis the central government
will support more the incumbent party.

This hypothesis is based on the following assumption: a state with more resources
of its own—that is, financially more decentralized—develops more transparent
institutions and thus, its citizens are more likely to reward the incumbent party. Greater
financial autonomy generates elements that increase accountability, such as transparency,
access to information, and/or political knowledge. According to this view, voters living
under such conditions are more likely to support the incumbent party.

Hypothesis 3: Revenues from the Federal Government
H3 proposes that the more dependent a state is on resources from the federal
government, the less support the incumbent party will get in state elections.

Ideally, rational voters will reject the notion of their state being heavily dependent
on the federal government, and support instead efforts designed to develop or increase
their state’s financial autonomy, as autonomy means accountability. Thus, it is expected
that incumbent parties will get less electoral support, the higher their total revenue
depends on transfers from the federal government.

Hypothesis 4: Experience under Non-PRI Rule
H4 proposes that states with experience under a party other than the PR] rule will

support less the incumbent party.

The logic behind this hypothesis is that states which have experienced rule by
parties other than the long-time hegemonic party, are more likely to have voters who

assign substantive content to their voting decisions than those states which have not

76

experienced alternation of power. Experience under a party different than the PRI means
that the state has high levels of electoral competition. As discussed earlier, this kind of
context is likely to have high levels of access to information and a more critical civic
society. Since voters have experienced rule under a different party, they tend to be less
risk-averse, and are therefore more likely to express dissidence with the incumbent party,

if they disapprove of the administration’s performance.

4.4 Data and Model

The central question explored in this chapter is whether greater financial
autonomy and alternation of political power at the state-level lead to economic voting.
Thus, the units of analysis are elections for governor and state legislatures in Mexico’s 31
states (see appendix for a list of the states and corresponding years of elections included
in the data set). A data set was created on a number of fiscal and political indicators, as
well as economic performance variables for subnational elections taking place between
1970 and 2004. The data was drawn primarily from Mexico’s Federal Institute of
Statistics (Instituto Nacional de Estadt'sticas, Geografia, e Informdtica, INEGI), Center
of Research for Development (CIDAC), Banamex, and Mexico’s Federal Electoral
Institute (Instituto Federal Electoral, IFE).
Dependent Variable (Electoral Incumbent Support)

In gubernatorial elections, incumbent support was measured by the percentage of
votes that the incumbent party won in a given election, regardless of whether it
maintained its control or not (see Figure 4.1). 34 In state legislature elections, incumbent

support refers to the percentage of seats that the incumbent party got, also regardless of

 

3" 1n the cases where a party ran in a coalition, I considered the major party.

77

whether it retained a majority or not. Both majority and proportional representation seats
were added and divided by the total number of seats.
Independent Variables
a) State-level Economic Performance

To measure a state’s economic performance, this analysis relies on a state’s
annual gross state product (GSP) growth rate. Growth in state-level GSP is expected to be
significant and positive. The well-known argument followed here is that economic voters
take into account the economic performance of the incumbent administration and will
reward at the polls its positive performance. Unfortunately, no other state-level economic
indicators were available.35
b) Financial Autonomy ’6

In this research, financial autonomy is measured by two dimensions: 37 A state’s
own revenues and federal transfers. First, each state’s locally-generated revenues were
calculated for every year. Own revenues refer to the locally-generated revenue composed
of taxes, fees, patrimonial goods, and benefit fees, which were calculated by adding the
figures of all four components and then obtaining the ratio vis-a-vis the total amount of
revenues for the state. Per capita figures were used in the analysis. Second, each state’s

federal revenues were calculated for every year. Federal revenues refer to the amount of

 

3’ No state-level data for inflation rate were available. Unemployment rates are available only according to
“major urban areas” (INEGI), which makes it nonsuited for a state-level indicator.

’6 It is important to make a few notes on the methodology used to calculate most of these variables. I relied
for the most part on figures published by the INEGI in different volumes of F inanzas Publicas y
Municipales (1970-2004). Since this information was not available in electronic form, I created an
extensive database including the figures for: taxes (impuestos),patrimonial goods (productos), fees
(derechos), benefit fees (aprovechamientos) and participaciones for the 32 states over a period of 34 years.
To make the figures comparable, I first converted all into a single unit (thousands of pesos), and then
applied consumer price index figures to deflate the numbers. I added taxes, patrimonial goods, fees, and
benefit fees in order to get the total of locally generated revenue. To obtain per capita figures, I relied on
Mexico’s National Population Council’s (Consejo Nacional de Poblacio'n, CONAPO) yearly population
information. Finally, 1 calculated an average of these numbers according to each state’s gubernatorial term.
’7 I followed Dial-Cayeros (2004) for the operationalization of subnational financial autonomy.

78

revenue sharing each state receives from the federal government. These
intergovernmental transfers are composed of two figures: Unconditional federal revenue
sharing (participaciones) and earmarked federal revenue (aportaciones), which means
that state governments have little or no discretion over its spending. The share was also
obtained vis-a-vis the total amount of revenues, and per capita figures were used in the
analysis.
0) Local Electoral Environment

To consider the effects of electoral competition and pluralism, states were
classified according to two categories: Those that have ever experienced rule under a
non-PRI party either in the governor’s office or a majority in the local legislature (Table
4.2).
Control Variables
a) Coattail Effects

To account for the effects of increased turnout and party voting in lower level
elections due to concurrent elections for higher office, I include two dummy variables,
Concurrent 1 (in regards to the local legislature) and Concurrent 2 (in relation to
presidential elections), to measure coattail effects.
b) Divided Government

Divided 1 is a dummy variable, taking a value of one when the incumbent
governor’s administration is of a different party than the local legislature’s major party.

Divided 2 refers to divided government in relation to the president’s party.

79

c) Fixed Effects

I used a dummy variable for each state, taking Mexico City as the baseline. With
this, I intend to capture differences in vote outcomes due to state-specific factors.
(1) National-level Economic Indicators

Two variables were included to control for the effects of national-level economic
indicators: Annual national GDP per capita and national level annual inflation rate.

To test the above hypotheses, I estimated a series of regressions using the
following general model:
Incumbent Support, = 130 + [31 GDP Growth + [32 Own Revenues + [33 Federal

Revenues + [34 Non-PRI Experience + [35 Divided 1 + [36
Concurrent 1 + [37 Divided 2 + [33 Concurrent 2 + e

4.5 Analysis and Results

I applied an OLS regression model to test the effects that financial autonomy and
experience under a non-PRI administration have upon support for the incumbent party.
Because of my interest in capturing the effects of both the more recent decentralizing
fiscal efforts and the growth of electoral competition, I considered 1985 as a threshold
year and focused on the elections taking place between that year and 2004.38 Generally,
five gubernatorial elections (six-year terms) and twelve local congresses’ elections (three-

year terms) were included for each state (see Appendix A). Specifically, the data set for

 

’8 1 applied the model using the data in three different time periods. First, 1 calculated an average per
administration. A few cases in the last elections were calculated using only the data available until 2003.
However, as different studies have shown, looking at the entire administration when trying to analyze
voting behavior is not very accurate/reliable. 1 also examined the figures of the election year (Kramer 1971;
Tufie 1978). In addition, I looked at the conditions of the previous year, as well as at the average of these
two years (prior and election years). All three models, however, yielded similar results. I decided to focus
on the one that looks at the year prior to the election, because it allows including observations from the
following 2004 elections.

80

gubernatorial elections includes 112 observations and the local congress elections data set
includes 209 cases. A case is defined as an election.

Table 4.2 presents the results for the regressions exploring gubernatorial elections.
It contains two models; Model 1 is an OLS estimate, while the regression in Model 2
includes a fixed effects variable that controls for unobserved state-specific correlates of
incumbent support. Overall, Model 2 confirms the first model’s robustness, and provides
a better fit for the model. The fixed-effects variable exposes the importance of
considering the uneven impacts of the variables under study across the Mexican states.
There is evidence suggesting that state-level factors related directly to decentralization
are significant in explaining incumbent support.

The classic economic voting argument (H1) is supported by the results of Table
4.2. A state’s economic performance appears to be an important explanatory factor in the
support of the incumbent governor’s party at the polls. Specifically, a percentage point of
grth in a state’s GSP translates into an increased support of 3.9 percent of the
incumbent party. This result is very significant, because it suggests that state-level factors
are included in the voting calculus, pointing specifically to a presence of economic voting
at the subnational level in the Mexican context.

H2 is supported as well. The results show that locally-generated revenues are
positively and significantly correlated with support for the incumbent party. This result
suggests that the incumbent party benefits electorally from higher locally-generated
revenues.The second indicator of state-level financial autonomy, however, federal
revenue, appears significant but in the opposite direction. The results might not be that

surprising, given the traditional use that federal transfers often have. In fact, this variable

81

has been used by other scholars (Remmer and Ge’lineau 2003) as an indicator of the
presence of patron-client relationships between the central government and the periphery.
Patronage, pork barrel, and clientelistic activities can be measured by the amount of
transfers the central government gives to provinces. As discussed in chapter 3, Mexican
presidents under the PRI-regime used extensively clientelistic practices to maintain a
tight electoral control across all governmental levels. This indicator was expected to
reflect voter’s dissatisfaction (in the more decentralized, autonomous states) with their
state being too dependent from the central government. On the other hand, citizens in
heavily dependent and centralized states, regard resources from the federal government as
necessary, and demand more of them for their development. But, they seem to never get
enough of them, and thus, cast less support for the incumbent party at the polls.
Experience under the opposition is significant and in the expected (negative)
direction. This result means that states that have experienced an opposition government
are significantly more likely to cast less support for the incumbent. The support of H4
suggests that indeed, the experience under the opposition creates certain conditions,
which contribute to more critical voters. H4 is strongly supported, and it suggests that a
state that has undergone an alternation of power may have developed a mechanism that
decreases support for the incumbent at the polls. As mentioned earlier, one possibility is
that voters in areas without some form of opposition experience remain risk-averse and,
not knowing what the opposition might look like in power, prefer to keep their familiar
political arrangement intact. In contrast, in states where electoral competition is surely

higher and perceived by voters as legitimate, the incumbent party tends to get a lesser

82

share of the votes. In this context, voters appear to perceive they have the real option to
reward or punish however they wish, which points to accountability.

Among the political control variables, divided government between the governor
and the president (Divided 2), is strongly significant with a negative coefficient. That is,
when a state’s administration is divided, the governor’s incumbent party gets less votes,
which confirms the prevailing explanatory power of partisan identity in Mexican
politics—a concept which is assumed to be included in this variable. Divided government
at the state level (Divided 1) reaches statistical significance in Model 2. In general,
divided government translates frequently into conflicting interests among power factions,
which constantly try to expose each other with the aim to throwing the other out of
power. These dynamics often lead to gridlock. In a new democratic setting, however,
divided government can be regarded as healthy to a certain degree (Lujambio 2000). This
regression result suggests that divided government at the state-level affects incumbent
support, just as divided government vis-a-vis the president does. It is important, because
it points to the visibility of state-level institutions, which can only result from electoral
competition and pluralism.

The results for the analysis of the state legislature elections are presented in Table
4.3. The results do not show support for the hypotheses 1-3, and the prevailing
importance of national-level factors in explaining incumbent support in state legislature’s
elections is very clear. In contrast with the gubernatorial elections analysis, Divided 1 is
significant in both models. A divided state administration means, according to these
results, less support for the incumbent majority party in the state legislature. The

importance of party identification is further supported with Divided 2, which suggests

83

that national-level factors have primordial importance in state congress’ elections.
Experience under a non-PRI rule does show significance with a negative coefficient.

Overall, these results need to be analyzed in further.

4.6 Conclusion

Although the literature on economic voting presents some disagreements on the
exact nature of the relationship between economic performance and public support for the
incumbent, overall, presidents are believed to be punished or rewarded for their economic
performance. This analysis has taken this hypothesis to the subnational level and has
found significant evidence at this level across the Mexican states. The sources for the
variation across the states cannot be discerned from this type of analysis, but based on the
theoretical framework that the study is based on, it is possible to present some
conclusions.

As shown in the preceding analysis, state-level economic performance plays an
important role in incumbent support in Mexican gubernatorial elections. Economic voting
appears to be taking place in gubernatorial elections. A state’s economic performance
entails, as discussed earlier, a wide array of factors that distinguish one state from
another. Generally, states performing better economically and with experience under the
opposition are more likely to gain electoral support for the incumbent government.

A state’s capacity to generate its own revenues is also found to be positively
correlated with the incumbent’s party support in gubernatorial elections. That is, those
state governments that manage to increase their economic autonomy vis-a-vis the center

are rewarded with more approval at the polls. This is an important finding, because a

84

state’s financial autonomy is a crucial indicator of the effectiveness of decentralizing
policies. The general argument of this dissertation is supported hereby.

Another important finding refers to the significance of the variable controlling for
a state’s experience under a non-PRI administration. As discussed before, my goal was to
capture the effects that an opposition party in power has on the institutional environment
in a given country. Based on the assumption that such experience under the “opposition”
brings about greater clarity of responsibility and thus, voters are likely to become
economic voters, the findings in this chapter suggest that indeed the strength of the
opposition within a state and its actual presence in government is strongly related with
the way the state’s population votes.

Clearly, there is a significant degree of heterogeneity within the group of Mexican
states that cannot be fully accounted for by the other explanatory variables in the
statistical model presented here. The next chapter looks with more detail into citizen’s

motivations and state-specific factors.

85

Table 4.1 Diverse Indicators of Financial Autonomy in Mexican States, 2004

Table 4.1a States receiving highest amount of revenues from taxes (% of GDP), 2004

Table 4.1b States receiving lowest amount of revenues from taxes (% of GDP), 2004

Table 4. 1c States receiving highest amount of federal revenues (% of GDP), 2004

Table 4.1d States receiving lowest amount of federal revenues (% of GDP), 2004

Source: http://www.cidac.org.mx

 

 

 

 

 

 

DF 61.44%
TAM 32.20%
COL 31.22%

NL 16.06%
VER l 1.12%

 

 

 

 

 

 

 

 

 

 

 

CHIS 2.42%
DGO 2.77%
OAX 3.07%
GRO 3.08%
NAY 3.38%
TAB 3.88%
CAMP 3.99%

 

 

 

 

 

 

 

 

 

 

CHIS 21.86%
TAB 21.73%
OAX 18.99%
ZAC 16.99%
NAY 16.53%
GRO 15.90%
TLAX 15.52%

 

 

 

 

 

 

 

 

 

 

DF 2.53%
NL 4.25%
CHIH 5.34%
BC 5.91%
QROO 5.92%
VER 6.47%
COAH 6.60%
JAL 6.66%

 

 

 

 

 

 

Figure 4.1 Total Votes for Incumbent Party, Gubernatorial Elections, 1985-2004

Percentage

 

 

 

Years

 

Source: http2//www.cidac.org

87

Table 4.2 Gubernatorial Elections across Mexican States, 1985-2004

Incumbent Support (% votes) OLS Coefficient (S.E.)

 

 

Variable Model 1 Model 2 (Fixed Effects)
GSP Growth Rate 3.553" 3.949"
(1.786) (1.935)
Own Revenues 36.172" 70.967***
(14.200) (19.924)
Federal Revenues 24.075*** 29.204***
(7.339) (9.387)
Non-PRI Experience -7.638*** -47.472**
(2.842) (20.976)
Concurrent 1 -0.472 -0.480
(3.576) (4.844) .
Divided 1 -7.859 44.745"
(5.639) (7.216)
Concurrent 2 -4.1 15 1 1.080
(4.548) (9.504)
Divided 2 -14.640*** -12.090***
(3.320) (3.729)
Constant 51.242*** 52.912***
(6.409) (1 1.998)
N 112 112
Adjusted R-squared 0.4057 0.4403

 

Notes: Standard errors are listed in parentheses.

* = statistically significant at .1, ** = statistically significant at .05, and *** = statistically

significant at .01 or better

Table 4.3 Local Legislature Elections across Mexican States, 1985-2004

Incumbent Support (%seats) OLS Coefficient (S.E.)

 

 

Variable Model 1 Model 2 (Fixed Effects)
GSP Growth Rate -13.115*** -15.192***
(3.273) (3.355)
Own Revenues 1.474 10.127
(6-891) (7.609)
Federal Revenues 18.509*** 14.650" *
(3.999) (4.667)
Non-PRI Experience -6.376*** -30.817***
(L612) (8.206)
Concurrent l -l.301 -0.258
(1.625) (1.708)
Divided l -l3.020*** -l3.875***
(3.143) (3.274)
Concurrent 2 -0.961 2.304
(2.283) (2.701)
Divided 2 40.153444: 94304:“
(2247) (2.303)
Constant 57.825*** 67.658***
(2.699) (7.312)
N 209 209
Adj. R-squared 0.4134 0.4453

 

Notes: Standard errors are listed in parentheses.

"‘ = statistically significant at .1, ** = statistically significant at .05, and *** = statistically

significant at .01 or better

Chapter 5. Individual-Level Analysis

5.1 Introduction

The findings of last chapter confirmed that, based on aggregate-level data, the
classic economic voting hypothesis holds at the state level in Mexico. That is,
macroeconomic indicators are linked to some degree to incumbent support in state-level
elections across the Mexican states. Furthermore, the results show that even though the
economic voting hypothesis holds across the Mexican states, it does so in differing
degrees of intensity among the different subnational contexts. However, due to its nature,
an aggregate-level model does not permit us to explore further this apparent variability in
the strength of economic voting. In this regard, the model presents two main weaknesses.
First, it assumes that the political and economic contexts across states are the same.
Second, the model assumes that citizens across the states respond (or not) in the same
way to their economic and political contexts when voting for their subnational leaders. In
short, this macro-level model treats both context and voters as homogeneous, and does
not distinguish variations between states. So, the question remains, what accounts for this
variability in economic voting across the states? Under what conditions are voters more
likely to use elections as an act of accountability? How rational is the Mexican voter after
all?

The analysis of this chapter makes use of pre-electoral survey data collected by
the newspaper group Reforma in the Mexican states of Chiapas, Chihuahua, Jalisco,
Michoacan, Nuevo Leon, and Querétaro. These states make an excellent sample because
they reflect the variation across states in three main ways: geographic location, economic

performance, and political experiences (Table 5.1 presents the main characteristics of this

90

sample of states). Thus, the purpose of this chapter is to expand last chapter’s findings, by
switching to the micro-level analysis in order to examine more closely both the different
contexts among states and individual motivations that lie behind economic voting. The
advantages of such a model are numerous, including the possibility to account for
multiple intervening factors such as political and socioeconomic cleavages, which are
considered to be traditionally very strong in new democratic regimes.

Overall, the findings of this chapter show support for the hypotheses tested here.
As expected, those states with better economic performance and direct experience under
opposition governments (Jalisco and Nuevo Leon) show the clearest support for the
theory. Citizens in Chihuahua and Querétaro, both states with experience under panista
administrations, show clear support of the theory as well. More importantly, the results of
this analysis confirm that the institutional capacity of subnational government differ

among states, and it has strong effects on citizens’ likelihood to be economic voters.

5.2 Theoretical Considerations

In the analysis of chapter 4, I tried to explain the presence of economic voting
through diverse variables, such as a state’s economic performance, a state’s level of
economic decentralization, and a state’s local political environment. When analyzing
gubernatorial elections, I found support for all three claims, which in short point to the
conclusion that “institutions matter.” Yet, a crucial question remains: Are the findings so
far accidental or have voters learned to use elections as an act of accountability? As
reviewed in chapter 3, Mexican voters were constrained by an overwhelming presidential

system, by powerless and, for the most part, subservient to the federal government

91

 

subnational leaders, and by elections that were mostly just an act of formality and
symbolism. I have argued that the decentralization of political and economic powers have
turned elections into an act of accountability. But, what about voters? Have they learned
to make well-informed decisions when deciding whom to vote for? 15 perhaps their
learning a product of their local context? The argument presented here is that they do,
however, this propensity varies according to their state’s level of political and economic
powers. That is, a state that has experienced alternation of power, divided government,
and is able to economically detach itself —at least to some extent— from the federal
government, is more likely to have a more informed citizenry that makes use of elections
in a new way.

When do performance evaluations become significant to citizens? So far, I have
argued that adequate decentralization of power strengthens subnational institutions in
such a way that elections take on a different meaning: they become an instrument of
accountability for voters. So, the next question refers to once leaders institutionalize
accountability in the act of elections, then what triggers voters to take into account
performance evaluations?

Following Diamond (1999277-93), legitimacy and regime performance are closely
intertwined, particularly in the early stages of the democratization process. As democracy
deepens, the legitimacy of the regime tends to depend less on performance evaluations
and more on citizens’ “default” commitment to democracy. This is what Diamond refers
to as “intrinsic, diffuse, or unconditional legitimacy.” As explained by Bratton, Mattes,
and Gyimah-Boadi (2005242), “performance evaluations cover at least two baskets of

public goods. In one basket are economic goods, like jobs, incomes, assets (for example,

92

 

land), consumer products, and an array of social services; in the other basket are political
goods including political order, civil liberties, electoral rights, human dignity, and
equality before the law.” Applying Diamond’s argument to the concept of political and
economic goods, the logic followed is that citizens in new democracies expect both
political and economic goods at once, and once they develop an unconditional
commitment to political goods, then they will support the system, and the system’s
economic performance will not put this support at risk.

If I am correct in arguing that economic voting is largely a function of
decentralization of political and economic powers, then it is expected that citizens living
in states that have already provided them with the required political goods (or
commitment to democracy) will perhaps feel more freely to express their preferences at
the voting polls, because they see no threat in doing 50. Conversely, citizens living in
states where it is still unclear to what level political elites are committed to democracy,
citizens will be less likely to use elections as a tool of accountability.

Next, I present a number of assumptions based on some specific characteristics of
the states contained in the sample. On the one hand, Chiapas is a southern state, and
perhaps best known to international observers for the indigenous uprising led by the
Zapatista Army of National Liberation (EZLN), which started in 1994.39 During
Mexico’s recent democratic transition process, Chiapas has been considered (along with
Puebla, Tabasco, Guerrero, Oaxaca, and Campeche) to be fertile ground for the “survival
and even strengthening of subnational authoritarian enclaves” (Cornelius 1999). With a

population largely rural (57% of its active population working in agriculture, INEGI

 

’9 See Harvey (1999) for an analysis of the Zapatista Movement within Mexico’s democratic transition
process.

93

2001) and composed by 24.7% of indigenous population (INEGI 2001), Chiapas is
among the poorest states in the country. The state’s socioeconomic characteristics made it
a traditional PRI stronghold, which developed a large base of support among its
agricultural workers. In general, the dire economic circumstances of this sector makes it
particularly vulnerable to authoritarian practices and impunity exercised to maintain
electoral control. In this context, struggling citizens are assumed to be particularly
susceptible to clientelist and vote buying practices.

Chihuahua, just like Nuevo Leon, borders to the North with the US. The border
itself presents unique characteristics compared to the rest of the nation. In general terms,
the natural and intensive interactions taking place between both sides of the border create
different environments. Politically, Chihuahua is a state with a comparatively long
history of experience with the opposition.40 In 1992 it became the third state (after Baja
California and Guanajuato) to experience alternation of power when the PAN won the
governor’s office. Furthermore, in the next elections in 1998, the PRI managed to win
back the govemorship. Thus, citizens living in Chihuahua are among the minority of
Mexican citizens who have experienced two alternations of power (along with Nuevo
Leon and Tlaxcala). Citizens in this state, according to the theoretical framework that is
being tested here, are expected to be aware that different parties in power offer different
styles of governing, and that elections are the mechanism through which they can express
which party they prefer to have in office.

The third state, Jalisco, is characterized by a large population (7% of total) that is
disproportionately concentrated in Guadalajara — the third largest city in the nation (after

Mexico City and Monterrey). Jalisco is largely an industrial state. It is characterized

 

’0 See Aziz (2000) for an in-depth account of Chihuahua’s recent political history.

94

along with Nuevo Leén, by its early progress in developing an independent press, which
is expected to increase citizen’s awareness and information of public affairs. As seen
earlier, there is a theoretical consensus on the need of an informed citizenship to improve
the quality of a democracy. In addition to formal education, which is the single most
important factor in improving cognitive awareness (Bratton, Mattes, and Gyimah-Boadi
2005:40-41), a free media contributes to the increase of people’s awareness and
knowledge of public affairs. In 1995 the PAN won the govemorship, and it has kept it
under control until now.

Michoacan is politically best identified by being the PRD’s stronghold. Founder
and long-time leader of the PRD, Cuauhtémoc Cardenas, is a native from Michoacan, as
was his father and former Mexican President Lazaro Cardenas—who nationalized the oil
industry in 1936 and is perhaps the most revered president in Mexican history. Before
Cardenas split from the PRI in 1987, he had been Michoacan’s governor, so when he
formed his own party (PRD in May of 1989), he could count on an already-built base of
support in his native state. After a long struggle spanning over ten years, the PRD won its
first govemorship in 2000.

Nuevo Leon is the second border state in the sample. Two characteristics make
this state stand out from the rest: a large, politically active business sector, and a strong,
serious, and independent media. On the one hand, Nuevo Leon developed largely as an
industrial state with the city of Monterrey as its economic center. As a result of the
economic crisis the nation went through during the 19805, a group composed primarily of
small- and medium-size entrepreneurs joined forces with the PAN to fight politically

against the PRI (Mizrahi 2003; Chand 2001). The PAN won eventually in 1997, and

95

experienced in 2003 again an alternation when the PR1 won the govemorship back. The
other trait that characterizes the state of Nuevo Leon is its long history of an independent
media. The newspaper group Reforma started here in the 19405 with the newspaper El
Norte.

Finally, the state of Querétaro has become one of PAN’s steadiest electoral
strongholds. It is composed by a population largely conservative, where the Catholic
Church counts with a strong presence. Since 1997, the PAN won control of both the
govemorship and the majority of seats in the local congress. It has kept tight control over

both since then.

5.3 Hypotheses

The next set of hypotheses is divided according to the two main dependent
variables, namely, approval of governor and vote for the incumbent governor’s party.41
Hypothesis 1: Sociotropic Evaluations
HIA proposes that individual evaluations of the governor vary positively with
evaluations of the state '5 economy. H] B proposes that electoral support for the
incumbent governor ’5 party will vary positively with individual evaluations of the state ’5
economy.
Hypothesis 2: Pocketbook Evaluations
H2A proposes that individual evaluations of the governor vary positively with the

respondent’s evaluation of his/her own economy. HZB proposes that electoral support for

 

41 . . . . . . . . .
Vote for the incumbent majorrty party 1n local congress electrons 15 not mentioned, it 15 however
assumed to be included in the hypotheses related to the electoral support for governor.

96

the incumbent governor ’s party will vary positively with the respondent ’s evaluation of
his/her own economy.

Hypothesis 1 and 2 deal with the classic economic voting variables of sociotropic
and pocketbook evaluations. The expectation is that when individuals are satisfied with
the economic performance of their state, they will support the sitting government, and
when they are disappointed with the state’s economic performance, they will punish the
incumbent’s party.

Hypothesis 3: In-Party

Since the objective here is to find out where, and to what extent Mexican voters
are economic voters, this hypothesis rests on a counter-argument of the referendum-
voting model, which builds on the premise that “the responsibility of managing the state
economy is laid at the feet of the governor” (Atkeson and Partin 19952 100). The
fundamental logic of the referendum model implies that individual voters send messages
to the central government when evaluating the performance of subnational leaders.
However, economic voters are expected to be able to distinguish between layers of
government when they assess their government’s economic performance. That is, when
economic voters are casting a vote at the state level, they are supposed to be evaluating
their governor’s administration performance, not their president’s. The approach
defended here assumes that voters recognize the independent policy-making role played
by governors.

H3A proposes that voters living in in-party states will show less gubernatorial approval.
Along the same line, H3B proposes that the probability to vote for the incumbent party

decreases if the incumbent governor’s party is the same as the president’s party.

97

Hypothesis 4: Fiscal Autonomy

H4A expects gubernatorial approval to be higher in the more decentralized states.
Accordingly, H48 expects that electoral support for the incumbent governor ’s party will
be higher the more decentralized a state is.

Within a country, economic evaluations are expected to be significant in varying
degrees, according to the different levels of clarity of responsibility that each state
develops (Powell and Whitten 1993). Economic evaluations are expected to be significant
the higher the clarity of responsibility a state has. When political conditions help voters
associate the candidate of the incumbent party with the outgoing government or, to put it
another way, when the context “eases the attribution of responsibility,” economic
assessments should have a greater impact on incumbent vote choice. The role governors
play in state and national economic management combined to their increasing level of
public visibility suggests that governors should be held accountable for state economic
performance, especially in the more decentralized (fiscally autonomous) states
(Chihuahua, Jalisco, Nuevo Leon, and Querétaro). Conversely, the less decentralized

states (Chiapas and Michoacan) are expected to exhibit the reverse behavior.

5.4 Data and Model

The hypotheses are explored on the basis of survey data collected by the
newspaper Reforma in six states prior to gubernatorial elections between 2000 and 2004.
Each survey was conducted on a representative sample of the population of each state
through face-to-face interviews. The sample size of the six surveys ranges from 840 to

1205 respondents (Table 5.1). The assembled data set offers enough variation in

98

economic and political conditions across subnational units to provide a robust test of the
hypotheses. As shown in Table 5.1, subnational economic performance varies from one
state to another. The states included in the analysis represent about 24 percent of the
country’s total population and 24 percent of its GDP. It is clear that regional economic
performance differs from the overall national performance, as well as across states. In
addition, the data set provides two cases where the incumbent governor does not belong
to the president’s party (Michoacan, Chihuahua) at the time of the survey. This is
especially important to test the effects of in-party.

In addition to variations across the subnational units selected for the analysis,
there are important differences with respect to the six states. As observed earlier, Mexico
initiated its process of political and economic decentralization. during the 19805 and more
so during the 19905. However, the extent to which these processes in each individual
state have led to the effective decentralization of the federal regime varies from one state
to another, especially in terms of financial decentralization. Figure 5.1 shows the varying
degrees of subnational capacity to generate own revenues. Within an average taken over
2000 and 2002, Chihuahua is clearly the state with the highest capacity to produce its
own revenues, with subnational government revenues accounting for almost 20 percent of
the state’s total revenues. Comparing this sample with all states and excluding Mexico
City, Chihuahua is still the most financially autonomous (national mean=l 1.5)
Dependent Variables ’2
a) Gubernatorial Approval

This variable measures individual approval of the incumbent governor, as

expressed by the respondent when asked to evaluate in retrospective the incumbent

 

’2 The complete list of the questions used and their recoding is included in the Appendix B.

99

governor’s job. The item could take four values: disapproves completely, disapproves
some, approves some, and approves completely.43
b) Incumbent Support in Gubernatorial Elections

This is a binary variable that reflects individual vote choice in gubernatorial
elections. It takes a value of one when the respondent voted for the incumbent governor’s
party, zero otherwise. In all six states, the question used in the analysis asked respondents
to cast their vote for governor in a ballot box that provided by the interviewer.
c) Incumbent Support in State Legislature’s Elections

Vote choice in local legislature’s elections, takes a value of one when the
individual voted for the incumbent local congress’ majority party, zero otherwise.
Chiapas is not included in this model, because it only held elections for governor in 2000.
Independent Variables of Main Interest
a) Sociotropic and Pocketbook Evaluations

The independent variables of most interest included in the model consist of two
individual retrospective44 economic assessments: the respondent’s evaluation of the
economy in his or her state, as well as his or her own economic situation at the end of the
incumbent administration. Fortunately, the surveys included for the most part the same
questions across states.45 The sociotropic item was measured with the question asking
respondents to assess the performance of their state economy. This question had three

ossible answers: “im roved,” “remained the same,” or “worsened.” The ocketbook
P

 

‘3 In the case of Chiapas, the choices were three: good government, indifferent, and bad government. In
order to get four categories, the “Don’t know” and “Did not respond” responses were recoded into
“Disapprove some.”

4‘ Unfortunately, only the most recent survey, Chihuahua, included a question on prospective economic
evaluation.

’5 In Chihuahua, respondents were asked to evaluate the overall condition of their state under the incumbent
governor, rather than specifically the economy of the state.

100

evaluation was measured with the item asking respondents about their past personal
situation, having the same three response choices as the sociotropic item.
b) Local Electoral Environment

In order to explain to some extent the variation across states, a state-level variable
was included in the model. This dummy variable takes a value of one when the
incumbent governor’s party is the same as the president’s party in turn, zero otherwise. It
is used in interactions with the sociotropic and pocketbook variables to compare the
impact of these variables on incumbent support in both in- and out-party states.
Control Variables
a) Partisan Identity

The model includes a control variable for party preferences. This variable consists
of the respondent’s self-declared party identification or preference. In all four states, the
original question asked “Generally speaking, do you consider yourself priista, panista, or
perredista?,” which refers to each of the three major parties (PRI, PAN, and PRD). The
answers included “very and some priista,” “very and some panista,” or “very and some
perredista,” and they were recoded into a dummy variable taking the value one if the
respondent claimed to identify with the incumbent governor’s political party.
Traditionally, partisanship has been the most powerful predictor of vote choice in the
Mexican electorate. It is assumed that the concept includes the effects of presidential
approval.
b) Political Knowledge and Interest

Political knowledge is represented in this analysis by two kinds of questions.

First, the respondent’s level of education, which was recoded into three categories

101

identifying individuals with “primary level completed,” “secondary/technical level
completed,” and “some university education or better.” The secont indicator of political
knowledge was taken from a question asking the respondent whether s/he knows when
the next elections are taking place. Interest was taken from a question asking respondents
how interested they are in the political campaigns going on for the next election.
c) Locality and Income

Finally, two dummy variables are included to control for the socioeconomic status
of the respondents. Locality took a value of one when the respondent lives in an urban or
mixed area, zero when the respondent lives in a rural area.
Income was used to control for the income level of the respondents; it was recoded intro
three categories identifying individuals earning “0 to 4,000 pesos”, “4,001 to 12,000
pesos”, and “12,001 & more pesos.” It is expected that individuals living in an urban
area, will be more likely to vote for the incumbent party. Income is also expected to have
a positive coefficient.

In order to test the hypotheses the analysis relies on an additive model of
individual governor’s approval and vote choice in gubernatorial and local legislature’s
elections. The general model for vote choice can be resumed as follows:

Gubernatorial Approval 1 = 130 + [31 State Economic Performance + [32 Personal
Economic Performance + 133 Partisan Identity + e,

Incumbent Support,- = 130 +13, State Economic Performance + [32 Personal Economic
Performance + [33 Partisan Identity + e,
5.5 Analysis and Results
The statistical analysis relies on both the ordered probit and probit regression

methods. First, the results are presented separately by state for each of the three

102

dependent variables (Tables 5.1-5.3). Tables 5.4 to 5.6 contain the results of the merged
data sets.

Table 5.2 presents the results for the model explaining gubernatorial approval.
The classic economic voting hypothesis is overall consistently supported across the six
states, although some variation can be observed across the states and between the two
measures of economic voting. First, sociotropic evaluations are clearly the strongest in
Jalisco and Nuevo Leon, even stronger than their respective party id coefficients.
Chiapas, Chihuahua, and Querétaro also strongly support HlA, suggesting that for voters
in these states, their state’s economic performance is quite important when evaluating
their governor’s performance. Michoacan is the only state where this variable is not
significant. Second, pocketbook evaluations (H2A) are statistically significant in all
cases. In Querétaro, this indicator has more explanatory power than the sociotropic one.
In Chiapas, Chihuahua, and Nuevo Leon the coefficient is clearly weaker than the
sociotropic one. Partisan identity is the only variable that is consistently significant across
all states.

Table 5.3 presents the results of the same model for the dependent variable of
incumbent support in gubernatorial elections. The results are clearly more mixed than the
one in the previous table. First, Chihuahua, Jalisco, and Nuevo Le6n present again the
strongest support for the economic voting sociotropic argument. The results in Querétaro
are significant still, but weaker than in the previous cases. Again, the pocketbook
coefficient is stronger than the sociotropic one. The coefficients of Chiapas and

Michoacan do not reach significance in any of the cases. Second, the pocketbook

103

 

 

hypothesis is significant only in the cases of Nuevo Leon and Queretaro. Third, party
identity remains strongly significant across all cases.

Table 5.4 presents the results of the model looking at incumbent support in state
legislature’s elections. Overall, this model shows weaker support for the economic voting
hypothesis. Again, the economic voting hypothesis is the strongest in Jalisco and Nuevo
Leon when observing the results for the sociotropic variable, although comparatively
much weaker than in the other models. Chihuahua and Querétaro are significant as well.
Michoacan, consistent with the other results, is not significant. When looking at the
results for the pocketbook hypothesis, the results present weaker explanatory power for
the economic voting argument. Chihuahua does not reach significance at all. Only Nuevo
Leon and Querétaro present support for the argument. For Querétaro, consistent with the
other models, the pocketbook argument appears to have stronger explanatory power than
the sociotropic one. The results in Michoacan are unexpected, because the coefficient is
significant but negative. Party identity remains the one variable consistently significant
and the strongest.

Table 5.5 presents the first results of the merged data set for the dependent
variable of gubernatorial support. The table includes the results for three models. Model 1
is the same model previously used with the individual states. The economic voting
hypothesis is strongly supported in both the sociotropic and the pocketbook hypotheses
across the three models. Consistent with previous results, the sociotropic hypothesis is'
supported clearly stronger. Model 2 includes the state-level variable in-party to help
account for state—level factors. As discussed earlier, my intuition is that economic voters

living in states where the governor belongs to the president’s party, will be more likely to

104

approve or support less the incumbent governor, which would be a sign that such voters
are able to distinguish between both layers of government. The results of Model 2
support this hypothesis. Model 3 includes an interaction term composed of the in-party
variable and the sociotropic and pocketbook evaluations. The expectation is that
economic voters living in in-party states and with positive economic performance
evaluations, will be less likely to approve or support the incumbent governor. This
hypothesis is supported in Model 3. Again, across all models, partisan identity is
consistently significant.

Table 5.6 presents the merged results for the dependent variable of incumbent
electoral support. The same three models are included here. Again, the results show
consistent support for the sociotropic hypothesis across the three models. The results for
the pocketbook hypothesis are more mixed. The hypothesis is supported in Models 1 and
2, but not so in Model 3. The in-party variable is significant and in the expected
direction, as it is the case with the interaction term.

Finally, Table 5.7 presents the results for the dependent variable of incumbent
support in state legislature’s elections. Similarly with the results of incumbent
gubernatorial result, the economic voting hypothesis is consistently supported with the
sociotropic argument. The in-party variable is significant and in the expected direction,

however, the interaction term does not reach significance.

5.6 Conclusion

The current chapter explored the relationship between individual economic

assessments and gubernatorial approval and vote choice on the basis of survey data

105

collected. in six Mexican states. Overall, the results presented in this chapter show support
for the economic voting argument explored here, but to different degrees across the
states.

The results offer clear support for the retrospective sociotropic hypothesis. The
findings suggest that governors are held accountable for state-level economic
performance. More precisely, positive assessments about the state’s economy increase the
likelihood of approving and/or voting for the governor’s party. Both variables of main
interest—an individual’s evaluation of state-level and personal economic perfonnance—
are found to impact differently across the states under study. The variability across the
states, however, supports the theoretical expectations.

First, economic voting is clearly the strongest in Chihuahua, Jalisco, and Nuevo
Leon, the three states that have experienced alternation of power. Both Chihuahua and
Nuevo Leon have experienced in fact two alternations. To a lesser degree, but also
consistently, citizens in Querétaro are economic voters as well.

The fact that governors are evaluated on the basis of state economic performance
in Jalisco and Nuevo Leon, more than in the other states, suggests that political
decentralization is associated with increased democratic accountability. More
decentralized regimes seem to provide political incentives for good subnational economic
management and public service provision. This shows that governors that manage the
economy well can actually gain electoral benefits from managing their state more
efficiently, while those who fail to generate good economic performance run the risk of

losing support.

106

The findings of this chapter support the argument that gubernatorial approval
reflects a combination of both national and state-level evaluations (Svoboda 1995; King
2001). Both dependent variables used in the analysis, gubernatorial approval and vote
choice present clear evidence that national-level factors have strong explanatory power.
For instance, partisan identity appears with stronger support in results for vote choice.
Here, the evidence supporting the economic voting theory are weaker, but still significant
in the expected states. States with higher levels of dependency on the federal government
(such as Chiapas) are expected to depend mostly on partisan loyalty. The different results

between approval and vote choice should be further examined.

107

can States, 1970-2004

ext

Figure 3.5 Municipal-Level Competition (Holbrook- Van Dunk Index), M

 

 

 

 

I 1970-1989
I 1990-2004

 

 

 

 

 

_
—

—-_er
—
—

— - ‘ ,L wruu.nmymfx. W’ .
_—
_—
_—
_—
_—
—t—|_"W- .
_
—
_
—
-———;._W~
_1_——-.W_
‘—-—— Wm
—

- . MW

1———
__-
1_1___
_——
_ - - _ I v.8”mwflm
_—

-h-II— “*NWM WW.
——_
__—_....

Kqu‘xu TM . .

-
=_ ”I ~ I-.:..r
--—-— t‘SW‘WW-vlv

-. .wm~

 

 

 

90

80

I I I I I I I I
C Q Q C O Q C O
F VD If) V M N '—

9” 3,9- «We

a

e 95 e o o e 4 o s e

 

 

108

Source: Table 4.

Figure 5.1 State ’5 Own Revenues, Average Percentage 2000-2002

 

 

NUEVO LEON
QUERETARO
MICHOACAN 223" " " " "’-"~"

““300 .. .

    

CHIHUAHUA

  
   

 

 
   

i 1
CHIAPAS I t 1 3 i I
l . . 1 | 1 I I , .
0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00
Percentage

 

 

 

Source: INEGI

109

1
1
1
1
1
1
1
1
1
1

 

State

 

 

 

 

 

 

1
weaned 1
1

F‘

 

Figure 5.2 Mexican States, Own Revenues (% of total state revenue), Average 1989-2003

110

Table 5.2 Gubernatorial Approval, Selected Mexican States (Ordered Probit Analysis)

 

Nuevo
Chiagts Chihuahua Jalisco Michoacan Lec’m Querétaro
State Economy .474*** 515*" l.164*** .051 1.478*** 182*”
(.078) (.055) (.065) (.063) (.145) (.060)
Personal
Finances 195*“ .152“ n.a. .184*** 185*" .448***
(.073) (.060) (.068) (.061) (.072)
Party ID .649*** .744*** .479*** .468*** .478*** .438***
(.088) (.083) (.082) (.085) (.097) (.097)
Interest .131 ** -.001 -.024 -.020 .007 .079
(.065) (.064) (.062) (.056) (.064) (.064)
Knowledge .038 .1 17" .036 -.006 .063 .1 14
(.047) (.053) (.047) (.043) (.067) (.071)
Education -.104 .012 .027 -.009 .092* -.027
(.064) (.059) (.055) (.025) (.054) (.061)
Income -.1 12 -.094 .056 -.135** 219*" .052
(.083) (.070) (.065) (.067) (.073) (.075)
Urban (=1) -.185** -.014 .042 .048 -.126 .097
(.088) (.1 17) (.089) (.086) (.185) (.092)
Log Likelihood -794.09 -829.75 -763.75 -872.93 -689.88 -681.66
Pseudo
R-Squared .091 l .1374 .2519 .0262 .1422 .0957
N 840 1000 1205 1008 1200 855
Cut 1 .496 1.139 -.922 2.273 .707
(.259) (.235) (.224) (.450) (.285)
Cut 2 2.146 3.947 .987 5.096 2.824
(.265) (.267) (.224) (.461) (.298)

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .1, ** = statistically
significant at .05, and *** = statistically significant at .01 or better.

111

 

 

 

 

 

 

 

Table 5.3 Vote Choice in Gubernatorial Elections, Selected Mexican States

 

(Probit Analysis)
Nuevo
Chiapas Chihuahua Jalisco Michoacan Leon Querétaro
State Economy .089 299*" .475*** .092 519*" .131*
(.095) (.065) (.061) (.086) (.186) (.074)
Personal
Finances .067 .088 n.a. -.130 .355*** .312***
(.095) (.072) (.091) (.076) (.088)
Party 1D 1857*" 1477*" l.470*** 2039*“ 2026*" 1.531 ***
(.1 10) (.097) (.089) (.034) (.1 10) (.l 13)
Interest .137 .034 .044 .109 .01 1 .157*
(.087) (.075) (.070) (.079) (.082) (.081)
Knowledge -.068 .054 -.002 .013 .109 -.040
(.062) (.064) (.052) (.059) (.085) (.087)
Education -.089 .141” .033 .039 .123* 205*“
(.087) (.070) (.061) (.034) (.065) (.076)
Income .046 -.140* .046 -.252*** .066 .008
(.109) (.084) (.074) (.093) (.090) (.094)
Urban (=1) -.006 -.041 .100 .086 -.056 .074
(.117) (.138) (.099) (.119) (.234) (.115)
Constant -1.478*** -1.577*** -2.21*** -l.123*** -3.551*** -2.389***
(.417) (.315) (.272) (.302) (.575) (.360)
N 840 1000 1205 1008 1200 855
Pseudo
R-Squared .3369 .2585 .2772 .3493 .3804 .2747
Log Likelihood -364.21 -513.92 -590.56 -397.62 -433.04 -392.16

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .1, ** = statistically
significant at .05, and *** = statistically significant at .01 or better.

112

Table 5.4 Vote Choice in Local Legislature Elections, Selected Mexican States

 

(Probit Analysis)
Nuevo
Chihuahua Jalisco Michoacan Leon Querétaro
State Economy 274*” 350*" .101 .434“ .168"
(.066) (.058) (.080) (.173) (.074)
Personal
Finances .006 n.a. -.184** 258*“ .302***
(.072) (.085) (.075) (.088)
Party 1D 1513*” 1.175*** 1585*” 1902*“ 1333*"
(.096) (.085) (.101) (.106) (.1 10)
Interest .037 -.045 .054 -.053 .098
(.075) (.066) (.073) (.081) (.081)
Knowledge .043 -.059 .053 .083 .062
(.064) (.050) (.055) (.085) (.090)
Education .004 .012 .040 127* .167”
(.070) (.058) (.031) (.065) (.075)
lncome -.103 * .086 .008 .001 -.091
(.084) (.070) (.085) (.089) (.094)
Urban (=1) .005 .029 .040 .007 .263
(.136) (.094) (.111) (.229) (.116)
Constant -l.326*** 1.457*** -l.214*** -2.963*** -2.581***
(.315) (.254) (.287) (.541) (.369)
N 1000 1205 1008 1200 855
Pseudo
R-Squared .2608 .1855 .2273 .3491 .2413
Log Likelihood -51 1.90 -663.97 -474.87 -441.34 -393.96

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .1, ** = statistically
significant at .05, and *** = statistically significant at .01 or better.

113

Table 5.5 Gubernatorial Approval, Selected Mexican States, Merged Data Set

(Ordered Probit Analysis)

 

Model 1 Model 2 Model 3
State Economy .473*** 463*" .547***
(.027) (.027) (.034)
Personal Finances .105*** .140*** .120***
(.026) (.027) (.042)
Party ID 638*" 609*" .607***
(.036) (.036) (.037)
Interest .009 .036 .025
(.025) (.026) (.026)
Knowledge .055“ 096*" 099*”
(.022) (.022) (.022)
Urban (=1) -.073*"‘ .056“ .049
(.031) (.033) (.034)
Education .004 -.035** -.035"
(.017) (.017) (.017)
Income -.042 .003 .008
(.028) (.029) (.029)
lnparty -.439*** -.229*
(.036) (.123)
State
Economy*lnparty -. 102* "‘
(.045)
Personal
Finances“ lnparty .013
(.054)
N 4903 4903 4784
Cut 1 .069 .132 .266
(.107) (.107) (.125)
Cut 2 (2.006) 2.080 2.206
(.109) .110 (.127)
Cut 3 2.913 3.029 3.156
(.113) .114 (.131)
Log Likelihood 4936.98 -4863.01 -4754.53
LR Chi-Squared 852.78 1000.72 1017.43
Pseudo R-Squared .0795 .0933 .0967

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .l. ** = statistically
significant at .05, and *** = statistically significant at .01 or better.
Merged data set does not include Jalisco.

114

Table 5.6 Vote Choice in Gubernatorial Elections, Selected Mexican States, Merged Data Set

 

(Probit Analysis)
Model 1 Model 2 Model 3
Sociotropic 222*" .21 1‘“M .274***
(.036) (.036) (.045)
Pocketbook .102*** .121 *** -.054
(.035) (.036) (.055)
Party 10 1826*" 1813*" 1795*”
(.046) (.046) (.047)
Interest .047 .062 .059*
(.034) (.034) (.035)
Date .014 .032 .024
(.029) (.029) (.030)
Locality -.025 .038 .035
(.042) (.044) (.046)
Education .055" .035 .036
(.022) (.023) (.023)
lncome -.034 -.01 1 -.01 1
(.03 8) (.038) (.039)
lnparty -.220*** -.562*"'*
(.047) (.168)
State
Economy“ lnparty -.107*
(.062)
Personal
F inances*lnparty 300*“
(.072)
Constant -1.802*** -1.829*** -l.599***
(.144) (.145) (.165)
N 4903 4903 4784
Log Likelihood -2180.90 -2169.98 -2116.74
LR Chi-Squared 1978.28 2000.12 1966.96
Pseudo R—Squared .3120 0.3155 .3172

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .1,
** = statistically significant at .05, and *** = statistically significant at .01 or better.

Merged data set includes all four states.

115

Table 5. 7 Vote Choice in Local Legislature ’s Elections, Selected Mexican States,
Merged Data Set (Probit Analysis)

 

Model 1 Model 2 Model 3
Sociotropic 238*" 225*" .256***
(.038) (.038) (.047)
Pocketbook .059 .080" -.1 13**
(.03 8) (.038) (.054)
Party ID 1663*" 1642*“ 1.616***
(.050) (.050) (.051)
Interest -.003 .019 .010*
(.037) (.037) (.038)
Date .049 .082" .081"
(.033) (.033) (.033)
Locality .084 .1 12 .083
(.060) (.060) (.061)
Education .071 ** .048" .050"
(.023) (.023) (.023)
Income -.070* -.024 -.024
(.041) (.042) (.042)
lnparty -.276*** -.999***
(.049) (.202)
State
Economy" lnparty .001
(.089)
Personal
Finances“ lnparty .373***
(078)
Constant -1.796*** -l.865*** -1.513***
(.155) (.156) (.173)
N 4063 4063 4012
Log Likelihood -1880. 10 -1864.58 -1831.68
LR Chi-Squared 1390.58 1421.63 1430.99
Pseudo R-Squared .2700 0.2760 .2809

 

Notes: Standard errors are listed in parentheses. * = statistically significant at .1,
** = statistically significant at .05, and *** = statistically significant at .01 or better.

Chiapas is not included.

116

Chapter 6. Conclusion

There is an overall scholarly consensus that the Mexican democratic transition
process culminated with the alternation of power in 2000, when the PRI was unseated by
President Fox’s PAN. Having observed these events unfold firsthand, and seeing
expectations and hopes rise among citizens, the question that underlies this dissertation
took form in the following way: to what extent were these national-level political changes
reaching Mexican citizens?

As I have stressed so far, one of the PRI-regime’s major shortcomings was its
lack of responsiveness to the public. For the most part, the basic form of political
participation—elections—had no real meaning to citizens, beyond being for the most part
a ritual of transfer of power within the PRI. So, within this new electoral context, in
which Mexico could finally be labeled an electoral democracy, did the government turn
accountable to its citizens? In other words, and as I ask in the introduction of this
dissertation, to what extent does the label “new democracy” guarantee working
mechanisms of accountability? Without doubt, establishing effective circuits of
accountability between politicians and electoral constituencies constitutes a latent
challenge not only in Mexico, but in new democracies more generally.

Exploring this question by looking at the Mexican states took me invariably to
hypothesize that democracy had reached citizens to different degrees, that is, state
governments vary according to their individual experiences. Within different contexts,
elections acquire different meanings. To the extent that state-level political leaders
become autonomous from the central government, accountability increases. The different

hypotheses in this study had the overall purpose to test under what conditions do

117

elections work as an instrument of accountability. The results support in different ways
the underlying argument that elections acquire a power of accountability in function of a
state’s level of political and economic decentralization.

In order to undertake this type of analysis, I had to answer one prior question: Is
the economy part of the voting calculus among the Mexican electorate? The macro-level
analysis in Chapter 4 looks at this question, and the findings show evidence that a state’s
economic performance has been indeed part of the voting decision in gubernatorial
elections during the last nineteen years. Once this prior question had been answered, it
was possible to look at the effects of political and economic decentralization on electoral
behavior. According to the expectations put forward here, both factors probed to have
effects on electoral behavior. States performing better in economic terms, with higher
levels of economic autonomy vis-a-vis the federal government, and with experience
under a non-PRI government, probed to be more likely to have economic voters among
its population, who in turn are capable of holding the government accountable for its
performance on election day.

The second part of the analysis took on the task of exploring the apparent
variations across the states by using micro-level data. Clearly, the results are limited in
several ways, and need to be further expanded, but the findings did support in different
ways the key argument of this work. Overall, the likelihood of Mexicans being economic
voters depends on the political and economic contexts they live in. The results of the
statistical analyses undertaken here show support for the economic voting theory, but
there is visible variation in the strength of the theory across the states. One important

result that came to the surface in this analysis is the distinction between approval and

118

vote choice. In general, the theory appeared consistently stronger in regard with approval.
A reason could be that respondent’s feel freer to express their opinion about the sitting
government when they are “just” being asked whether they approve or not of the
government’s performance. On the other hand, when asked for whom they would actually
vote, then voters seem to go back to “traditional” cleavages, mostly to their partisan
identity. However, based on the findings of chapter 5, it is safe to say that some form of
economic voting is taking place in the Mexican states, with stronger evidence in those
states that are both economically better off and have experienced rule under a party other
than the PRI.

These findings have different implications for the study of the quality of new
democracies. First, and perhaps this dissertation’s most valuable contribution, this study
sheds light on the dynamics taking place at the subnational level in a new democracy. It
shows that democratization reaches citizens unevenly, and that the manner in which they
experience democracy is a function to a great extent of state-level factors. For one, the
nature of the opposition is of crucial importance. As argued by Manin, Przeworski, and
Stokes (1999:48), “Citizens have two agents, not just one: the incumbents who govern
and the opposition that wants to become the government. . .The opposition has incentives
to monitor the government ant to inform (truthfully or not) voters about the performance
of the incumbents.” In fact, opposition parties constitute a key element in developing
partisan competition, which in turn causes citizens to be more attentive to politics (Muller
and Seligson 1994).

Furthermore, this dissertation joins the claim developed in different forms by

scholars of Mexican politics (Cornelius 1999; Fox 1994; Lawson 2000), and which was

119

 

so clearly put by Hiskey and Bowler (2005:67) who see the local context “as a critical
frontline for these struggling democracies, one where support for and engagement with a
democratic system of government can either be fostered or frustrated.” In other words, it
would be a mistake to see the nation through the national-level changes, and make
assumptions based on such. Local politics matter, and they produce different types of
electorate as well.

Second, by studying economic voting at the subnational level in a new
democracy, this study contributes to the broad field of electoral behavior. Studying the
way in which voters elect their leaders is essential to understanding the functioning of
any democracy. This exercise becomes more relevant in nascent democracies, where
voters are “learning democracy” (Anderson and Dodd 2005). The findings of this
dissertation shed light on this learning process, one which is widely acknowledged but
difficult to operationalize. Overall, this dissertation probes the broader applicability of the
theory, but it also warns on the importance of taking into account variations in the
economic and political contexts.

Finally, these findings have direct implications for the study of accountability in
new democratic contexts. Both the aggregate and individual-level analyses suggested that
opposition (or non-PRI) governments develop different styles of governing that involve
most of the times greater communication with citizens (greater transparency and access to
information). These conclusions are very positive in regard to the development of a
culture of accountability, which was not fostered during the one-party regime. They
suggest that accountability has greater chances of appearing and staying, when it

becomes part of the political survival strategy of subnational leaders. That is, elections

120

 

will be used by voters as an instrument of accountability to the extent that subnational
political leaders are in need of their local constituents—a practice that is quite new in

Mexico.

 

121

APPENDICES

122

Appendix A. States and Election Years, Chapter 4

Table A. Gubernatorial Elections, Mexican States, 1985-2004

 

 

 

 

 

 

Election
State year State Election year State Election year
AGS 1986 H00 1987 SLP 1985
A08 1992 HGO 1993 SLP 1991
AGS 1998 H00 1999 8 LP 1997
AGS 2004 1 AL 1988 SLP 2003
BC 1989 I AL 1995 SIN 1986
BC 1995 J AL 2000 SIN 1992
BC 2001 MEX 1987 SIN 1998
BCS 1987 MEX 1993 SIN 2004
BCS 1993 MEX 1999 SON 1985
BC S 1999 MICH 1986 SON 1991
CAMP 1985 MIC H 1992 SON 1997
CAMP 1991 MIC H 1995 SON 2003
CAMP 1997 MICH 2001 SON 1985
CAMP 2003 MOR 1988 SON 1991
COAH 1 987 MOR 1994 SON 1 997
COAH 1 993 MOR 2000 SON 2003
COAH 1 999 NAY 1987 TAB 1988
COL 1985 NAY 1993 TAB 1994
COL 1991 NAY 1999 TAB 2001
COL 1 997 N L 1 985 TAM 1 986
COL 2003 NL 1991 TAM 1992
CHIS 1988 N L 1997 TAM 1998
CHIS 1994 N L 2003 TAM 2004
CHIS 2000 OAX 1986 TLAX 1986
C HIH 1 986 OAX 1 992 TLAX 1 992
C HIH 1992 OAX 1998 TLAX 1998
C HIH 1998 OAX 2004 TLAX 2004
C HIH 2004 PUE 1986 VER 1986
D00 1 986 PU E 1992 VER 1992
D00 1992 PU E 1998 VER 1998
DGO 1998 PUE 2004 VER 2004
D60 2004 QRO 1985 YUC 1987
GTO 1985 QRO 1991 YUC 1993
GTO 1991 QRO 1997 YUC 1995
GTO 1995 QRO 2003 YUC 2001
GTO 2000 QROO 1987 ZAC 1 986
GRO 1986 QROO 1993 ZAC 1992
GRO 1993 QROO 1999 ZAC 1998
GRO 1 999 ZAC 2004
DF 2000
Source: IF E.

123

 

Table B. State Legislature Elections, Mexican States, 1985-2004

 

 

 

 

 

 

 

 

 

 

Election Electionl Election Election Election Electionl
State Year State Year State Year State Year State Year State Year
AGS 1985 CHIH 1985 MICH 1985 QRO 1987 QRO 1987 TAM 1986
AGS 1988 CHIH 1988 MICH 1988 QRO 1 990 QRO 1990 TAM 1 989
AGS 1991 CHIH 1991 MICH 1991 QRO 1993 QRO 1993 TAM 1992
AGS 1994 CHIH 1994 MIC H 1994 QRO 1996 QRO 1 996 TAM 1 995
AGS 1997 CHIH 1997 MICH 1997 QRO 1999 QRO 1 999 TAM 1 998
AGS 2000 CHIH 2000 MICH 2000 QRO 2002 QRO 2002 TAM 2000
AGS 2003 C HIH 2003 MICH 2003 QROO 1986 QROO 1986 TAM 2003
BC 1985 DGO 1985 MOR 1987 QROO 1989 QROO 1989 TLAX 1 985
BC 1 988 DGO 1988 MOR 1 990 QROO 1992 QROO 1 992 TLAX 1988
BC 1991 DGO 1991 MOR 1993 QROO 1995 QROO 1995 TLAX 1991
BC 1 994 DGO 1 994 MOR 1 996 QROO 1 998 QROO 1 998 TLAX 1994
BC 1997 DGO 1997 MOR 1999 QROO 2001 QROO 2001 TLAX 1997
BC 2000 DGO 2000 MOR 2002 SLP 1986 SLP 1986 TLAX 2000
BC 2003 DGO 2003 NAY 1986 SLP 1989 SLP 1989 TLAX 2003
BC S 1986 GTO 1987 NAY 1989 SLP 1992 SLP 1992 VER 1985
BCS 1989 GTO 1990 NAY 1992 SLP 1996 SLP 1996 VER 1988
BCS 1992 GTO 1993 NAY 1995 SLP 1999 SLP 1999 VER 1991
BCS 1995 GTO 1996 NAY 1998 S LP 2002 SLP 2002 VER 1994
BCS 1998 GTO 1999 NAY 2001 SIN 1985 SIN 1985 VE R 1997
BCS 2001 GTO 2002 NL 1987 SIN 1988 SIN 1988 VER 2000
CAMP 1985 GRO 1986 NL 1990 SIN 1991 SIN 1991 VER 2003
CAMP 1988 GRO 1989 NL 1993 SIN 1994 SIN 1994 YUC 1987
CAMP 1991 GRO 1992 NL 1996 SIN 1997 SIN 1997 YUC 1990
CAMP 1993 GRO 1995 NL 1999 SIN 2000 SIN 2000 YUC 1993
CAMP 1996 GRO 1998 NL 2002 SIN 2003 SIN 2003 YUC 1994
CAMP 1999 GRO 2001 OAX 1985 SON 1987 SON 1987 YUC 1995
CAMP 2002 HGO 1986 OAX 1988 SON 1990 SON 1990 YUC 1 997
COAH 1987 HGO 1989 OAX 1991 SON 1993 SON 1993 YUC 2000
COAH 1990 HGO 1992 OAX 1994 SON 1996 SON 1996 YUC 2003
C OAH 1993 HGO 1 995 OAX 1997 SON 1999 SON 1999 ZAC 1985
COAH 1996 HGO 1998 OAX 2000 SON 2002 SON 2002 ZAC 1988
COAH 1998 HGO 2001 OAX 2003 TAB 1985 TAB 1985 ZAC 1991
COAH 2001 J AL 1985 PUE 1986 TAB 1988 TAB 1988 ZAC 1994
COL 1987 1 AL 1988 PU E 1989 TAB 1991 TAB 1991 ZAC 1997
COL 1990 J AL 1991 PU E 1992 TAB 1994 TAB 1994 ZAC 2000
COL 1993 J AL 1994 PU E 1995 TAB 1997 TAB 1997 ZAC 2003
COL 1996 JAL 1997 PU E 1997 TAB 1999 TAB 1999 DE 1987
COL 1999 JAL 2000 PUE 2000 TAB 2002 TAB 2002 DF 1990
COL 2002 JAL 2002 PU E 2003 DP 1993
CHIS 1987 MEX 1986 DF 1996
CHIS 1990 MEX 1989 DF 1999
CHIS 1994 MEX 1992 DF 2002
CHIS 1997 MEX 1995
CHIS 2000 MEX 1999
CHIS 2003 MEX 2002
Source: [F E.

124

Appendix B. Question Wording and Coding, Chapter 5

Variables from pre-electoral surveys collected by Reforma in Chiapas, Chihuahua,
Jalisco, Michoacan, Nuevo Leon, and Querétaro.

CHIAPAS

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the
elections for governor were taking place today, for whom would you vote? (Ask
respondent to place ballot in the ballot box)]
Recoded: 1=Incumbent Party (PRI); 0=Not Incumbent

(Entregar boleta con candidatos a1 entrevistado y preguntar): Le voy a entregar una boleta
con los nombres de los candidatos a Gobemador. Si hoy hubiera elecciones para
Gobemador del estado, gpor quie'n votaria usted? (Pedir que marque su voto en la boleta
y solicitar que esta se deposite en la urna)

Vote Choice in Local Legislature Elections
Not asked in Chiapas.

Gubernatorial Approval

[In general, how do you rate Governor Roberto Albores Guillén’s administration?]
1. Good job

2. Poor job

3. Neither good nor bad

4. Don’t know/Did not answer

Recoded: 1=Poor job, 2 + 4=Neither good nor bad, 3=Good job

En general, gusted considera que el Gobiemo de Roberto Albores Guillén ha sido un
buen gobiemo 0 ha sido un mal gobiemo?

1. Buen gobiemo

2. Mal gobiemo

3. Ni bueno ni malo

4. No sabe/No contesto

State Economy

[How do you rate economic conditions in the state compared to a year ago?]
1. Improved

2. Stayed the same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2 + 4=Stayed the same, 3=Improved

Comparada con hace un afio, gdiria usted que la situacion economica del estado mejoro,
empeoro, o sigue igual?

125

1. Mejoré

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Personal Finances .

[How do you rate your personal economic conditions compared to a year ago?]
1. Improved

2. Stayed the same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2 + 4=Stayed the same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacién personal mejoro, empeoro, o
sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Party ID
[Do you consider yourself ...(priista, panista, or perredista)? Very much or somewhat?)
1. Very
2. Somewhat
3. Don’t know/Did not answer
Recoded: 1=Incumbent (PRI) [includes “very priista” and “somewhat priista”], O=Not
Incumbent

gSe considera usted como muy (PRIISTA / PANISTA / PERREDISTA) o algo
(PRIISTA / PANISTA / PERREDISTA)?

1.Muy

2. Algo

3. No sabe/No contesto

Interest

[How interested are you in the political campaigns going on related to the gubernatorial
elections taking place in the state?]

1. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2+3+5)=Some; 3=A lot

gQue’ tan interesado esta usted en las campafias de los candidatos a la gubematura del
estado:

126

l. Mucho
2. Algo

3. Poco
4. Nada

Knowledge

[Can you tell me the date when the next gubematurial elections will take place in the
state?]

1. Mentioned exact date

2. Mentioned incomplete date

3. Does not remember

4. Does not know

Recoded: 1 (3+4)=Does not know; 2=Incomplete; 3=Complete

gMe podria decir la fecha en que se llevaran a cabo las elecciones para Gobemador del
Estado, 0 en este momento no recuerda?

1. Menciono fecha completa

2. Mencioné fecha incompleta

3. No recuerda

4. No sabe

Education

[What is the highest level of education you have completed?]
1=No formal schooling

2. Primary schooling

3. Secondary/technical schooling

4. High school/equivalent

5. University and more

6. Did not answer

Recoded: 1=Secondary or less, 2=Preparatory; 3=University

gHasta que afio estudio usted? gCual es su ultimo grado de estudios?
1. No tiene estudios

2. Primaria

3. Secundaria/Técnica

4. Preparatoria/Equivalente

5. Universidad o mas

6. No contesto

Income
[How much money do you (and your household together) earn per month?]
1. Between $0 and $1,000
2. Between $1,001 and $2,000
3. Between $2,001 and $4,000
4. Between $4,001 and $6,000
5. Between $6,001 and $8,000

127

6. Between $8,001 and $12,000

7. Between $12,001 and $16,000

8. Between $16,001 and $20,000

9. Over $21,000

10. Don’t know/Did not answer

Recoded: 1=Between $0-$4,000; 2=$4,001-12,000; 3=Over $12,001

Entre todas las personas que viven con usted, gaproximadamente cuanto ganan a1 mes?
1. De $0 a $1,000

2. De $1,001 a $2,000
3. De $2,001 a $4,000
4. De $4,001 a $6,000
5. De $6,001 a $8,000
6. De $8,001 a $12,000
7. De $12,001 a $16,000
8. De $16,001 a $20,000
9. Mas de $21,000

10. No sabe/No contesto

Locality
[Type of locality: 1=Urban; 2=Rura1]

Tipo de localidad:
1. Urbano
2. Rural

CHIHUAHUA

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the elections
for governor were taking place today, for whom would you vote? (Ask respondent to
place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PRI); 0=Not Incumbent

(Desprender boleta con candidatos a Gobemador) Con esta boleta no oficial y que es 3610
para propésitos de esta encuesta, digame por favor, si hoy hubiera elecciones para
Gobemador de Chihuahua, [,pOI' quién votaria usted? (Pida que deposite 1a boleta en la
uma)

Vote Choice, Local Legislature

[(Hand out ballot containing ballot for local representatives to the respondent and ask) If
the elections for local representatives were taking place today, for whom would you vote?
(Ask respondent to place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PRI); 0=Not Incumbent

128

(Desprender boleta para eleccion de Diputados Locales) Con esta boleta no oficial y que
es 5610 para propositos de esta encuesta, digame por favor, si hoy hubiera elecciones para
diputados locales, gpor cual partido votaria usted? (Pida que deposite la boleta en la urna)

Gubernatorial Approval

[In general, do you approve or disapprove of the way Patricio Martinez is doing his job as
govemor?]

1. Approves completely

2. Approves some

3. Disapproves some

4. Disapproves completely

5. Don’t know/Did not answer

Recode: 1=Disapproves completely; 2=Approves/Disapproves some; 3=Approves
completely

En general, busted aprueba o desaprueba Ia forma en que Patricio Martinez esta haciendo
su trabajo como Gobemador? (Insistir: lo...totalmente o algo?)

I. Aprueba totalmente

2. Aprueba a1 go

3. Desaprueba algo

4. Desaprueba totalmente

5. No sabe/No contesto

State Economy

[In general, do you consider conditions in Chihuahua have improved, remained the same
or worsened under the administration of Patricio Martinez?]

1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2 + 4=Same, 3=Improved

En general, gusted considera que la situacién de Chihuahua ha mejorado, sigue igual 0 ha
empeorado con el trabajo de Patricio Martinez como Gobemador? (Insistir: ...totalmente
o algo?)

1. Mejorado totalmente

2. Mejorado algo

3. Sigue igual

4. Empeorado algo

5. Empeorado totalmente

6. No sabe/No contesto

Personal Finances

[How do you rate your own economic conditions compared to a year ago?]
1. Improved

2. Same

I29

3. Worsened
4. Don’t know/Did not answer
Recoded: 1=Worsened, 2 + 4=Same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacion economica personal mejoro,
empeoro, o sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Party ID
[In general, do you consider yourself priista, panista, or perredista?]
1. Very priista
2. Somewhat priista
3. Very panista
4. Somewhat panista
5. Very perredista
6. Somewhat perredista
7. Other
8. None
Recoded: 1=Incumbent (PRI) [includes “very priista” and “somewhat priista”], 0=Not
Incumbent

Generalmente, gusted se considera priista, panista o perredista?
1. Muy priista

2. Algo priista

3. Muy panista

4. Algo panista

5. Muy perredista

6. Algo perredista

7. Otro

8. Ninguno

9. No sabe/No contesto

Interest

[How interested are you in the political campaigns going on related to the gubernatorial
elections taking place in the state?]

1. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2 + 3 + 5)=Some; 3=A lot

130

gQué tan interesado esta usted en las campafias de los candidatos a la Gubernatura del
Estado?

1. Mucho

2. Algo

3. Poco

4. Nada

5. No sabe/No contesto

Knowledge

[Can you tell me the date when the next gubernatorial elections will take place in the
state?]

1. Mentioned exact date

2. Mentioned incomplete date

3. Does not remember

4. Does not know

5. Did not answer

Recoded: 1=Does not know; 2 (2+3+5)=Incomplete; 3=Complete

gSabe usted la fecha de las proximas elecciones para Gobemador de Chihuahua?
1. Sabe la fecha completa

2. Supo solo e1 mes y el afio

3. Supo solo el afio

4. No la sabe
5. No contesto

Education

gHasta que afio o grado aprobo (paso) en la escuela?
l. Ninguno

2. Preescolar o Kinder

3. Primaria

4. Secundaria

5. Preparatoria o bachillerato

6. Normal

7. Carrera te’cnica o comercial

8. Profesional

9. Maestria o Doctorado

10. No sabe/No contesto

Recoded: 1=Secondary or less, 2=Preparatory; 3=University

Income
See Chiapas.

Locality

[Type of locality: 1=Urban; 2=Rura1; 3=Mixed]
Recoded: 1=Urban/Mixed, 0=Rural

I31

Tipo de localidad:
l. Urbano

2. Rural

3. Mixto

JALISCO

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the elections
for governor were taking place today, for whom would you vote? (Ask respondent to
place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

Le voy a entregar una boleta con Ios nombres de los candidatos parecida a la que se usara
el dia de las elecciones. Esta boleta no es oficial y solo se utilizara para propésitos de esta
encuesta. Por favor marque en ella, si hoy hubiera elecciones para Gobemador de J alisco,
gpor quién votaria usted? Solicitar que la boleta se deposite en la uma.

Vote Choice, Local Legislature

[(Hand out ballot containing ballot for local representatives to the respondent and ask) If
the elections for local representatives were taking place today, for whom would you vote?
(Ask respondent to place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

(Entregar boleta con partidos y preguntar: Si hoy hubiera elecciones para diputados
locales, gpor cual partido votaria usted? (Pedir que deposite la boleta en la uma)

Gubernatorial Approval

[In general, do you approve or disapprove of the way Alberto Cardenas is doing his job
as govemor?]

1. Approves completely

2. Approves some

3. Disapproves some

4. Disapproves completely

5. Don’t know/Did not answer

En general, gusted aprueba o desaprueba 1a forma como Alberto Cardenas esta haciendo
su trabajo como Gobemador? (Insistir: lo...totalmente o algo?)

1. Aprueba totalmente

2. Aprueba algo

3. Desaprueba a1 go

4. Desaprueba totalmente

5. No sabe/No contesto

132

State Economy

[In general, do you consider conditions in Jalisco have improved, remained the same or
worsened under the administration of Alberto Cardenas?]

1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

En general, gusted considera que el Gobiemo de Alberto Cardenas ha sido un buen
Gobiemo 0 un mal Gobiemo?

I. Buen Gobiemo

2. Mal Gobiemo

3. Ni bueno ni malo

4. No sabe/No contesto

Personal Finances
This question was not asked in Jalisco.

Party ID
[In general, do you consider yourself priista, panista, or perredista?]
1. Very priista
2. Somewhat priista
3. Very panista
4. Somewhat panista
5. Very perredista
6. Somewhat perredista
7. Other
8. None
Recoded: 1=Incumbent (PAN) [includes “very panista” and “somewhat panista”], 0=Not
Incumbent

Independientemente del partido por el que usted votaria, gcon cual partido politico se
identifica mas?...Muy o algo?

1. Muy priista

2. Algo priista

3. Muy panista

4. Algo panista

5. Muy perredista
6. Algo perredista

7. Menciono otro

8. No sabe/No contesto
9. Ninguno

133

Interest

[How interested are you in the political campaigns going on related to the gubernatorial
elections taking place in the state?]

1. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2+3+5)=Some; 3=A lot

gQué tan interesado esta usted en las campafias de los aspirantes al gobiemo de Jalisco:
muy, algo, poco o nada interesado?

1. Muy

2. Algo

3. Poco

4. Nada

5. No sabe/No contesto

Knowledge

[Can you tell me the date when the next gubernatorial elections will take place in the
state?]

I. Mentioned exact date

2. Mentioned incomplete date

3. Does not remember

4. Does not remember/Does not know

Recoded: I (3+4)=Does not know; 2=Incomplete; 3=Complete

gMe podria decir la fecha en que se llevaran a cabo las pro'ximas elecciones para
Gobemador de J alisco. . .0 en este momento no recuerda?

1. Menciono fecha completa

2. Menciono fecha incompleta

3. No recuerda/No sabe

Education

gHasta que afio o grado aprobo (paso) en la escuela?
. Ninguno

. Preescolar o Kinder

. Primaria incompleta

. Primaria completa

. Secundaria incompleta

. Secundaria completa

. Preparatoria o bachillerato incompleto
. Preparatoria o bachillerato completo

. Carrera técnica o comercial

10. Normal

11. Universidad incompleta

OWQO‘M-RUJNI—t

I34

12. Universidad completa

13. Maestria o Doctorado incompleto

14. Maestn’a o Doctorado complete

15. No sabe/No contesto

Recoded: 1=Secondary or less, 2=Preparatory; 3=University

Income
See Chiapas.

Locality
[Type of locality: 1=Urban; 2=Mixed; 3=Rural]
Recoded: 1=Urban/Mixed, 0=Rural

Tipo de localidad:
1. Urbano

2. Mixto

3. Rural

MICHOACAN

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the
elections for governor were taking place today, for whom would you vote? (Ask
respondent to place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PRD); 0=Not Incumbent

(Entregar boleta con candidatos a1 entrevistado y preguntar): Le voy a entregar una boleta
con los nombres de los candidatos a Gobemador. Si hoy hubiera elecciones para
Gobemador del estado, por quien votaria usted? (Pedir que marque su voto en la boleta y
solicitar que esta se deposite en la uma)

Vote Choice, Local Legislature

(Entregar boleta para Diputados Locales al entrevistado y preguntar): Le voy a entregar
otra boleta. Si hoy hubiera elecciones para diputados locales, (3pm cual partido votaria
usted? (Pedir que marque su voto en la boleta y solicitar que esta se deposite en la urna)
Recoded: 1=Incumbent Party (PRD); 0=Not Incumbent

[(Hand out ballot containing ballot for local representatives to the respondent and ask) If
the elections for local representatives were taking place today, for whom would you vote?
(Ask respondent to place ballot in the ballot box)]

Gubernatorial Approval

[In general, do you approve or disapprove of the way Victor Manuel Tinoco Rubi is
performing his job as govemor?]

l. Approves completely

I35

2. Approves some

3. Disapproves some

4. Disapproves completely

5. Don’t know/Did not answer

Recoded: 1=Disapproves completely, 2=Approves/Disapproves some, 3=Approves
completely

En general, gusted aprueba o desaprueba la forma como Victor Manuel Tinoco Rubi esta
haciendo su trabajo como gobemador del estado? (Insistir): lo (aprueba/desaprueba)
totalmente o algo?

1. Aprueba totalmente

2. Aprueba algo

3. Desaprueba a1 go

4. Desaprueba totalmente

5. No sabe/No contesto

State Economy

[How do you rate economic conditions in the state compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

Comparada con hace un afio, gdiria usted que 1a situacion econémica del estado mejoro,
empeoro, o sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Personal Finances

[How do you rate your own economic conditions compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacion personal mejoro, empeoro, o
sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

I36

Party ID

[In general, do you consider yourself priista, panista, or perredista?]
. Very priista

. Somewhat

. Very panista

. Somewhat panista

. Very perredista

. Somewhat perredista

. Other

. None

9. Don’t know/Did not answer

Recoded: 1=Incumbent (PRD) [includes “very perredista” and “somewhat perredista”],
0=Not Incumbent

OO\10\M-BUJNt—‘

Generalmente, gusted se considera priista, panista o perredista?
. Muy priista

. Algo priista

. Muy panista

. Algo panista

. Muy perredista

. Algo perredista

. Otro

. Ninguno

9. No sabe/No contesto

OOQONUI-bDJNt-d

Interest

[How interested are you in the political campaigns going on related to the gubernatorial
elections taking place in the state?]

1. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2+3+5)=Some; 3=A lot

(,Que' tan interesado esta usted en las campafias de los candidatos a la gubematura del
estado?

1. Mucho

2. Algo

3. Poco

4. Nada

5. No sabe/No contesto

Knowledge

[Can you tell me the date when the next gubematurial elections will take place in the
state?]

137

 

 

1. Mentioned exact date

2. Mentioned incomplete date

3. Does not remember

Recoded: 1=Does not know; 2=Incomplete; 3=Complete

gMe podria decir la fecha en que se llevaran a cabo las elecciones para Gobemador del
estado. . .0 en este momento no recuerda?

1. Menciono fecha completa

2. Menciono fecha incompleta

3. No recuerda

Locality
[Type of locality: 1=Urban; 2=Rural]

Tipo de localidad:
1. Urbano
2. Rural

Education

[What is the highest level of education you have completed?]
1=No formal schooling .
2. Primary schooling

3. Secondary/technical schooling

4. High school/equivalent

5. University and more

6. Did not answer

Recoded: 1=Secondary or less, 2=Preparatory; 3=University

gHasta que afio estudio usted? gCual es su ultimo grado de estudios?
1. No tiene estudios

2. Primaria

3. Secundaria/Te’cnica

4. Preparatoria/Equivalente

5. Universidad o mas

6. No contesto

Income
See Chiapas.

NUEVO LEON

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the
elections for governor were taking place today, for whom would you vote? (Ask
respondent to place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

I38

 

(Desprender boleta con candidatos a gobemador). Con esta boleta no oficial y que solo es
para propésitos de esta encuesta, digame por favor, si las elecciones para Gobemador de
Nuevo Leon fueran hoy, gpor cual candidato votaria usted? (Solicitar que la boleta se
deposite en la urna).

Vote Choice, Local Legislature

[(Hand out ballot containing ballot for local representatives to the respondent and ask) If
the elections for local representatives were taking place today, for whom would you vote?
(Ask respondent to place ballot in the ballot box)]
Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

Con esta boleta no oficial y que es solo para propositos de esta encuesta digame por
favor, si las elecciones para diputados locales fueran hoy, gpor cual partido votaria usted?
(Que deposite en la urna)

Gubernatorial Approval

[In general, do you approve or disapprove of the way Fernando Elizondo Barragan is
performing his job as govemor?]

1. Approves completely

2. Approves some

3. Disapproves some

4. Disapproves completely

5. Don’t know/Did not answer

Recoded: 1=Disapproves completely, 2=Approves/Disapproves some, 3=Approves
completely

En general, gusted aprueba o desaprueba la forma como Fernando Elizondo Barragan esta
haciendo su trabajo como Gobemador? (Insistir): lo (aprueba/desaprueba) totalmente o
algo?

1. Aprueba totalmente

2. Aprueba algo

3. Desaprueba algo

4. Desaprueba totalmente

5. No sabe/No contesto

State Economy

[How do you rate economic conditions in the state compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

En general, gUsted considera que la situacion de Nuevo Leon ha mejorado, sigue igual 0
ha empeorado con el trabajo de Fernando Elizondo como Gobemador?

139

1. Mejorado totalmente
2. Mejorado algo

3. Sigue igual

4. Empeorado algo

5. Empeorado totalmente
6. No sabe/No contesto

Personal Finances

[How do you rate your own economic conditions compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacion personal mejoro, empeoro, o
sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Party ID

[In general, do you consider yourself priista, panista, or perredista?]
1. Very priista

2. Somewhat

3. Very panista

4. Somewhat panista

5. Very perredista

6. Somewhat perredista

7. Other

8. None

9. Don’t know/Did not answer

Recoded: 1=Incumbent (PAN) [includes “very panista” and “somewhat panista”], 0=Not
Incumbent

Generalmente, gusted se considera priista, panista o perredista?...muy o algo?
1. Muy priista

2. Algo priista

3. Muy panista

4. Algo panista

5. Otro

6. Ninguno

7. No sabe/No contesto

140

 

 

Interest

[How interested are you in the political campaigns going on related to the gubernatorial
elections taking place in the state?]

1. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2+3+5)=Some; 3=A lot

gQué tan interesado esta usted en votar en las elecciones de Gobemador de Nuevo Leon
de este afio?

1. Mucho

2. Algo

3. Poco

4. Nada

5. No sabe/No contesto

Knowledge

[Can you tell me the date when the next gubematurial elections will take place in the
state?]

1. Knows exact date

2. Knows only the year

3. Knows only the month and year

4. Does not know

5. Did not answer

Recoded: 1 (3+4+5)=Does not know; 2 (2+3)=Incomplete; 3=Complete

éSabe usted la fecha de las proximas elecciones para Gobemador de Nuevo Leon?
1. Sabe la fecha completa

2. Supo solo el afio

3. Supo solo el mes y el afio

4. No la sabe

5. No contesto

Locality
[Type of locality: 1=Urban; 2=Rural]

Tipo de localidad:
1. Urbano
2. Rural

Education

[What is the highest level of education you have completed?]
1=No formal schooling

2. Primary schooling

141

3. Secondary/technical schooling

4. High school/equivalent

5. University and more

6. Did not answer

Recoded: 1=Secondary or less, 2=Preparatory; 3=University

gHasta que afio estudio usted? gCual es su ultimo grado de estudios?
1. Ninguno

2. Preescolar o Zinder

3. Primaria

4. Secundaria

5. Preparatoria o bachillerato
6. Normal

7. Carrera te’cnica o comercial
8. Profesional

9. Maestria o Doctorado

99. No sabe/No contesto

Income
See Chiapas.

QUERETARO

Vote Choice, Gubernatorial Elections

[(Hand out ballot containing candidates’ names to the respondent and ask) If the elections
for governor were taking place today, for whom would you vote? (Ask respondent to
place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

(Entregar boleta con candidatos a1 entrevistado y preguntar): Le voy a entregar una boleta
con los nombres de los candidatos a Gobemador. Si hoy hubiera elecciones para
Gobemador del estado, gpor quién votaria usted? (Pedir que marque su voto en la boleta
y solicitar que esta se deposite en la urna)

Vote Choice, Local Legislature

[(Hand out ballot containing ballot for local representatives to the respondent and ask) If
the elections for local representatives were taking place today, for whom would you vote?
(Ask respondent to place ballot in the ballot box)]

Recoded: 1=Incumbent Party (PAN); 0=Not Incumbent

Entregar boleta para Diputados Locales al entrevistado y preguntar): Le voy a entregar
otra boleta. Si hoy hubiera elecciones para diputados locales, gpor cual partido votaria
usted? (Pedir que marque su voto en la boleta y solicitar que esta se deposite en la uma)

142

Gubernatorial Elections

[In general, do you approve or disapprove of the way Ignacio Loyola Vera is performing
his job as govemor?]

1. Approves completely

2. Approves some

3. Disapproves some

4. Disapproves completely

5. Don’t know/Did not answer

Recoded: 1=Disapproves completely, 2=Approves/Disapproves some, 3=Approves
completely

En general, gusted aprueba o desaprueba la forma como Ignacio Loyola Vera esta
haciendo su trabajo como gobemador del estado? (Insistir): Io (aprueba/desaprueba)
totalmente o algo?

1. Aprueba totalmente

2. Aprueba algo

3. Desaprueba algo

4. Desaprueba totalmente

5. No sabe/No contesto .

State Economy ,

[How do you rate economic conditions in the state compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacion economica del estado mejoro,
empeoro, o sigue igual?

1. Mejoro

2. Sigue igual

3. Empeoro

4. No sabe/No contesto

Personal Finances

[How do you rate your own economic conditions compared to a year ago?]
1. Improved

2. Same

3. Worsened

4. Don’t know/Did not answer

Recoded: 1=Worsened, 2=Same, 3=Improved

Comparada con hace un afio, gdiria usted que su situacion personal mejoro, empeoro, o

sigue igual?
1. Mejoro

143

2. Sigue igual
3. Empeoro
4. No sabe/No contesto

Party [D
[In general, do you consider yourself priista, panista, or perredista?]
1. Very priista
2. Somewhat priista
3. Very panista
4. Somewhat panista
5. Very perredista
6. Somewhat perredista
7. Other
8. None
Recoded: 1=Incumbent (PAN) [includes “very panista” and“somewhat panista”], O=Not
Incumbent

Generalmente, gusted se considera priista, panista o perredista?
1. Muy priista

2. Algo priista

3. Muy panista

4. Algo panista

5. Muy perredista

6. Algo perredista

7. Otro

8. Ninguno

9. No sabe/No contesto

Interest

[How interested are you in the political campaigns going on related to the gubernatorial .
elections taking place in the state?]

I. A lot

2. Some

3. A little

4. Not at all

5. Does not know/Did not answer

Recoded: 1=Nothing; 2 (2+3+5)=Some; 3=A lot

LQué tan interesado esta usted en las campafias de los candidatos a la gubematura del
estado: mucho, algo, poco o nada interesado?

1. Mucho

2. Algo

3. Poco

4. Nada

5. No sabe/No contesto

144

Knowledge

[Can you tell me the date when the next gubernatorial elections will take place in the
state?]

1. Mentioned exact date

2. Mentioned incomplete date

3. Does not remember

4. Does not know/Did not answer

Recoded: 1 (3+4)=Does not know; 2=Incomplete; 3=Complete

gMe podria decir la fecha en que se llevaran a cabo las elecciones para Gobemador del
estado. . .0 en este momento no recuerda?

1. Menciono fecha completa

2. Menciono fecha incompleta

3. No recuerda

4. No sabe/No contesto

Locality
[Type of locality: 1=Urban; 2=Rural]

Tipo de localidad:
1. Urbano
2. Rural

Education
[What is the highest level of education you have completed?]
1=No formal schooling
2. Primary schooling
3. Secondary/technical schooling
4. High school/equivalent
5. University and more
6. Did not answer
Recoded: 1=Secondary or less, 2=Preparatory; 3=University

gHasta que afio estudio usted? gCual es su ultimo grado de estudios?
1. No tiene estudios

2. Primaria

3. Secundaria/Técnica

4. Preparatoria/Equivalente

5. Universidad o mas

6. No contesté

Income
See Chiapas.

145

 

Appendix C. List of States

AGS
BC
BCS
CAMP
COAH
COL
CHIS
CHIH
DF
DGO
GTO
GRO
HGO
JAL
MEX
MICH
MOR
NAY
NL
OAX
PUE
QRO
QROO
SLP
SIN
SON
TAB
TAM
TLAX
VER
YUC
ZAC

State

Aguascalientes
Baja California
Baja California Sur
Campeche
Coahuila
Colima
Chiapas
Chihuahua
Distrito Federal
Durango
Guanajuato
Guerrero
Hidalgo
Jalisco

México
Michoacan
Morelos
Nayarit

Nuevo Leon
Oaxaca

Puebla
Querétaro
Quintana Roo
San Luis Potosi
Sinaloa

Sonora
Tabasco
Tamaulipas
Tlaxcala
Veracruz
Yucatan
Zacatecas

146

BIBLIOGRAPHY

Almond, Gabriel A. and Sidney Verba. 1963. The Civic Culture: Political Attitudes and
Democracy in Five Nations. Princeton, NJ: Princeton University Press.

Alt, James E. and Robert C. Lowry. 1994. “Divided Government, Fiscal Institutions, and

Budget Deficits: Evidence from the States.” American Political Science Review
88(4):811-828.

Alt, James E., David Dreyer Lassen, and David Skilling. 2002. “Fiscal Transparency,
Gubernatorial Approval, and the Scale of Government: Evidence from the States.”
State Politics and Policy Quarterly 2(3):230-250.

Ames, Barry. 1987. Political Survival: Politicians and Public Policy in Latin America.
Berkeley, CA: University of California Press.

Ames, Barry. 1970. “Bases de apoyo del partido dominante en México.” F oro
Internacional l l(1):50-76.

Anderson, Christopher J. 2000. “Economic Voting and Political Context: A Comparative
Perspective.” Electoral Studies 19:151-170.

Anderson, Leslie E. and Lawrence C. Dodd. 2005. Learning Democracy: Citizen
Engagement and Electoral Choice in Nicaragua, 1990-2001. Chicago, IL: The
University of Chicago Press.

Arellano Cadena, Rogelio (ed). 1996. Mexico Hacia un Nuevo F ederalismo Fiscal.
Mexico City: Fondo de Cultura Economica.

Atkeson, Lonna Rae and Randall W. Partin. 1995. “Economic and Referendum Voting: A

Comparison of Gubernatorial and Senatorial and Senatorial Elections.” American
Political Science Review 89(1):99-107.

Aziz Nassif, Alberto. 2000. Los ciclos de la democracia: Gobiemo y elecciones en
Chihuahua. Mexico City: Miguel Angel Porrua.

Aziz Nassif, Alberto. 1994. Chihuahua: Historial de una Alternativa. Mexico City: La
Jornada/CIESAS.

Bahl, Roy and Johannes Linn. 1986. “Public Expenditure Decentralization in Developing
Countries.” Government Policy 4:405-418.

Bailey, John. 1994. “Centralism and Political Change in Mexico: The Case of National
Solidarity,” in Cornelius, Wayne A., Ann L. Craig, and Jonathan Fox (eds.).
Transforming State-Society Relations in Mexico: The National Solidarity
Strategy. La Jolla, CA: Center for US-Mexican Studies, UCSD.

I47

Becerra, Ricardo, Pedro Salazar, and Jose’ Woldenberg. 2000. La mecdnica del cambio
politico en México: Elecciones, partidos y reformas. Mexico City: Ediciones Cal
y Arena.

Beer, Caroline C. 2003. Electoral Competition and Institutional Change in Mexico. Notre
Dame, IN: University of Notre Dame Press.

Beer, Caroline C. 2004. “Electoral Competition and Fiscal Decentralization in Mexico,”
in Alfred P. Montero and David J. Samuels (eds.). Decentralization and
Democracy in Latin America. Notre Dame, IN: University of Notre Dame Press.

Besley, Timothy and Anne Case. 1995. “Does Electoral Accountability Affect Economic

Policy Choices? Evidence from Gubernatorial Term Limits.” The Quarterly
Journal of Economics 1 10(3):769-798.

Bickers, Kenneth and Robert M. Stein. 1996. “The Electoral Dynamics of the Federal
Pork Barrel.” American Journal of Political Science 40(4): 1300-1326.

Bird, Richard M. and Francois Vaillancourt. 1998. Fiscal Decentralization in Developing
Countries. New York, NY: Cambridge University Press.

Bratton, Michael. 1998. “Second Elections in Africa.” Journal of Democracy 9(3):51-66.

Bratton, Michael, Robert Mattes, and E. Gyimah-Boadi. 2005. Public Opinion,
Democracy, and Market Reform in Afiica. New York, NY: Cambridge University
Press.

Bratton, Michael and Nicholas van de Walle. 1997. Democratic Experiments in Afiica:
Regime Transitions in Comparative Perspective. New York, NY: Cambridge
University Press.

Bruhn, Kathleen. 1997. Taking on Goliath: The Emergence of a New Left and the
Struggle for Democracy in Mexico. University Park, PN: The Pennsylvania State

University Press.

Buendia Laredo, Jorge. 1996. “Economic Reform, Public Opinion and Presidencial
Approval in Mexico, 1988-1993.” Comparative Political Studies 29:566-592.

Cabrero, Enrique. 1995. La nueva gestion municipal en México: Andlisis de experiencias
innovadoras en gobiernos locales. Mexico City: Miguel Angel Porrua/CIDE.

Camp, Roderic Ai. 2003. Politics in Mexico. The Democratic Transformation. New
York, NY: Oxford University Press.

Camp, Roderic Ai. 1996. Politics in Mexico. Second Edition. New York, NY: Oxford

148

University Press.
Camp, Roderic Ai. 1993. Politics in Mexico. New York, NY: Oxford University Press.

Camp, Roderic Ai. 1980. Mexico’s Leaders: Their Education and Recruitment. Tucson,
AZ: University of Arizona Press.

Carsey, Thomas M. and Gerald C. Wright. 1998. “State and National Factors in

Gubernatorial and Senatorial Elections.” American Journal of Political Science
42(3):994-1002.

Centro de Investigacion Para e1 Desarrollo, A. C. (CIDAC). 2005. Base de datos de
elecciones locales, 1980-2005. http://www.cidac.org

Chand, Vikram Khub. 2001. Mexico’s Political Awakening. Notre Dame, IN: Notre
Dame University Press.

Chappell, Henry W. Jr. and William R. Keech. 1985. “A New View of Political
Accountability for Economic Performance.” American Political Science Review
79:10-27.

Chubb John E. 1985. “The Political Economy of Federalism.” American Political Science
Review 79(4):994-1015.

Chubb, John E. 1988. “Institutions, the Economy, and the Dynamics of State Elections.”
American Political Science Review 82(1):81-154.

Clarke, Harold D. and Marianne C. Stewart. 1994. “Prospections, Retrospections, and
Rationality: The ‘Bankers’ Model of Presidential Approval Reconsidered.”
American Journal of Political Science 38(4):] 104-1 123.

Coleman, Kenneth M. 1975. “The capital city electorate and Mexico’s Accion Nacional:

Some survey evidence on conventional hypotheses.” Social Science Quarterly
(56).

Collier, David and Steven Levitsky. 1997. “Democracy with Adjectives: Conceptual
Innovation in Comparative Research.” World Politics 49(3):430-451.

Consejo Nacional de Poblacion (CONAPO). Proyecciones de poblacio'n a mitad de afio
por sexo y entidadfederativa, 1970-2030.

Cornelius, Wayne A. 1999. “Subnational Politics and Democratization: Tensions
Between Center and Periphery in the Mexican Political System,” in Subnational
Politics and Democratization in Mexico. Wayne A. Cornelius, Todd A.
Eisenstadt, and J. Hindley (eds.). La Jolla, CA: Center for US-Mexican Studies,
UCSD.

I49

Cornelius, Wayne A. 1969. “Urbanization as an Agent in Latin American Political
Instability: The case of Mexico.” American Political Science Review 63(3):833-
857.

Cornelius, Wayne A., and Ann L. Craig. 1991. The Mexican Political System in

Transition. Monograph Series 35. La Jolla, CA: Center for US-Mexican Studies,
UCSD.

Cornelius, Wayne A., Ann L. Craig, and Jonathan Fox (eds.). 1994. Transforming State-
Society Relations: The National Solidarity Strategy. US-Mexico Contemporary
Perspectives Series, 6. La Jolla, CA: Center for US-Mexican Studies, UCSD.

Craig, Ann L. and Wayne Cornelius. 1995. “Houses Divided: Parties and Political
Reform in Mexico,” in Scott Mainwaring and Timothy R. Scully (eds.). Building
Democratic Institutions: Party Systems in Latin America. Stanford, CA: Stanford
University Press.

Dahl, Robert A. 1971. Polyarchy: Participation and Opposition. New Haven, CT: Yale
University Press.

Dennis, Jack. 1991. “The Study of Electoral Behavior,” in William Grattz (ed.). Political
Science Looking to the Future. Political Behavior 3:51-89.

Diamond, Larry. 1999. Developing Democracy: Toward Consolidation. Baltimore and
London: The Johns Hopkins University Press.

Diamond, Larry and Leonardo Morlino (eds.). 2005. Assessing the Quality of
Democracy. Baltimore, MD: Johns Hopkins University Press.

Diamond, Larry and Doh Chull Shin. 2000. “Introduction: Institutional Reform and
Democratic Consolidation in Korea,” in Larry Diamond and Doh Chull Shin
(eds.). Institutional Reform and Democratic Consolidation in Korea. Stanford,
CA: Hoover Institution Press.

Diamond, Larry, Marc F. Plattner, and Andreas Schedler. 1999. “Introduction,” in
Andreas Schedler, Larry Diamond, and Marc F. Plattner (eds.). The Self-
Restraining State: Power and Accountability in New Democracies. Boulder, CO:
Lynne Rienner.

Diamond, Larry, Jonathan Hartlyn, Juan J. Linz, and Seymour Martin Lipset. 1999.
“Introduction: Politics, Society, and Democracy in Latin America,” in Larry
Diamond, Jonathan Hartlyn, Juan J. Linz, and Seymour Martin Lipset.
Democracy in Developing Countries: Latin America. Boulder, CO: Lynne
Rienner Publishers.

Diamond, Larry and Marc F. Plattner (eds.). 1994. Nationalism, Ethnic Conflict, and

150

 

 

Democracy. Baltimore, MD: Johns Hopkins University Press.

Diaz-Cayeros, Alberto. 1995. Desarrollo Econémico e Inequidad Regional: Hacia un
Nuevo Pacto Federal en Me'xico. Mexico City: F undacién Friedrich Naumann en
México, Centro de Investigacion para el Desarrollo, A.C., and Miguel Angel
Porrua.

Dominguez, Jorge I. and Chappell Lawson (eds.). 2004. Mexico 's Pivotal Democratic
Election. Candidates, Voters, and The Presidential Campaign of 2000. Stanford
and La Jolla, CA: Stanford University Press and Center for U.S.-Mexican Studies,
UCSD.

Dominguez, Jorge I. and Alejandro Poiré. 1999. Toward Mexico’s Democratization. New
York, NY and London: Routledge.

Dominguez, Jorge I. and James A. McCann. 1996. Democratizing Mexico. Public
Opinion and Electoral Choices. Baltimore: The Johns Hopkins Press.

Dominguez, Jorge I. and James A. McCann. 1995. “Shaping Mexico’s Electoral Arena:
The Construction of Partisan Cleavages in the 1988 and 1991 National Elections.”
American Political Science Review 89(1):34-48.

Dorussen, Han and Harvey D. Palmer. 2002. “The Context of Economic Voting: An
Introduction,” in Dorussen, Han and Michaell Taylor. Economic Voting. New

York, NY: Routledge/ECPR Studies in European Political Science.

Downs, Anthony. 1957. An Economic Theory of Democracy. New York, NY: Harper &
Brothers.

Duch, Raymond M. 2001. “A Developmental Model of Heterogeneous Economic Voting
in New Democracies.” American Political Science Review 95(4):895-910.

Easton, David. 1965. A Systems Analysis of Political Life. New York, NY: John Wiley.
Escobar-Lemmon, Maria. 2003. “Political Support for Decentralization: An Analysis of l
the Colombian and Venezuelan Legislatures.” American Journal of Political ‘

Science 47(4):683-697.

Eisenstadt, Todd. 2004. Courting Democracy in Mexico. Party Strategies and Electoral
Institutions. New York, NY: Cambridge University Press.

Erickson, RS. 1989. “Economic Conditions and the Presidential Vote.” American
Political Science Review 83(2): 567-573.

Espinoza Valle, Victor Alejandro. 1999. “Altemation and Political Liberalization: The

151

PAN in Baja California,” in Cornelius, Wayne A., Todd A. Eisenstadt and Jane

Hindley (eds.). Subnational Politics and Democratization in Mexico. La Jolla,
CA: Center for U.S.-Mexican Studies, UCSD.

Fagen, Richard R., and William S. Tuohy. 1972. Politics and Privilege in a Mexican
City. Stanford, CA: Stanford University Press.

F iorina, Morris P. 1981. Retrospective Voting in American National Elections. New
Haven, CT: Yale University Press.

Fiorina, Morris P. 1978. “Economic Retrospective Voting in American National
Elections: A Micro-Level Analysis.” American Journal of Political Science
22:246-243.

Fox, Jonathan. 1994. “The Difficult Transition from Clientelism to Citizenship: Lessons
from Mexico.” World Politics 46(2): 151-184.

Furtak, Robert K. 1969. “El Partido Revolucionario Institucional: integracion nacional y
movilizacion electoral.” F oro Internacional 9(2):339-352.

Garman, Christopher, Stephan Haggard, and Eliza Willis. 2001. “Fiscal Decentralization:
A Political Theory with Latin American Cases.” World Politics 53:205-236.

Garrido, Luis Javier. 1989. “The Crisis of Presidentialism,” in Wayne Cornelius, Judith
Gentleman, and Peter Smith (eds.). Mexico’s Alternative Political Futures. La
Jolla, CA: Center for US-Mexican Studies, UCSD.

Ge’lineau, Francois. 2002. “Economic Voting in Volatile Contexts: National and
Subnational Politics in Latin America.” Ph.D. diss. University of New Mexico.

Golden, David G. and James M. Poterba. 1980. “The Price of Popularity: The Political
Business Cycle Reexamined.” American Journal of Political Science 24(4):696-
714.

Gomez, Brad T. and J. Matthew Wilson. 2006. “Cognitive Heterogeneity and Economic
Voting: A Comparative Analysis of Four Democratic Electorates.” American
Journal of Political Science 50(1): 127-145.

Gomez, Brad T. and J. Matthew Wilson. 2001. “Political Sophistication and Economic
Voting in the American Electorate: A Theory of Heterogeneous Attribution.’
American Journal of Political Science 45(4):899-914.

9

Gomez-Tagle, Silvia. 1992. “Balance de las elecciones de 1991 en Mexico.” Revista
Mexicana de Sociologia 54(1):253-287.

Gomez-Tagle, Silvia. 1989. “La dificultad de perder: el partido oficial en la coyuntura de

152

1988.” Revista Mexicana de Sociologia 52(4): 239-260.
Gonzalez Casanova, Pablo. 1965. La Democracia en México. Mexico City: Era.

Gonzalez Graf, Jaime (ed.). 1988. Las Elecciones de 1988 y la Crisis del Sistema Politico
Mexicana. Mexico City: Instituto Mexicano de Estudios Politicos/Diana.

Gyimah-Boadi, E. 1998. “The Rebirth of African Liberalism.” Journal of Democracy
9(2): 1 8-3 1.

Haggard, Stephan and Steven Webb (eds.). 1994. Voting for Reform: Democracy,

Political Liberalization, and Economic Adjustment. New York, NY: Oxford
University Press/World Bank.

Hagopian, Frances. 1990. “Democracy by Undemocratic Means: Elites, Political Pacts,
and Regime Transition in Brazil.” Comparative Political Studies 23: 147-170.

Harper, Marcus A. G. 2000. “Economic Voting in Postcommunist Eastern Europe.”
Comparative Political Studies 33(9): 1 191-1227.

Hibbs, Douglas A., Jr., Douglas Rivers, and Nicholas Vasilatos. 1982. “On the Demand
for Economic Outcomes: Macroeconomic Performance and Mass Political
Support in the United States, Great Britain, and Germany.” The Journal of
Politics 44:426-462.

Hiskey, Jonathan. 2005. “The Political Economy of Subnational Economic Recovery in
Mexico.” Latin American Research Review 40(1):30-55.

Hiskey, Jonathan T. and Shaun Bowler. 2005. “Local Context and Democratization in
Mexico.” American Journal of Political Science 49(1):57-71.

Holbrook, Thomas M. and Emily Van Dunk. 1993. “Electoral Competition in the
American States.” American Political Science Review, 87(4): 955-962.

Holbrook-Provow, Thomas M. 1987. “National factors in gubernatorial elections.”
American Politics Quarterly 15:471-483.

Howell, Susan E., and J .M. Vanderleeuw. 1990. “Economic Effects on State Governors.”
American Politics Quarterly 15:471-483.

Huntington, Samuel. 1991. The Third Wave: Democratization in the Late Twentieth
Century. Norman, OK: University of Oklahoma Press.

Instituto Federal Electoral (IFE). Atlas Electoral Federal de México 1970-2003.

Instituto Nacional de Estadistica, Geografia e Informatica, INEGI. F inanzas Pziblicas

153

 

Estatales y Municipales de México, 1970-2004 (various issues). Aguascalientes,
Mexico: INEGI. http://www.inegi.gob.mx

Instituto Nacional de Estadistica, Geografia e Inforrnatica, INEGI. 2001. XII Censo
General de Poblacién y Vivienda. http://www.inegi.gob.mx

Jones, Mark P. 1995. Electoral Laws and the Survival of Presidential Democracies.
Notre Dame, IN: University of Notre Dame Press.

Jones, Mark P., Pablo Sanguinetti, and Mariano Tommassi. 1999. “Politics, Institutions,
and Public-Sector Spending in the Argentina Provinces,” in Poterba, James M.
and Jfirgen von Hagen (eds.). Fiscal Institutions and Fiscal Performance.
Chicago, IL: The University of Chicago Press.

Karl, Terry Lynn. 1990. “Dilemmas of Democratization in Latin America.” Comparative
Politics 22: 1-20.

Kaufman, Clifford. 1972. “Urbanization, material satisfaction and mass political
involvement: The poor in Mexico City.” Comparative Political Studies 4(3).

Key, V.O. Jr. 1966. The Responsible Electorate, Rationality in Presidential Elections.
Cambridge, MA: Harvard University Press.

Key, V.O. Jr. 1964. Politics, Parties, and Pressure Groups. New York, NY: Thomas Y.
Crowell.

Kiewiet, D. Roderick. 1981. “Policy-Oriented Voting in Response to Economic
Issues.” American Political Science Review 75:448-459.

Kinder, Donald R. and D. Roderick Kiewiet. 1978. “Economic Discontent and Political
Behavior: The Role of Personal Grievances and Collective Economic Judgments

in Congressional Voting.” American Journal of Political Science 23(3):495-527.

Kinder, Donald R. and D. Roderick Kiewiet. 1981. “Sociotropic Politics: The American
Case.” British Journal of Political Science 1 1:129-161.

King, J .D. 2001. “Incumbent popularity and vote choice in gubernatorial elections.”
Journal of Politics 63:585-597.

Klesner, Joseph L. 1995. “The 1994 Mexican elections: manifestation of a divided
society?” Mexican Studies/Estudios Mexicanos l 1(1):137-149.

Klesner, Joseph L. 1988. Electoral Reform in an Authoritarian Regime: The Case of
Mexico. Ph.D. diss., Massachusetts Institute of Technology.

Klorman, R. 1978. “Trend in Personal Finances and the Vote.” Public Opinion Quarterly

154

 

 

42:31-48.

Kramer, Gerald H. 1971. “Short-Tenn Fluctuations in US. Voting Behavior, 1986-
1964.” The American Political Science Review 65: 131-143.

Lawson, Chappell. 2002. Building the Fourth Estate: Democratization and the Rise of a
Free Press in Mexico. Berkeley, CA: University of California Press.

Levy, Daniel C. and Kathleen Bruhn, with Emilio Zebadua. 2001. Mexico. The Struggle
for Democratic Development. Berkeley, CA: University of California Press.

Lewis-Beck, Michael S. 1988. Economics and Elections: The Major Western
Democracies. Ann Arbor, MI: The University of Michigan Press.

Lewis-Beck, Michael S. and Mary Stegmaier. 2000. “Economic Determinants of
Electoral Outcomes.” Annual Review of Political Science 3: 183-219.

Lewis-Beck, Michael S. and Martin Paldam. 2000. “Economic Voting: An Introduction.”
Electoral Studies 19:1 13-121.

Linz, Juan J. 1990. “The Perils of Presidentialism.” Journal of Democracy 1(1):51-69.

Linz, Juan J. and Alfred Stepan. 1996. Problems of Democratic Transition and
Consolidation: Southern Europe, South America, and Post-Communist Europe.
Baltimore, MD: Johns Hopkins University Press.

Lowry, Robert C., James E. Alt, and Karen E. Ferree. 1998. “Fiscal Policy Outcomes and
Electoral Accountability in American States.” American Political Science Review
92(4):759-774.

Lujambio, Alonso. 2000. El Poder Compartido. Mexico City: Ed. Océano.

MacKuen, Michael B., Robet S. Erikson, and James A. Stimson. 1989.
“Macropartisanship.” American Political Science Review 83(4)21125-1142.

MacKuen, Michael B., Robert S. Erikson, and James S. Stimson. 1992. “Peasants or
Bankers? The American Electorate and the US. Economy.” American Political
Science Review 86(3):597-611.

Magaloni, Beatriz. 1997. The Dynamics of Ruling Party Decline: The Mexican
Transition to Multipartism. Ph.D. diss., Duke University.

Magaloni, Beatriz. 1994. “Eleccion racional y voto estratégico: Algunas aplicaciones
para el caso mexicano.” Politica y Gobiemo l(2):309-344.

Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave of

155

Democratization. The Case of Brazil. Stanford, CA: Stanford University Press.

Mainwaring, Scott. 1993. “Presidentialism, Multipartism, and Democracy: The Difficult
Combination.” Comparative Political Studies 26(2):]198-1228.

Manin, Bernard, Adam Przeworski, and Susan C. Stokes. 1999. “Elections and
Representation,” in Adam Przeworski, Susan C. Stokes, and Bernard Manin
(eds.). Democracy, Accountability, and Representation. New York, NY:
Cambridge University Press.

Markowski, Radoslaw, and Gabor Toka. 1998. “Left Turn in Poland and Hungary Five
Years after the Collapse of Communism.” Sisyphus 1(9):75-99.

Markus, Gregory B. 1988. “The Impact of Personal and National Economic Conditions
on the Presidential Vote: A Pooled Cross-Sectional Analysis.” American Journal
of Political Science 32: 137-154.

Martinez Assad, Carlos (ed.). 1992. La Sucesién Presidencial en México, 1928-1988.
Mexico City: Nueva Imagen.

Merino, Mauricio (ed.). 1994. En busca de la democracia municipal. La participacio'n
ciudadana en el gobiemo local mexicano. Mexico City: El Colegio de México.

Meyer, Lorenzo, and Hector Aguilar Camin. 1993. In the Shadow of the Mexican
Revolution: Contemporary Mexican History, 1910-1989. Austin, TX: University
of Texas Press.

Meyer, Lorenzo. 1986. “Un tema afiejo siempre actual: El centro y las regiones en la
historia mexicana,” in Blanca Torres (ed.). Descentralizacion y Democracia en
México. Mexico City: El Colegio de Mexico.

Middlebrook, Kevin. 1986. “Political Liberalization in an Authoritarian Regime: The
Case of Mexico,” in Paul W. Drake and E. Silva (eds.). Elections and
Democratization in Latin America, 1985-1985. La Jolla, CA: Center for US-
Mexican Studies, UCSD.

Middlebrook, Kevin J. (ed.). 2001. Party Politics and the Struggle for Democracy in

Mexico. National and State-Level Analyses of the Partido Accion Nacional. La
Jolla, CA: Center for U.S.-Mexican Studies, UCSD.

Mishler, William, and Richard Rose. 1996. “Trajectories of Fear and Hope: Support for
Democracy in Post-Communist Europe.” Comparative Political Studies

28(4):553-581.

Mizrahi, Yemile. 2003. From Martyrdom to Power: The Partido Accion Nacional in
Mexico. Notre Dame, IN: University of Notre Dame Press.

156

Mizrahi, Yemile. 1995. “Entrepreneurs in the Opposition: Modes of Political
Participation in Chihuahua,” in Rodriguez, Victoria E. and Peter M. Ward.
Opposition Government in Mexico. Alburqueque, NM: University of New Mexico
Press.

Molina, J osé. 2001. “The electoral effect of underdevelopment: government turnover and
its causes in Latin-American, Caribbean and industrialized countries.” Electoral
Studies 20:427-446.

Molinar Horcasistas, Juan. 1986. “The Mexican electoral system: continuity by change,”
in Paul W. Drake and Eduardo Silva (eds.). Elections and democratization in
Latin America, [980-1985. La Jolla, CA: Center for Iberian and Latin American
Studies, UCSD.

Molinar Horcasitas, Juan, and Jeffrey Weldon. 1990. “Elecciones de 1988 en Mexico:
crisis del autoritarismo.” Revista Mexicana de Sociologia 52(4):229-262.

Monsivais, Carlos. 1992. “‘Just Over that Hill’: Notes on Centralism and Regional
Cultures,” in Eric Van Young (ed.). Mexico ’s Regions: Comparative History and
Development. La Jolla, CA: Center for U.S.-Mexican Studies, UCSD.

Montero, Alfred P. and David J. Samuels (eds.). 2004. Decentralization and Democracy
in Latin America. Notre Dame, IN: University of Notre Dame Press.

Moreno, Alejandro. 2003. El Votante Mexicana. Democracia, Actitudes Politicas y
Conducta Electoral. Mexico City: Fondo de Cultura Econémica.

Moreno, Alejandro. 1993. “Agosto de 1991: éPor qué se voto por el PRI?” Este Pat's.
(Dec):26-28.

Moreno, Alejandro and Keith Yanner. 1995. “Predictors of voter preferences in Mexico’s
1994 presidential elections.” Paper presented at the Latin American Studies
Association Meeting, Washington, DC.

Morgenstem, Scott and Elizabeth Zechmeister. 2001. “Better the Devil you know than

the Saint you Don’t? Risk Propensity and Vote Choice in Mexico.” Journal of
Politics 63(1):93-119.

Muller, Edward N. and Mitchell A. Seligson. 1994. “Civic Culture and Democracy: The
Question of Causal Relationships.” American Political Science Review 88(3):635-
652.

Nadeau, Richard, Richard G. Niemi, and Antoine Yoshinaka. 2002. “A cross-national

analysis of economic voting: taking account of the political context across time
and nations.” Electoral Studies 21:403-423.

157

Nannestad, P. and M. Paldam. 1994. “The VP-Function: A Survey of the Literature on
Vote and Popularity Functions after 25 Years.” Public Choice 79:213-245.

Niemi, Richard G., Harold W. Stanley, and Ronald J. Vogel. 1995. “State Economies and
State Taxes: Do Voters Hold Governors Accountable?” American Journal of
Political Science 39(4):936-957.

Nickson, R. Andrew. 1995. Local Government in Latin America. Boulder, CO: Lynne
Rienner Publishers.

Norpoth, Helmut. 1996. “Presidents and the Prospective Voter.” The Journal of Politics
58(3):776-792.

Norpoth, Helmut. 1984. Economics, Politics and the Cycle of Presidential Popularity.
Political Behavior 6:252-273.

O’Donnell, Guillermo. 1994. “Delegative Democracy.” Journal of Democracy 5(1):55-
69.

O’Donnell, Guillermo, Philippe C. Schmitter, and Laurence Whitehead (eds.). 1986.
Transitions from Authoritarian Rule. Latin America. Baltimore, MD: The Johns
Hopkins University Press.

Pacek, Alexander, and Benjamin Radcliff. 1995. “The Political Economy of Competitive
Elections in the Developing World.” American Journal of Political Science
39(3):745-759.

Pacheco, Guadalupe. 1986. El PR1 y los procesos electorales de 1961 a 1985. Mexico
City: UAM-Xochimilco.

Partin, R. 1995 “Economic Conditions and Gubernatorial Elections: Is the State
Executive Held Accountable?” American Politics Quarterly 23:81-95.

Peltzman, Sam. 1992. “Voters as Fiscal Conservatives.” Quarterly Journal of Economics
107(2):327-361.

Peltzman, Sam. 1990. “How Efficient is the Voting Market?” Journal of Law &
Economics 33:27-63.

Peltzman, Sam. 1987. “Economic Conditions and Gubernatorial Elections.” American
Economic Review 77:293-297.

Piereson, James E. 1975. “Presidential Popularity and Midterm Voting at Different
Electoral Levels.” American Journal of Political Science 29(4):683-694.

158

Popkin, Samuel L. 1991. The Reasoning Voter: Communication and Persuasion in
Presidential Campaigns. Chicago, IL: University of Chicago Press.

Posner, Daniel N. and David J. Simon. 2002. “Economic Conditions and Incumbent

Support in Africa’s New Democracies. Evidence from Zambia.” Comparative
Political Studies 35(3):313-336.

Poterba, James M. 1994. “State Responses to Fiscal Crises: The Effects of Budgetary
Institutions and Politics.” Journal of Political Economy 102(4):799-821.

Poterba, James M. and Jiirgen von Hagen (eds.). 1999. Fiscal Institutions and Fiscal
Performance. Chicago, IL: The University of Chicago Press.

Powell, Bingham G., Jr. and Guy D. Whitten. 1993. “A Cross-National Analysis of
Economic Voting: Taking Account of the Political Context.” American Journal of
Political Science 37(2):391-414.

Prud’homme, Rémy. 1995. “The Dangers of Decentralization.” The World Bank
Observer 10:201-220.

Przeworski, Adam. 1996. “Public Support for Economic Reforms in Poland.”
Comparative Political Studies 29:520-543.

Przeworski, Adam, Stokes and Manin (eds.). 1999. Democracy, Accountability, and
Representation. New York, NY: Cambridge University Press.

Putnam, Robert D. with Robert Leonardi and Raffaella Y. Nanetti. 1993. Making
Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton
University Press.

Quinn, Dennis P. And John T. Woolley. 2001. “Democracy and National Economic
Performance: The Preference for Stability.” American Journal of Political Science
45(3):634-657.

Ranney, Austin. 1976. “Parties in State Politics,” in Jacob, Herbert and Kenneth Vines
(eds.). Politics in the American States. Boston, MA: Little Brown & Co.

Remmer, Karen L. 1993. “The Political Economy of Elections in Latin America, 1980-
1991.” American Political Science Review 87(2):393-407.

Remmer, Karen L. 1991. “The Political Impact of Economic Crisis in Latin America.”
American Political Science Review 85:777-800.

Remmer, Karen L. And Erik Wibbels. 2000. “The Subnational Politics of Economic

Adjustment. Provincial Politics and Fiscal Performance in Argentina.”
Comparative Political Studies 33(4):419-45 1 .

159

Remmer, Karen L. and Francois Gélineau. 2003. “Subnational Electoral Choice.
Economic and Referendum Voting in Argentina, 1983-1999.” Comparative
Political Studies 36(7):801-821 .

Rezk, Ernesto. 1998. “Argentina: Fiscal Federalism and Decentralization,” in Richard M.
Bird and Francois Vaillancourt (eds.). Fiscal Decentralization in Developing
Countries. New York, NY: Cambridge University Press.

Roberts, Bryan. 1992. “The Place of Regions in Mexico,” in Eric Van Young (ed.).
Mexico ’s Regions: Comparative History and Development. La Jolla, CA: Center
for U.S.-Mexican Studies, University of California, San Diego.

Roberts, Kenneth M. and Eric Wibbels. 1999. “Party Systems and Electoral Volatility in
Latin America: A Test of Economic, Institucional, and Structural Explanations.”
American Political Science Review 93(3):575-590.

Rodden, Jonathan and Eric Wibbels. 2002. “Beyond the Fiction of Federalism.
Macroeconomic Management in Multitiered Systems.” World Politics 54:494-
531.

Rodriguez, Victoria. 1997. Decentralization in Mexico: From Reforma Municipal to
Solidaridad to Nuevo F ederalismo. Boulder, CO: Westview Press.

Rodriguez, Victoria E. and Peter Ward . 1996. New Federalism, State and Local
Government in Mexico. Austin, TX: Mexican Center of the Institute of Latin
American Studies, University of Texas at Austin.

Rodriguez, Victoria E. and Peter Ward (eds.) 1995. Opposition Government in Mexico.
Albuquerque, NM: University of New Mexico.

Rodriguez, Victoria E. and Peter Ward. 1994. “Disentangling the PRI from the
Government in Mexico.” Mexican Studies/Estudios Mexicanos 10(1): 163-186.

Rondinelli, Dennis. 1989. “Decentralizing Public Services in Developing Countries:
Issues and Opportunities.” Journal of Social, Political, and Economic Studies
14(1):77-98.

Rondinelli, Dennis. 1981. “Government Decentralization in Comparative Perspective.”
International Review of Administrative Sciences 2: 133-145.

Rose, Richard, William Mishler, and Christian Haerpher. 1998. Democracy and its
Alternatives: Understanding Post-Communist Societies. Baltimore, MD: Johns

Hopkins University Press.

Rowland, Allison M. 2001. “Population as a Determinant of Local Outcomes under

160

Decentralization: Illustrations from Small Municipalities in Bolivia and Mexico.”
World Development 29(8): 1373-1389.

Samuels, David. 2003. “Presidents, Assemblies, and Accountability.” Paper presented at
the Annual Meeting of the American Political Science Association, Philadelphia.

Sartori, Giovanni. 1970. “Concept Misforrnation in Comparative Politics,” American
Political Science Review 64(4): 1033-1053.

Schattschneider, Elmer Eric. 1942. Party Government. New York, NY: Farrar and
Rinehart.

Schedler, Andreas. 1999. “Conceptualizing Accountability,” in Andreas Schedler, Larry
Diamond, and Marc F. Plattner (eds.). The Self-Restraining State: Power and
Accountability in New Democracies. Boulder, CO: Lynne Rienner.

Schmitter, Philippe C. and Terry Lynn Karl. 1991. “What Democracy Is. . .and Is Not.”
Journal of Democracy 2(3):78-

Schmitter, Philippe C. 2004. “The Ambiguous Virtues of Accountability.” Journal of
Democracy 15(4):47-60.

Schmitter, Philippe. 1997. “Civil Society East and West,” in Larry Diamond, Marc F.
Plattner, Yun-han Chu, and Hung-mao Tien (eds.). Consolidating the Third Wave
Democracies: Themes and Perspectives. Baltimore, MD: Johns Hopkins
University Press.

Segovia, Rafael. 1975. La Politizacién del Nifio Mexicana. Mexico City: El Colegio de
México, Coleccién Centro de Estudios Internacionales (l4).

Seong, Kyoung-Ryung. 2000. “Delayed Decentralization and Incomplete Democratic
Consolidation,” in Larry Diamond and Doh Chull Shin (eds.). Institutional
Reform and Democratic Consolidation in Korea. Stanford, CA: Hoover
Institution Press.

Shugart, Matthew Soberg and John M. Carey. 1992. Presidents and Assemblies:
Constitutional Design and Electoral Dynamics. New York, NY: Cambridge
University Press.

Simon, Dennis M. 1989. “Presidents, Governors, and Electoral Accountability.” Journal
of Politics 51:286-304.

Simon, Dennis M., Charles W. Ostrom, and Robin F. Marra. 1991. “The President,

Referendum Voting, and Subnational Elections in the United States.” American
Political Science Review 85:1 177-1 192.

161

Smith, Peter. 1979. Labyrinths of Power: Political Recruitment in Twentieth Century
Mexico. Princeton, NJ: Princeton University Press.

Solt, Frederick. 2004. “Electoral Competition, Legislative Pluralism, and Institutional
Development: Evidence from Mexico’s States.” Latin American Research Review
39(1):]55-167.

Solt, Frederick. 2003. “Explaining the Quality of New Democracies: Actors, Institutions,
and Socioeconomic Structure in Mexico’s States.” Ph.D. diss. University of North
Carolina at Chapel Hill.

Stein, Robert M. 1990. “Economic Voting for Governor and US. Senator: The Electoral
Consequences of Federalism.” Journal of Politics 52(1):29-53.

Stepan, Alfred. 2000. “Brazil’s Decentralize Federalism: Bringing Government Closer to
the Citizens?” Daedalus 129(2): 145-169.

Svoboda, Craig J. 1995. “Retrospective Voting in Gubernatorial Elections: 1982 and
1986.” Political Research Quarterly 48: 135-150.

Tiebout, Charles M. 1956. “A Pure Theory of Local Expenditures.” Journal of Political
Economy 64:416-424. '

Tomasen, Jacques and Hermann Schmitt. 1997. “Policy Representation.” European
Journal of Political Research 32:165-184.

Tufte, Edward R. 1978. Political Control of the Economy. Princeton, NJ: Princeton
University Press.

United Nations Development Program. 1993. “People and Governance,” in Human
Development. New York, NY: Oxford University Press.

Vargas Llosa, Mario. 2000. “De la Dictadura Perfecta a una Democracia Dificil.”
Reforma. July 3:IIA.

Ward, Peter M. 1993. “Social Welfare Policy and Political Opening in Mexico.” Journal
of Latin American Studies 25:613-628.

Ward, Peter M. 1995. “Policy Making and Policy Implementation Among Non-PRI
Governments: The PAN in Ciudad Juarez and in Chihuahua,” in Peter M. Ward
and Victoria E. Rodriguez (eds.). Opposition Government in Mexico.
Alburqueque: University of New Mexico Press.

Ward, Peter M. and Victoria E. Rodriguez. 1999. “New Federalism, Intra-govemmental

Relations and Co-govemance in Mexico.” Journal of Latin American Studies
31:673-710.

162

Weatherford, M. Stephen. 1987. “How Does Government Performance Influence
Political Support?” Political Behavior 9(1):5-28.

Weldon, Jeffrey A. 1997. “The Political Sources of Presidencialismo in Mexico,” in
Mainwaring, Scott and Matthew Shugart (eds.). Presidentialism in Latin America.

New York, NY: Cambridge University Press.

Weyland, Kurt. 2003. “Economic Voting Reconsidered. Crisis and Charisma in the
Election of Hugo Chavez.” Comparative Political Studies 36(7):822-848.

Wibbels, Erik. 2000. “Federalism and the Politics of Macroeconomic Policy and
Performance.” American Journal of Political Science 44(4):687-702.

Wilkin, Sam, Brandon Haller, and Helmut Norpoth. 1997. “From Argentina to Zambia: a
World-Wide Test of Economic Voting.” Electoral Studies 16(3):301-316.

Willis, Eliza, Christopher da CB, and Stephan Haggard. 1999. “The Politics of
Decentralization in Latin America.” Latin American Research Review 34(1):7-56.

Zavala, Ivan. 1991. “Factores sociales de la votacién por Carlos Salinas de Gortari.”
Estudios Politicos 8:43-54. ‘

163

     

GA

uliltiltglltljlflltjlllwill?"