American Political Science Review Vol. 106, No. 3 August 2012
doi:10.1017/S0003055412000287
Inequality and Regime Change:
Democratic Transitions and the Stability of Democratic Rule
STEPHAN HAGGARD University of California at San Diego
ROBERT R. KAUFMAN Rutgers University
R
ecent work by Carles Boix and Daron Acemoglu and James Robinson has focused on the role
of inequality and distributive conflict in transitions to and from democratic rule. We assess these
claims through causal process observation, using an original qualitative dataset on democratic
transitions and reversions during the “third wave” from 1980 to 2000. We show that distributive conflict,
a key causal mechanism in these theories, is present in just over half of all transition cases. Against
theoretical expectations, a substantial number of these transitions occur in countries with high levels of
inequality. Less than a third of all reversions are driven by distributive conflicts between elites and masses.
We suggest a variety of alternative causal pathways to both transitions and reversions.
A
re inequality and distributive conflicts a driving
force in the transition to democratic rule? Are
unequal democracies more likely to revert to
authoritarianism? These questions have a long pedigree
in in the analysis of the transition to democratic
rule in Europe (Lipset 1960; Marshall 1963; Moore
1966), and have been raised again in newer comparative
historical work on democratization (Collier 1999;
Rueschemeyer, Stephens, and Stephens 1992). More
recently, an influential line of theory has attempted
to ground the politics of inequality on rationalist assumptions
about citizens’ preferences over institutions
(Acemoglu and Robinson 2000; 2001; 2006; Boix 2003;
2008; Przeworski 2009). These distributive conflict approaches
conceptualize authoritarian rule as an institutional
means through which unequal class or group
relations are sustained by limiting the franchise and the
ability of social groups to organize. The rise and fall of
democratic rule thus reflect deeper conflicts between
elites and masses over the distribution of wealth and
income.
Despite its logic, there are several theoretical and
empirical reasons to question the expectations of these
new distributive conflict models. Socioeconomic inequality
plays a central role in these models, but has
cross-cutting effects. The more unequal a society, the
greater the incentives for disadvantaged groups to
press for more open and competitive politics. Yet the
Stephan Haggard is Lawrence and Sallye Krause Distinguished Professor,
Graduate School of International Relations and Pacific Studies,
University of California at San Diego, 9500 Gilman Drive, La
Jolla, CA 92093 (shaggard@ucsd.edu).
Robert R. Kaufman is Professor, Department of Political Science,
Rutgers University, 89 George Street, New Brunswick, NJ 08901
(kaufrutger@aol.com).
The authors thank Carles Boix, Michael Bratton, T.J. Cheng,
Ruth Collier, Javier Corrales, Ellen Commisso, Sharon Crasnow,
Anna Grzymala-Busse, Allan Hicken, Jan Kubik, James Long, Irfan
Noorudin, Grigore Pop-Eleches, Celeste Raymond, Andrew
Schrank, and Nic VanDewalle for comments on earlier drafts, including
on the construction of the dataset. We received useful feedback
from a presentation at the Watson Institute, Brown University, and
comments from Ronald Rogowski and anonymous reviewers of the
APSR. Particular thanks as well to Christian Houle for making his
dataset available. We also thank Vincent Greco, Terence Teo, and
Steve Weymouth for research assistance.
wider the income disparities in society, the more elites
have to fear from the transition to democratic rule and
the greater the incentives to repress challenges from
below. Given this potential indeterminacy, theoretical
models have hinged on a variety of other parameters,
such as the cost of repression or the mobility of assets.
Even with these refinements, attempts to demonstrate
the relationship between inequality and regime
type have yielded only mixed results. In cross-section,
there is a relationship between income distribution and
the level of democracy: Ceteris paribus, more equal societies
are more democratic. Yet the causal relationship
between inequality and either transitions to democratic
rule or reversions from it is much less robust.
We focus on regime change during the “third wave”
of democratic transitions from 1980–2000. This period
was marked by the spread of democracy to a wide range
of developing and postsocialist countries. These included
not only middle-income nations in Latin America,
Eastern Europe, and East and Southeast Asia but
also a substantial number of lower income countries, including
in Africa (Bratton and van de Walle 1997). Although
democratic transitions outnumber reversions
from democratic rule, the period also saw a number of
transitions to authoritarian rule.
Not only does this temporal focus on the third wave
capture a wide-ranging sample of regime changes but it
also overlaps with important changes in international
context. During the Cold War era, both right- and leftwing
dictators could exploit great power rivalries to win
support from external patrons. During the 1980s and
1990s, the decline and ultimate collapse of the Soviet
Union created a much more permissive international
environment for democratic rule (Boix 2011).
Using an extremely generous definition of “distributive
conflict” transitions, we find that between 55% and
58% of the democratic transitions during this period
conformed—even very loosely—to the causal mechanisms
specified in the distributive conflict models. Thus,
even with an expansive definition of distributive conflict,
more than 40% did not conform at all. Moreover,
a substantial number of the distributive conflict transitions
occurred under conditions of high inequality,
a result that is at odds with the expectations of the
495
Inequality and Regime Change August 2012
theory. Approximately 30% of all transitions occurred
in countries that ranked in the top tercile in terms of inequality,
and a substantial majority of these transitions
resulted from distributive conflict; this finding is robust
to alternative measures of inequality. These findings do
not necessarily overturn distributive conflict theories,
but suggest that they are underspecified with respect
to scope conditions and only operate under very particular
circumstances.
Given the substantial incidence of nondistributive
conflict transitions, we find several alternative causal
pathways to democratic rule. External actors were decisive
in some cases. In many cases, however, other domestic
causal factors induced incumbents to relinquish
power in the absence of strong challenges from below.
Elite incumbents were sometimes challenged by elite
outgroups or defectors from the ruling coalition who
saw gains from democratic openings. In other cases,
elite incumbents ceded power in the absence of mass
pressure because they believed they could control the
design of democratic institutions in ways that protected
their material interests.
An even smaller percentage of reversions—less than
a third—conformed to the elite-mass dynamics postulated
in the theory, and once again, we found little relationship
between the incidence of these transitions and
socioeconomic inequality. However, we did find several
alternative causal mechanisms. In several cases, incumbent
democratic governments were overthrown not by
socioeconomic elites seeking to block redistribution,
but by authoritarian populist leaders promising more
redistribution. Even more commonly, however, reversions
were driven by conflicts that either cut across class
lines or arose from purely intra-elite conflicts, particularly
conflicts in which factions of the military staged
coups against incumbent office holders.
Our analysis is motivated by methodological as well
as substantive concerns. In contrast to quantitative
tests of the relationship between inequality and regime
change, we have constructed a qualitative dataset
of within-case causal process observations (Haggard,
Kaufman, and Teo 2012). Our approach differs from
other such designs in that it examines all discrete
country-years that have been coded as transitions or
reversions in two prominent datasets: Polity IV and the
dichotomous coding scheme developed by Przeworski
et al. (2000) and extended by Cheibub, Ghandi, and
Vreeland (2010).
Critics of “medium-N” designs have argued that
such designs lack both the detail of individual case
studies or smaller-N designs and the precision of wellspecified
larger-N econometric models. Yet we argue
that they are particularly useful for evaluating whether
the causal mechanisms stipulated in formal models—
which typically involve complex sequences of strategic
interactions—are in fact present in the cases. The
approach is particularly useful for testing theories of
relatively rare events, such as democratic transitions
and reversions, civil wars, genocides, financial crises,
and famines. In cross-national quantitative models of
these phenomena, the number of country-years in the
panel is large, but the number of cases to be explained
is limited, permitting more intensive treatment of the
relevant cases and thus more robust inference.
We begin in the first section by reviewing distributive
conflict models of regime change, focusing on the
contributions by Boix (2003; 2008), Acemoglu and
Robinson (2000; 2001; 2006), and Przeworski (2009).
The next section discusses methodological issues. The
remainder of the article is structured around a consideration
of transitions to democracy and reversions
to authoritarian rule. Our causal process observations
show not only that transitions occur across cases with
very different levels of inequality—as the null findings
in econometric models already attest—but also that a
large number of democratic transitions and reversions
occur in the absence of significant redistributive conflict
altogether.
The returns from this exercise are both substantive
and methodological. First, the findings cast doubt
on the prevalence of the core causal mechanisms at
work in the underlying model, including the relationship
between inequality and particular types of elite
and mass behavior. In the conclusion, we raise questions
about alternative approaches and suggest several
ways in which the theory might be modified: There
may be other channels through which inequality can
destabilize democratic rule, and there might be other
economic and institutional factors that condition the
capacity of low-income groups to engage in collective
action. Second, our methodological contribution raises
important questions about the validity of reduced-form
panel designs, including with respect to the coding of
regime type itself. More positively, it suggests a fruitful
way of combining quantitative and qualitative methods
that focuses attention on alternative transition paths
rather than the partial-equilibrium treatment effects
of favored variables.
THEORY
Adam Przeworski (2009, 291) poses the puzzle of
democratic transitions in the clearest terms: “Why
would people who monopolize political power ever
decide to put their interests or values at risk by sharing
it with others? Specifically, why would those who
hold political rights in the form of suffrage decide to
extend these rights to anyone else?” The seminal work
of Meltzer and Richard (1981) provides the point of
departure for all current distributive conflict models of
regime change.1
The Meltzer-Richard model posits that
the distribution of productivity and income is skewed
to the right, with most citizens falling at the lower and
middle range of the distribution and a smaller tail constituting
the rich; the mean income exceeds the median.
Where voting rules result in appeals to the median
voter, the wider the divergence between the median
and mean income, the more is to be gained from redistribution.
Put differently, in countries with more
skewed income distributions, the poor have more to
gain from redistribution and should have more generous
tax and transfer programs as a result.
1 See also Romer (1975) and Roberts (1977).
496
American Political Science Review Vol. 106, No. 3
In the distributive conflict theories of regime change,
most notably in the work of Boix (2003) and Acemoglu
and Robinson (2000; 2001; 2006), these expectations
are modified and expanded to endogenize the very
existence of democratic governments. These models
differ in ways we explicate later, but both rest on complex
causal chains including both structural and gametheoretic
components: inequality, distributive conflict,
and strategic interactions between incumbents and oppositions
over the nature of political institutions. In
models of democratic transitions, low-income groups—
sometimes in coalition with middle-class forces—mobilize
in favor of redistribution and against the authoritarian
institutions that sustain inequalities. These theories
are vague about how collective action problems are
solved, but posit that they can be overcome by changes
in information with respect to the solidity of incumbent
power (Boix) or by increasing returns from mobilization
as inequality rises (Acemoglu and Robinson).
Faced with the threat of being displaced by force—in effect,
through revolution—elites calculate the net cost of
repression vs. concession, including institutional ones.
At very high levels of inequality, the threats posed by
democratization are too high to accept and they choose
to repress. Yet at low or medium levels of inequality,
redistributive demands can be managed through class
compromises over institutions and policy that permit
democratic transitions.
Carles Boix’s (2003) Democracy and Redistribution
is a significant exemplar of this broad approach. Boix
defines a right-wing authoritarian regime as one in
which the political exclusion of the poor sustains existing
economic inequalities. According to Boix (2003, 37)
“a more unequal distribution of wealth increases the redistributive
demands of the population. . .. [However]
as the potential level of transfers becomes larger,
the authoritarian inclinations of the wealthy increase
and the probabilities of democratization and democratic
stability decline steadily.” The translation of
these demands into a change in institutions hinges on
the balance of power between the wealthy and the
poor. Boix offers an informational model in which
regime changes are triggered by exogenous shocks that
weaken the elite or reveal its weakness (28–30). A
necessary (although not sufficient) mechanism driving
regime change is pressure from below: “As the least
well off overcome their collective action problems, that
is, as they mobilize and organize in unions and political
parties, the repression cost incurred by the wealthy
rise[s],” forcing elites to make institutional compromises
(13).
Boix also emphasizes the role played by capital mobility
in mitigating this relationship (see also Freeman
and Quinn 2012). High levels of capital mobility enhance
the bargaining power of elites. Fixed assets, by
contrast, limit the options of the wealthy and make
them vulnerable to democratic redistribution and thus
more resistant to it. Given the decision of the poor to
mobilize, the incentives of upper-class incumbents to
repress are a function of the level of inequality and
mobility of assets. Transitions are most likely when
inequality is low, asset mobility is high, and elites
have less to lose from competitive politics. Elites have
stronger incentives to repress as inequality increases
and when assets are fixed.
Although broadly similar in spirit, Acemoglu and
Robinson’s (2006) Economic Origins of Dictatorship
and Democracy introduces several innovations. Acemoglu
and Robinson concur with Boix that regime type
is a function of the balance of power between highand
low-income groups. Although elites monopolize
de jure power, masses potentially wield de facto power
through their capacity to mobilize against the regime.
Like Boix, Acemoglu and Robinson elaborate more
complex “three-class” models in which the pressure on
elites comes from coalitions of middle- and low-income
groups. However, the establishment of mass democracy
presupposes the engagement of low-income sectors because
of their sheer weight. Because they constitute the
majority, masses can sometimes “challenge the system,
create significant social unrest and turbulence, or even
pose a serious revolutionary threat” (25).
High inequality increases the incentives for authoritarian
elites to repress these political demands
for redistribution. To this observation, Acemoglu and
Robinson add an important point about credible commitments.
When elites are confronted by mobilization
from below, they can make short-run economic concessions
to diffuse the threat. Yet politically and economically
excluded groups are aware that elites can renege
on these concessions when pressures from below subside.
Because there is a cost to subsequently reversing
democracy after a transition has occurred, democratic
institutions provide a means for elites to credibly commit
to a more equal distribution of resources not only
in the present but into the future as well.
Acemoglu and Robinson agree with Boix that, although
inequality increases the incentive for excluded
groups to press for democracy, it also increases elite
incentives to repress. High inequality is inauspicious
for democracy. However, Acemoglu and Robinson argue
that democratization is also unlikely to occur in
authoritarian governments with low levels of inequality
because the demand for it is also attenuated; despite
political restrictions, excluded groups nonetheless
share in the distribution of societal income. They
conclude that the relationship between inequality and
democratic transitions should exhibit an inverted-U
pattern, with transitions to democratic rule most likely
to occur at intermediate levels of inequality.
It is important to emphasize that the theory is not
simply a structural one but operates through strategic
interactions between elites and masses: incentives for
collective action on the part of the masses and repression
or concessions on the part of elites. At middle levels
of inequality, grievances are sufficient to motivate
the disenfranchised to mobilize, but not threatening
enough to invite repression (see also Burkhart 1997;
Epstein et al. 2006).
Boix (2003) and Acemoglu and Robinson (2006)
extend their arguments to a consideration of the stability
of democratic rule and reversion to autocracy
as well. Implicit in the theory is the assumption that
high-inequality democracies are rare; for that reason,
497
Inequality and Regime Change August 2012
less attention is given to the relationship between inequality
and democratic breakdown. Nonetheless Boix
postulates a direct linear relation between the degree
of inequality and the likelihood of reversion to dictatorship.
In high-inequality democracies, redistributive
pressures from lower-class groups will be more intense,
motivating elites to deploy force against incumbents in
order to reimpose authoritarian rule. Although Acemoglu
and Robinson posit an inverted-U shaped relation
between inequality and democratic transitions,
they agree that countries that do manage to democratize
at high levels of inequality “do not consolidate
because coups are attractive” (38). The costs for elites
of mobilizing against democratic rule are less than
the losses arising from redistribution under democratic
rule.
METHOD:
CAUSAL PROCESS OBSERVATIONS
At first glance, these theories appear amenable to relatively
straightforward tests. Is the level of inequality
associated with transitions to and from democratic rule
or not? Yet empirical tests are complicated by the fact
that different measures of inequality capture different
socioeconomic cleavages and the quality of the data is
notoriously poor. We also find that measures of democracy
commonly used in panel designs leave much to be
desired.
The problems of testing these theories are not limited
to the constraints posed by the data: They are also related
to the reduced-form nature of most cross-national
panel designs. These quantitative models typically omit
the intervening causal processes and focus directly on
the relationship between some antecedent condition—
in this case, levels of inequality—and the outcome variable,
regime change in this instance. However, as the literature
on process-tracing and causal process observation
has pointed out,2
the empirical question is not only
whether antecedent conditions are linked statistically
to the outcome but whether they also do so through the
stipulated causal mechanisms. In this case, we want to
know not only whether inequality is associated with
regime change but also whether its effects operate
through the particular causal mechanisms postulated
in distributive conflict theory.
Our method of causal process observation includes
two stages: (1) within-case analysis and coding and (2)
aggregation across the population of cases (Haggard,
2 The concept of causal process observation (Collier, Brady, and
Seawright 2010) grew out of an earlier stream of methodological
work on process-tracing initiated by Alexander George (Bennett and
George 2005; George and McKeown 1985) and subsequently joined
by work on the empirical testing of formal models, including through
“analytic narratives” (Bates et. al. 1998). Although Collier, Brady,
and Seawright distinguish between causal process observation and
process-tracing, we see them as essentially the same. However, we
prefer the term “causal process observation” because it underscores
the link to the testing of a particular theory; we suggest later the
particular way in which this approach can be used to leverage causal
inference. A related strand of work is associated with the “mechanism”
approach to causation (Falletti and Lynch 2010; Gerring 2007b;
2010; Hedstrom and Ylikoski 2010).
Kaufman, and Teo 2012). Selecting on the dependent
variable is a central feature of this approach, which
is designed to test a particular theory and thus rests
on identification of the causal mechanism leading to
regime change. In contrast to the more common practice
of purposeful (Gerring 2006; 2007a; 2007b) or random
(Fearon and Laitin 2011) selection of cases for
more intensive analysis, our approach is to select all
transition and reversion cases in the relevant sample
period (1980–2000). The cases included in our dataset
come from the dichotomous coding of transitions and
reversions in Cheibub, Ghandi, and Vreeland (CGV;
2010) and from Polity IV. For the continuous Polity IV
metric, we use a cutoff of 6 to indicate a transition, a
benchmark used in the dataset itself. We have, however,
examined alternative cutoff points of 7 and 8 and find
that as the bar is raised, the percentage of distributive
conflict transitions in our sample actually declines to
47.9 and 42.1%, respectively, suggesting that the results
are in fact robust.
Within-case causal process observation involves the
reconstruction of an empirical sequence of actor decisions,
ultimately strategic in form, that are postulated
by the theory to yield the given outcome. Within-case
analysis codes whether and to what extent individual
cases conform with the stipulated causal logic. This coding
can then be aggregated in a second stage to consider
characteristics of the whole population or subsets of it.
In constructing the dataset of causal process observations
on regime change, we begin with the stipulated
causal mechanisms that run from inequality through
the following elements: the mobilization of distributive
grievances by the poor or—more commonly—by
coalitions of low- and middle-income groups; elite calculations
about the costs of repressing these challenges
or offering political concessions; the iterated strategic
response of the masses to those elite decisions; and the
ultimate outcome of regime maintenance or change
(see particularly Boix 2003, 27–36, and Acemoglu and
Robinson 2006, 181–220, for explication of the basic
models). In the first instance, we seek to establish
whether distributive conflict is present or not and, if so,
whether and how it affects the decisions that result in or
constitute regime change. For democratic transitions,
we first identify the decisions made by authoritarian
leaders to make political concessions or withdraw altogether.
For reversions, we identify actions taken by
challengers within or outside the government that result
in the overthrow of democratic rule. For each transition
and reversion, we then provide a narrative that
reconstructs the causal process and assesses whether
the key political decisions in question were a result of
distributive conflicts. We then provide a justification of
the coding and references used to make the decision.3
The selection of all cases for a given time period has
the advantage of permitting what we call “stage two”
3 We personally researched all cases cited in the dataset and consulted
closely with each other on each coding decision and consistency
across cases. Country and regional experts also reviewed coding
decisions, particularly in ambiguous cases (see Haggard, Kaufman,
and Teo 2012).
498
American Political Science Review Vol. 106, No. 3
analysis: the aggregation of the individual causal process
observations to permit analysis of the population
as a whole or relevant subsets of it. For example, we pay
particular attention to high- and low-inequality cases
because the theory has particular expectations about
how such cases should behave.
The method of causal process observation has several
advantages that can enrich the testing of formal
theories through quantitative empirical designs; we see
it as a complement to such approaches, not a substitute.
In a quantitative model, the effects of either structural
variables, such as inequality, or behavioral ones such
as protest are estimated across a heterogeneous set
of cases, some of which transition as a result of the
stipulated causal mechanism and some of which do
not. The focus on average treatment effects masks
the heterogeneity of transition paths; the variable in
question is either significant or not. By contrast, causal
process observations do not ask whether the variable in
question is significant, but whether the transition path
in the cases conforms with the causal process stipulated
in the theoretical model.
As we see later, the quantitative work on inequality
and regime change is highly inconclusive at best
and is even more limited for the third wave transitions.
Nonetheless, causal process observations can
complement quantitative analysis in two ways that can
strengthen causal inference. First, if causal process observations
showed that elite-mass conflicts did drive
transitions in a significant number of cases, it could
reopen null statistical findings. The causal process observations
would suggest, for example, the need for
better specification of the quantitative model or more
appropriate measures of inequality. However, if regime
change was not driven by such conflicts in a significant
number of cases, the finding could be considered
disconfirmatory. More importantly, the finding could
be disconfirmatory even if inequality were statistically
significant in the quantitative analysis; this would occur
if causal process observation showed that the effects
of inequality work through causal channels not posited
by the game-theoretic models.
In addition to its advantages in more closely testing
the actual mechanisms specified in causal models,
causal process observations also address a second
important problem in standard quantitative panel designs:
the mismatch between the temporal framework
of a stipulated causal process and the constraints of
country-year coding of cases. In cross-national panels,
each country-year is coded as a transition or nontransition
year; these codings constitute the dependent variable.
The causal covariates are similarly either contemporaneous
or antecedent with some lag structure. Yet
the causal sequence of actor choices associated with
transitions and reversions may be more compressed
or extended, not constant across cases, and thus not
well captured by the artifact of the country-year coding
constraint typical of the panel design. As we see later,
many cases that are coded as transitions prove to be
dubious when a more extended but variable temporal
context is taken into account, a point emphasized more
generally in the work of Pierson (2004).
In principle, multistage models can be constructed
that work from structural causes through intervening
behaviors to institutional effects (King, Keohane, and
Verba 1994, 85–87). Some critics of the mechanisms
approach have argued that mechanisms may be nothing
more than such chains of intervening variables (Beck
2006; 2010; Gerring 2007b; 2010; Hafner-Burton and
Ron 2009). Although possible in principle, the continued
reliance on reduced-form specification suggests
that this problem is in fact not addressed, in part because
of the labor intensity of recoding existing datasets
to conform more precisely with the theory being tested.
In each of the remaining sections on democratic
transitions and reversions, we begin with a review of
the quantitative findings on the relationship between
inequality and regime change and then present both
aggregate and select case study findings from the causal
process observations in our dataset. We show that the
support for the distributive conflict model of regime
change is weak, even under highly generous coding
rules. When these rules are tightened, the evidence is
weaker still.
TRANSITIONS TO DEMOCRATIC RULE
Acemoglu and Robinson (2006) do not present systematic
empirical evidence in support of their claims.4
Much of their book is taken up with a discussion of
the underlying intuition of the theory (1–47, 80–87)
and the presentation of a family of formal models of
democratic and nondemocratic regimes (89–172) and
of regime change (173–320). Acemoglu and Robinson
do present scatterplots showing a positive relationship
between equality and the level of democracy across a
global sample of countries (58–61) and provide short
case studies of Great Britain, Argentina, South Africa,
and Singapore (1–14). Yet these correlations and cases
are illustrative at most.
In his analysis of democratic transitions over the very
long run (1850–1980), Boix (2003) finds that the distribution
of land, proxied by the share of family farms, has
an effect on the transition to democratic rule. More unequal
societies are both less likely to make a transition
to democracy and less stable when they do (90–97).
Boix also explores a highly uneven panel of countries
for the 1950–90 period (only 587 observations), including
developed ones (71–88). Using a Gini index as his
measure of inequality, Boix finds some evidence that
increases in the level of inequality reduce the likelihood
of a democratic transition, but the findings are
not altogether robust (see for example, 79: Model 2A).
More recently, other quantitative studies have taken
up the challenge raised by Boix (2003) and Acemoglu
and Robinson (2006), but with mixed results.
4 Earlier work (Acemoglu and Robinson 2000; 2001) was motivated
by experiences in nineteenth-century Europe and early twentiethcentury
Latin America. However, in those articles, as well as in the
later book, the formal theory is cast in general terms, without specifying
scope conditions that might apply to third wave transitions. In the
book, moreover, the illustrations from South Africa and Singapore
rely on much more recent developments.
499
Inequality and Regime Change August 2012
Like Boix, Ansell, and Samuels (2010) consider both
long-historical and postwar samples (1850–1993, 1955–
2004). They find that land concentration makes democratization
less likely, but that increases in income
inequality make it more likely. They argue that increasing
income inequality reflects the emergence of a new
capitalist class that challenges landed elites, a dynamic
consistent with Boix’s (2003, 47–59) and Acemoglu and
Robinson’s (2006, 266–86) extension of their models
into three-class variants.
The limited number of other tests in the literature
generally fail to find a relationship between inequality
and democratic transitions. A cross-sectional design by
Dutt and Mitra (2008) finds a relationship between
inequality measured by the Gini coefficient and “political
instability,” but fails to find a relationship between
inequality and transitions to democratic rule. Christian
Houle (2009) creates a dataset using an alternative
measure of inequality: capital’s share of income
in the manufacturing sector. Using the dichotomous
coding scheme developed by Przeworski et al. (2000)
and Cheibub and Ghandi (2004), Houle shows that inequality
bears no systematic relationship to democratic
transitions over the 1960–2000 period, but is a significant
predictor of reversions to authoritarian rule. In a
wide-ranging study of the determinants of democratization,
Teorell (2010, 60) also fails to find a relationship
between a Gini coefficient and democratic transitions.
Distributive Conflict and
Nondistributive Conflict Transitions
In sum, the quantitative work on inequality and regime
change is highly inconclusive at best, and even more
limited for the third wave transitions. Many of these
tests do not empirically model the underlying causal
processes stipulated in the most significant formal
models. Therefore, to undertake causal process observations,
we need to interpret the underlying causal
mechanisms at work in the theory. Two mechanisms
appear central. First, elites must confront politicalcum-distributive
pressure from below, or a “clear and
present danger” of it. In the absence of such pressures,
it is not clear why elites would be motivated to cede
power at all, as Przeworski’s trenchant question suggests.
Second, there must be some evidence—minimally
in the temporal sequence of events—that the repression
of these challenges appears too costly and that elites
make institutional compromises as a result.
We therefore code “distributive conflict” transitions
as ones in which both of the following occurred:
• The mobilization of redistributive grievances on
the part of economically disadvantaged groups or
representatives of such groups (parties, unions,
NGOs) posed a threat to the incumbency of ruling
elites.
• And the rising costs of repressing these demands
appear to have motivated elites to make political
compromises or exit in favor of democratic
challengers, typically indicated by a clear temporal
sequence (mass mobilization followed by authoritarian
withdrawal).
In coding the cases, we were deliberately permissive,
writing coding rules that gave the benefit of the doubt
to the theory (Haggard, Kaufman, and Teo 2012). Our
coding allowed us to consider a variety of distributive
conflicts that may not be captured by any single inequality
measure, from urban class conflicts to ethnic
and regional ones. Yet such conflicts must be fought
around distinctive and identifiable inequalities. The
economically disadvantaged or the organizations representing
them need not be the only ones mobilized in
opposition to the existing regime. Although mass mobilization
must partly reflect demands for redistribution,
it can be motivated by other grievances as well.
An important coding issue is the question of “potential”
threats in the absence of actual mobilization. As
Acemoglu and Robinson (2006) note in distinguishing
between de jure and de facto power, the poor can be
considered a potential threat in virtually every case.
However, the strategic basis of reforms aimed at preempting
potential long-term threats rests on probability
estimates and time horizons on the part of elites that
differ quite substantially from those that drive elite responses
to more immediate challenges. Moreover, we
are also wary of the coding challenge: Virtually any case
could be coded as one in which there was a “potential”
challenge from below, with a corresponding decline
in analytic leverage. However, we do take potential
threats into account where there has been a recent
history of mass mobilization demanding democratic
reforms.
We coded all cases in which such threats from below
did not occur at all or appeared to play only a marginal
causal role as “nondistributive transitions.” Why, in the
absence of significant pressure from below, would elites
withdraw or make institutional compromises that risk
the redistribution of assets and income not only in the
present but also into the indefinite future? As others
have argued (Huntington 1991; Linz and Stepan 1996;
Collier 1999; O’Donnell, Schmitter, and Whitehead
1986), there are a variety of routes from closed political
systems to democracy. We identify three: those driven
by international pressures, those involving intra-elite
conflicts and defections, and those in which incumbent
authoritarian elites withdraw in the belief that they can
control the post-transition democratic order in ways
that limit democracy’s redistributive impact.
International factors played a decisive role in a
number of third wave transitions (Boix 2011; Whitehead
1996). In a handful of cases—including Grenada
(1984), Panama (1989), and Haiti (1994)—outside intervention
took a military form. Yet particularly in the
wake of the end of the Cold War, aid donors—both
multilateral and bilateral—became less tolerant of undemocratic
regimes that appeared guilty of economic
mismanagement and outright corruption. Threats or
withdrawal of aid played an important role in transitions
in a group of low-income African countries in
particular.
500
American Political Science Review Vol. 106, No. 3
Even if we set aside the role of international pressures,
threats from below are by no means the only
domestic pressures that can cause elites to acquiesce to
democratizing institutional changes. A common cause
of transition in the nondistributive conflict cases is
intra-elite rivalries. These rivalries may stem from competition
among the political, military, and economic
elites that constitute the authoritarian coalition—for
example, when factions within the regime seek to displace
incumbents—or from elite challenges from outside
the regime altogether (Slater and Smith 2012). In
a number of cases, we found that concessions to elites
rather than mass challenges appear as important as
distributive conflicts pitting rich against poor.
Even when elites remain relatively unified, they may
still acquiesce to—or even lead—democratic reform if
they believe they can retain leverage over the political
process while reducing the costs of repression. Incumbent
elites can do this in several ways, including through
the design of political institutions that give them effective
vetoes or through the organization of political parties
that exploit other cleavages to dampen distributive
conflicts. Dominant parties provide incumbent political
elites particular organizational advantages that can be
redeployed in a more competitive context.
Note that each of the alternative domestic causal
mechanisms we have sketched—intra-elite conflict or
defection and authoritarian elites ceding office because
of confidence in their post-transition chances—may in
fact be related precisely to the weakness of immediate
threats from below. Where such threats are limited,
elites are more likely to control the transition. Societies
in which the poor are not mobilized through programmatic
parties, unions, or other organizations may be
especially prone to vote buying, patronage, and other
forms of clientelistic control that would guarantee elite
control of politics, even in nominally democratic settings
(Kitschelt and Wilkinson 2006).
Inequality and the Incidence of
Distributive Conflict Transitions
Table 1 shows the distributive and nondistributive transitions,
using the definition of transitions in the CGV
dataset; in the text, we also report the distribution of
these types of cases based on Polity transition coding.
The cases are arrayed according to three measures
of inequality: Christian Houle’s (2009) measure
of capital’s share of income in the manufacturing sector
(capshare), a Gini coefficient from the University
of Texas Inequality Project’s Estimated Household Income
Inequality (EHII) dataset (2008), and the Vanhanen
(2003) measure of land inequality. We divide
the sample of all developing countries into terciles of
high-, medium-, and low-inequality cases and identify
the transitions that fall into each tercile.
The table shows that transitions occurred at all levels
of inequality, regardless of which measure is used.
Twenty-nine percent of transitions occurred in the upper
third of countries ranked by capshare and Gini
inequality, and about 34% occurred in the top tercile
of countries ranked in terms of land distribution. More
problematic, and against theoretical expectations in
both Boix (2003) and Acemoglu and Robinson (2006),
there was a substantial incidence of distributive conflict
transitions among the high-inequality cases. When inequality
is measured using the Gini, about 75% of highinequality
transitions were distributive conflict transitions;
the incidence of such transitions is 60% using the
land inequality measure, and 57% using capital’s share
of income.
Table 2 offers additional insight into the causal role
of distributive conflict. Columns 2 and 3 divide the
CGV transitions into distributive and nondistributive
types; columns 4 and 5 replicate the exercise for Polity
transitions. We also identify the non-overlapping cases.
The last column shows the average Polity score from
the time of the transition through either the end of the
sample period or until an outright reversion to authoritarian
rule.
The information contained in Table 2 raises serious
questions about the validity of the coding of democratic
transitions in these two major datasets and, as a result,
casts doubt on the inferences that have been drawn in
the quantitative work that employs them. Only 55.4%
of the CGV transitions are also Polity cases, and 21 of
the 65 CGV transitions had Polity scores of less than 6.
Even where the two datasets are in agreement, moreover,
our examination of the cases raises questions
about the validity of the coding process. Insiders and
elites repressed opposition and/or exercised disproportionate
control over them in the nominallydemocratic
cases of Croatia, Niger, and Thailand. Transitions in
Guatemala (1986) and Honduras (1982) empowered
nominally democratic governments that actually intensified
repression of social movements that had redistributive
objectives. Death squads continued to terrorize
the opposition in El Salvador after the transition
in 1984. In at least four cases—Ghana under Rawlings;
Kenya under Moi; Malawi, where an “insider” won
the transitional election; and Romania—the military or
incumbent elites continued to exercise disproportionate
influence over the allocation of resources after the
transition. In all of these cases, the transitions appear to
conform more closely to what Levitsky and Way (2010)
call “competitive authoritarianism” than to democracy.
Because we seek to engage the quantitative analysis
that deploys such data, however, we do not discard
or reclassify cases identified as transitions in the two
datasets.
What about the theoretical expectations of the role
of distributive conflict in democratic transitions? We
found that distributive conflict played some causal role
in propelling transitions in about 55% of CGV and
58% of Polity transition cases. These are substantial,
but by no means overwhelming shares of the cases. In
combination with the findings in Table 1, the large percentage
of nondistributive transitions suggests strongly
that the link between inequality and distributive conflict
transitions is conditional at best.
Yet even these findings need to be tempered by the
generosity of our coding rules. Although pressure from
below did play an unambiguously significant role in a
501
InequalityandRegimeChangeAugust2012
TABLE 1. Distributive and Nondistributive Transitions by Level of Inequality, 1980–2000
Inequality Measures
Capital Share of Income in Manufacturing Sector Gini coefficient (Texas Inequality dataset) Share of Family Farms (Vanhanen)
Level of Inequality Distributive Nondistributive Distributive Nondistributive Distributive Nondistributive
High Bolivia (1982) Chile (1990) Armenia (1991) Ghana (1993) Albania (1992) Belarus (1991)
Brazil (1985) Ghana (1993) Benin (1991) Paraguay (1989) Bolivia (1982) Chile (1990)
Burundi (1993) Mexico (2000) Bolivia (1982) Sierra Leone (1996) Brazil (1985) Czechoslovakia (1989)
Indonesia (1999) Nicaragua (1984) Brazil (1985) Sierra Leone (1998) Bulgaria (1990) Honduras (1982)
Nigeria (1999) Sierra Leone (1996) Burundi (1993) Estonia (1991) Hungary (1990)
Peru (1980) Sierra Leone (1998) Congo (1992) Guatemala (1986) Nicaragua (1984)
Sri Lanka (1989) Guatemala (1986) Latvia (1991) Panama (1989)
Thailand (1992) Kenya (1991) Lithuania (1991) Paraguay (1989)
Malawi (1994) Mongolia (1990)
Mongolia (1990) Peru (1980)
Nepal (1990) Romania (1990)
Romania (1990) Ukraine (1991)
Percentage of
distributive and
nondistributive
conflict cases
57.1 42.9 75.0 25.0 60.0 40.0%
Medium Albania (1991) Bangladesh (1986) Argentina (1983) Central African Republic Armenia (1991) Central African Republic
Argentina (1983) Croatia (1991) El Salvador (1984) (1993) Argentina (1983) (1993)
Benin (1991) Hungary (1990) Fiji (1992) Chile (1990) Benin (1991) Comoros (1990)
Bulgaria (1990) Pakistan (1988) Indonesia (1999) Honduras (1982) Congo (1992) Mexico (2000)
El Salvador (1984) Panama (1989) Peru (1980) Pakistan (1988) El Salvador (1984) Pakistan (1988)
Guatemala (1986) Paraguay (1989) The Philippines (1986) Panama (1989) Fiji (1992) Senegal (2000)
Kenya (1998) Senegal (2000) Sri Lanka (1989) Senegal (2000) Kenya (1998)
Latvia (1991) Turkey (1983) Suriname (1988) Suriname (1991) Malawi (1994)
Madagascar (1993) Thailand (1992) Turkey (1983) The Philippines (1986)
Malawi (1994) Uruguay (1985) Uganda (1980) Sudan (1986)
Nepal (1990) Uruguay (1985)
The Philippines (1986)
Poland (1989)
South Korea (1988)
Sudan (1986)
Uruguay (1985)
502
AmericanPoliticalScienceReviewVol.106,No.3
TABLE 1. Continued.
Inequality Measures
Capital Share of Income in Manufacturing Sector Gini coefficient (Texas Inequality dataset) Share of Family Farms (Vanhanen)
Level of Inequality Distributive Nondistributive Distributive Nondistributive Distributive Nondistributive
Percentage of
distributive and
nondistributive
conflict cases
66.7 33.3 52.6 47.4 68.8 31.2%
Low Fiji (1992) Central African Republic Albania (1992) Bangladesh (1986) Burundi (1993) Bangladesh (1986)
Niger (1993) (1993) Bulgaria (1990) Cape Verde (1990) Indonesia (1999) Croatia (1991)
Niger (2000) Cyprus (1983) Latvia (1991) Croatia (1991) Madagascar (1993) Ghana (1993)
Romania (1990) Honduras (1982) Lithuania (1991) Cyprus (1989) Mali (1992) Guinea-Bissau (2000)
Macedonia (1991) Madagascar (1993) Czech Republic (1989) Nepal (1990) Macedonia (1991)
Uganda (1980) Nigeria (1999) Hungary (1990) Niger (1993) Serbia (2000)
Poland (1989) Macedonia (1991) Niger (2000) Sierra Leone (1996)
South Korea (1988) Mexico (2000) Nigeria (1999) Sierra Leone (1998)
Ukraine (1991) Nicaragua (1984) Poland (1989) Taiwan (1996)
Serbia (2000) South Korea (1988) Turkey (1983)
Taiwan (1996) Sri Lanka (1989) Uganda (1980)
Thailand (1992)
Percentage of
distributive and
nondistributive
conflict cases
44.4% 55.6% 45.0% 55.0% 52.2% 47.8%
Missing Data Armenia (1991) Belarus (1991) Estonia (1991) Belarus (1991) Suriname (1988) Cape Verde (1991)
Congo (1992) Cape Verde (1990) Mali (1992) Comoros (1990) Cyprus (1983)
Estonia (1991) Comoros (1990) Niger (1993) Grenada (1984) Grenada (1984)
Lithuania (1991) Czechoslovakia (1989) Niger (2000) Guinea-Bissau (2000) Sao Tome and Principle (1991)
Mali (1992) Grenada (1984) Sudan (1986) Sao Tome and Principe
(1991)
Suriname (1991)
Mongolia (1990) Guinea-Bissau (2000)
Suriname (1988) Sao Tome and Principe
Ukraine (1991) (1991)
Serbia (2000)
Suriname (1991)
Taiwan (1996)
Sources: Transitions: Cheibub, Ghandi, and Vreeland (2010); transition types: Haggard, Kaufman, and Teo (2012); capital share: Houle (2009); Gini: University of Texas Inequality Project
(2008); share of family farms: Vanhanen (2003).
503
Inequality and Regime Change August 2012
TABLE 2. Distributive and Nondistributive Transitions, 1980–2000
CGV Transitions Polity Transitions
Country/Year Distributive Nondistributive Distributive Nondistributive Polity score
Albania 1991 X Not a Polity transition 4.1
Argentina 1983 X X 7.4
Armenia 1991 X X 7.0
Bangladesh 1986 X Not a Polity transition 2.3
Bangladesh 1991 Not a CGV transition X 6.0
Belarus 1991 X X 7.0
Benin 1991 X X 6.0
Bolivia 1982 X X 8.8
Brazil 1985 X X 7.8
Bulgaria 1990 X X 8.0
Burundi 1993 X Not a Polity transition −1.6
Cape Verde 1990
(CGV), 1991 (Polity)
X X 7.1
Central African
Republic 1993
X Not a Polity transition 5.0
Chile 1990 (CGV),
1989 (Polity)
X X 8.1
Comoros 1990 X Not a Polity transition 2.6
Congo 1992 X Not a Polity transition 5.0
Croatia 1991 X Not a Polity transition −2.0
Croatia 2000 Not a CGV transition X 8.0
Cyprus 1983 X Not a Polity transition 10.0
Czechoslovakia 1989
(CGV), 1990 (Polity)
X X 8.2
Dominican Republic
1996
Not a CGV transition X 8.0
El Salvador 1984 X X 6.6
Estonia 1991 X X 6.0
Fiji 1992 X Not a Polity transition 5.1
Fiji 1999 Not a CGV transition X 5.5
Ghana 1993 X Not a Polity transition 1.0
Grenada 1984 X Not a Polity transition N.A.
Guatemala 1986 X Not a Polity transition 4.7
Guatemala 1996 Not a CGV transition X 8.0
Guinea-Bissau 2000 X Not a Polity transition 5.0
Guyana 1992 Not a CGV transition X 6.0
Haiti 1990 Not a CGV transition X 7.0
Haiti 1994 Not a CGV transition X 7.0
Honduras 1982 X X 6.0
Honduras 1989 Not a CGV transition X 6.2
Hungary 1990 X X 10.0
Indonesia 1999 X X 6.0
Kenya 1998 X Not a Polity transition −2.0
Latvia 1991 X X 8.0
Lesotho 1993 Not a CGV transition X 8.0
Lithuania 1991 X X 10.0
Macedonia 1991 X X 6.0
Madagascar 1992 X X 8.2
Malawi 1994 X X 6.0
Mali 1992 X X 6.5
Mexico 1997 Not a CGV transition X 6.5
Mexico 2000 X Not a Polity transition 8.0
Moldova 1993 Not a CGV transition X 7.0
Mongolia 1990(CGV),
1992 (Polity)
X X 8.2
Nepal 1990 X Not a Polity transition 5.2
Nepal 1999 Not a CGV transition X 6.0
Nicaragua 1984 X Not a Polity transition 4.2
Nicaragua 1990 Not a CGV transition X 7.1
Niger 1993 (CGV),
1992 (Polity)
X X 8.0
504
American Political Science Review Vol. 106, No. 3
TABLE 2. Continued.
CGV Transitions Polity Transitions
Country/Year Distributive Nondistributive Distributive Nondistributive Polity score
Niger 2000 X Not a Polity transition 5.0
Nigeria 1999 X Not a Polity transition 4.0
Pakistan 1988 X X 7.8
Panama 1989 X X 8.6
Paraguay 1989 X Not a Polity transition 5.7
Paraguay 1992 Not a CGV transition X 6.9
Peru 1980 X X 7.2
The Philippines 1986
(CGV), 1987 (Polity)
X X 7.5
Poland 1989 (CGV),
1991(Polity)
X X 8.0
Romania 1990 X Not a Polity transition 6.4
Romania 1996 Not a CGV transition X 8.0
Russia 2000 Not a CGV transition X 6.0
Sao Tome and
Principe 1991
X Not a Polity transition N.A.
Senegal 2000 X X 8.0
Serbia 2000 X X 7.0
Sierra Leone 1996 X Not a Polity transition 4.0
Sierra Leone 1998 X Not a Polity transition 0.0
South Africa 1992 Not a CGV transition X 8.6
South Korea 1988 X X 6.5
Sri Lanka 1989 X Not a Polity transition 5.0
Sudan 1986 X X 7.0
Suriname 1988 X Not a Polity transition N.A.
Suriname 1991 X Not a Polity transition N.A.
Taiwan 1992 Not a CGV transition X 8.0
Taiwan 1996 X Not a Polity transition 8.8
Thailand 1992 X X 9.0
Turkey 1983 X X 7.7
Uganda 1980 X Not a Polity transition 2.5
Ukraine 1991 X X 6.0
Ukraine 1994 Not a CGV transition X 6.9
Uruguay 1985 X X 9.8
Zambia 1991 Not a CGV transition X 6.0
N/% 36/55.4% 29/44.6% 33/57.9% 24/42.1% 6.3
Note: In the dataset, we treat any transitions that are coded within a two-year window as the same case (for example, the CGV coding
of the Philippines transition occurring in 1986, the Polity coding as 1987). Outside of this two-year window (for example, Paraguay) or
where there is an intervening reversion (Sierra Leone), we treat them as separate cases. There are no Polity scores for Grenada, Sao
Tome, and Suriname.
Sources: CGV transitions from Cheibub, Ghandi, and Vreeland (2010); Polity transitions from Marshall, Gurr, and Jaggers (2010);
transition types from Haggard, Kaufman, and Teo (2012).
number of middle-income countries such as Argentina,
South Korea, and South Africa, our expansive coding
rules also necessitated the classification of cases as distributive
conflict where there was considerable ambiguity
about its causal weight. The ambiguity in specific
cases stemmed from one or more of three factors.
1. First, some distributive conflict transitions occurred
in small open economies that were highly vulnerable
to pressure from donors or other international
actors, and this pressure may have been decisive.
2. The class basis of protest constituted a second
source of ambiguity; in many cases, protest was
dominated by middle- or even upper-middle-class
groups, calling into question the class dynamics of
the model even if we allow for cross-class coalitions
including the poor.
3. A third source of ambiguity involved judgments
about the role played by redistributive grievances
in opposition demands; in many instances, it was
difficult to separate redistributive demands from
grievances that focused on a defense of privileged
positions, generalized dissatisfaction with authoritarian
incumbents, or nationalist claims.
Table 3 lists the cases in the dataset where international
pressures, the class composition of the protestors,
or the nature of their redistributive grievances made
505
Inequality and Regime Change August 2012
TABLE 3. Ambiguous Cases of Distributive Conflict Transitions
CGV Dataset Polity Dataset
Country Source of Ambiguity Country Source of Ambiguity
Armenia Grievance Armenia Grievance
Benin Class Benin Class
Bulgaria Grievance Bulgaria Grievance
Congo Class
El Salvador International El Salvador International
Estonia Class/Grievance Estonia Class/Grievance
Fiji International
Kenya International
Lesotho Class/International
Latvia Class/Grievance Latvia Class/Grievance
Lithuania Class/Grievance Lithuania Class/Grievance
Malawi Class/International Malawi Class/International
Mali Class Mali Class
Mongolia Class/Grievance Mongolia Class/Grievance
Niger Class/Grievance/International Niger Class/Grievance/International
Sri Lanka Grievance
Suriname International
Ukraine Class/Grievance Ukraine Class/Grievance
Total 17 13
Percent of
total
transitions
26.2 22.8
the coding of the case ambiguous. Of particular significance
is the coding of several African “distributive conflict”
transitions, in which incumbent regimes—in the
midst of severe economic recessions—were vulnerable
both to intense donor pressure and the protest of relatively
well-off public employees and student groups.
Niger provides an example. The pivotal decision in
this case was an agreement by the military strongman,
General Ali Saibou, to convene a National Conference,
which then assumed the role of a transitional
government and organized competitive elections. Distributive
protests played a role in Saibou’s decision
to yield authority. Yet the opposition came primarily
from the Nigerien Workers Union, which represented
Niger’s 39,000 civil servants, and the Union of Nigerian
Scholars, which represented about 6% of the country’s
school-aged population (Gervais 1997, 93). Both
groups bitterly opposed tough adjustment programs
demanded by the International Monetary Fund, but
the conflicts did not appear to engage the poor. As
Gervais (1997, 105) writes, “the political stakes raised
by . . . adjustment policies tended to compromise the
benefits of the organized groups of the modern sector
as much as the privileges of the traditional political
class.” Notwithstanding our generous coding decision,
it is ambiguous at best to claim that the transition
process mapped directly to the underlying MeltzerRichard
model in which the interests of the poor or
even middle classes are pitted against the rich.
Similar questions can be raised about the class composition
of protest in other African cases, including
Benin, Congo, Lesotho, Malawi, and Mali. Even
though all of these cases meet our coding rules because
of the presence of mobilization “from below”
that affected the transition, protest was primarily limited
to civil servants, students, and other sectors of the
urban middle class. Moreover, several African cases
(Lesotho, Kenya, Malawi, and Niger) were also ambiguous
with respect to the role of international pres-
sures.
The nature of the grievances associated with the secession
from Yugoslavia and the former Soviet Union
also warrants special mention. In three such cases—
Croatia, Macedonia, and Belarus—the coding was unambiguously
nondistributive because mass mobilization
on distributive lines was altogether absent or there
is strong evidence that the political process of independence
occurred as a result of intra-elite processes. Yet
in the Baltic cases, as well as in Ukraine, Mongolia,
and Armenia, there is ambiguity as to the nature of
the claims made by groups engaged in mass mobilization.
Several regional specialists whom we consulted in
constructing our coding objected that these cases did
not fall easily into the distributive conflict category and
should be seen as the outcome of cross-class secessionist
or nationalist movements and the resulting collapse
of multinational empires. In these cases, we believed
that the evidence of conflicts within the polity between
indigenous populations and the Russians warranted a
“distributive conflict” coding, but it is important to acknowledge
the pivotal importance of strong nationalist
aspirations that cut across class lines.5
If we were to
5 These cases also posed a second coding problem: whether they
should be treated as democratic transitions at all given that they
506
American Political Science Review Vol. 106, No. 3
shift all of the ambiguous cases in Table 3 from the distributive
to nondistributive categories, the incidence of
distributive conflict transitions would fall to only 29.2%
of the CGV transitions and to about 31% of the Polity
transitions.
Even with the expansive coding of distributive conflict
transitions, we found a large share of cases—44.6%
using the CGV measure and 42.1% of Polity cases—in
which distributive conflict played only a marginal role
in the transition process. These cases followed the alternative
causal pathways we identified earlier: transitions
driven by international pressures or by intra-elite
conflicts, and elite-led transitions in which incumbents
believed they could control the democratic process to
limit its redistributive impact.
We have already noted that, in several of the “ambiguous
cases” discussed earlier, popular protest unfolded
in the context of severe international pressure.
However, in other cases, protest was weak or entirely
absent, and outside intervention was unambiguously
decisive. Transitions in Grenada (1984) and Panama
(1989) hinged almost entirely on U.S. military operations.
In Haiti (1994), the military ruler negotiated his
exit as an international force of 21,000 troops prepared
to land on the island. External political and economic
pressures from donors or great power patrons were also
decisive in Comoros (1990), Cape Verde (1990), the
Central African Republic (1993), and Cyprus (1983).
Intra-elite conflicts appear significant in a number
of nondistributive conflict cases. The 1989 transition in
Paraguay provides an illustration. The key decision was
a palace coup that ousted the aging dictator Alfredo
Stroessner and initiated a process of constitutional reform
and competitive presidential elections. The coup
was led by General Andres Rodriguez, Stroessner’s
second in command, and by a faction of the ruling Colorado
party that hoped to extend one-party rule by engineering
a “nonpersonalist” transition. Mass protest
did not pose a serious threat to the regime (Lambert
2000).
In Mexico the ruling PRI was challenged primarily
by business elites and an opposition party (PAN) that
was outside the regime and its ruling coalition and
wanted less rather than more redistribution. Popular
protest over alleged fraud in local elections strengthened
the bargaining leverage of the PAN in its negotiations
with the ruling party, but the political left played
only a marginal role in pushing the regime out of power.
Among other cases in which elite concessions to other
elites appeared significant are the military’s acquiescence
in the assumption of power by parliamentary
politicians in Pakistan, the Thai military’s accommodation
of emerging political-economic elites from the
Northern part of the country, and the Kuomintang’s
accommodation of native Taiwanese elites.
Finally, in a number of cases incumbent authoritarian
elites opened politics under the assumption—justified
or mistaken—that they could effectively control the
are entirely new countries. We chose to include them in the dataset
because they cross standard thresholds (Polity) or appear as new
democracies (CGV); see Haggard, Kaufman, and Teo (2012).
political system to limit its redistributive impact. The
Turkish military transferred power to a new civilian
government in 1983, but only after crushing violent
left and right factions that had been a feature of Turkish
politics in the late 1970s. As it gradually reopened
the political space in early 1983, the military vetoed
most of the new parties that had formed around established
politicians and designed institutions that gave it
veto power over crucial areas of policy. Although the
military elite was surprised by the victory of the one
opposition party it had allowed to function, there is no
indication that threats of mass mobilization influenced
the decision to allow the elections or to permit the
results to stand.
We see similar processes of reform in Chile, where
outgoing governments built in quite specific mechanisms
through which the military would continue
to exercise oversight and supporters of the outgoing
government would be overrepresented (Haggard and
Kaufman 1995). These mechanisms included the establishment
of national security councils with a veto
role for the military establishment, constitutional and
judicial guarantees limiting the authority of incoming
governments, and the allocation of Senate seats to be
filled by the head of the outgoing regime. In Kenya,
Mexico, and Taiwan, incumbents ceded power gradually
while competing aggressively and successfully in
the newly liberalized environment. Several communist
transitions, including Hungary and Mongolia, also fit
this pattern.
Two conclusions emerge from our discussion of
democratic transitions. First, although certainly some
democratic transitions are driven by distributive conflict
in ways that conform with the theory, these cases do
not appear to be related in any systematic way with the
level of inequality, as the lack of quantitative findings
already suggests. Second, the assumption that elites
do not yield power in the absence of mass pressure
from below is called into question by the high incidence
of alternative transition paths. Taken together,
these conclusions indicate that the theory is, at best,
underspecified and needs to delineate more explicitly
the conditions in which redistributive conflicts emerge.
We return to these issues in the conclusion.
The Collapse of Democratic Rule:
Causal Process Observations
Although Acemoglu and Robinson (2006) and Boix
(2003) offer diverging predictions about transitions
to democracy, they agree that when democracies do
emerge at high levels of inequality, they are more
likely to revert to authoritarian rule. As Acemoglu and
Robinson put it succinctly, “in democracy, the elites are
unhappy because of the high degree of redistribution
and, in consequence, may undertake coups against the
democratic regime” (222). This view comports with an
earlier generation of theory on “bureaucratic authoritarian”
installations in the Southern Cone (O’Donnell
1973; for critiques: see Collier 1979; Linz and Stepan
1978; Valenzuela 1978): Brazil (1964), Argentina (1966,
507
Inequality and Regime Change August 2012
and 1976), and Chile and Uruguay (both 1973), with
extensions to other regions as well (for example, Im
1987 on Korea).
Unlike the quantitative evidence on transitions,
there is somewhat stronger cross-national evidence
that inequality is incompatible with democratic stability
(e.g., Dutt and Mitra 2008; Reenock, Bernhard,
and Sobek 2007). Houle (2009) deploys his innovative
measure of capital share to test the relationship
and finds that high return to capital relative to labor
significantly undermined democratic stability between
1960 and 2000. We replicated his model using the Gini
and Vanhanen index of land inequality. Although land
shows no effects, the Gini was a significant determinant
of democratic breakdowns, both for the entire 1960–
2000 period and for the third wave between 1980 and
2000.6
Do these findings hold up when subjected to closer
qualitative scrutiny? To what extent do the causal process
observations comport with the expectations of
distributive conflict theories? As with the transition
cases, we considered whether political pressures for
redistribution drove regime change, in this case the
breakdown of democratic rule. We identified a category
called “elite-reaction” reversions that conform with
the distributive conflict model. In these cases, elites
undermine democracy either by (a) seeking to oust
incumbent governments that rely on the political support
of lower class or excluded groups and are actively
committed to the redistribution of assets and income
or by (b) imposing restraints on political competition in
order to prevent coalitions with explicitly redistributive
aims from taking office. In these cases, distributive conflicts
are in evidence and elites are acting against governments,
parties, and organized social forces that are
actively committed to greater redistribution through
the democratic process.
We also identified a second type of distributive conflict
reversion in which the incumbent democratic government
is overthrown by authoritarian populist leaders.
These types of reversion do not comport with our
expectation that reversions are driven by the right, but
they clearly involve redistributive conflict and are given
some attention in Boix (2006, 18, 214–19). Whereas
in elite-reaction reversions, challengers to democratic
rule appeal to elite interests and target the masses for
repression, in “populist reversions,” authoritarian challengers
appeal to the masses and target the elite.
Finally, “nondistributive” reversions are unambiguous
instances of the null hypothesis, but we distinguished
two alternative subtypes. In some cases,
support for a reversion cuts across class lines: Authoritarian
challengers exploit wide disaffection with
the performance of democratic incumbents and invoke
broad valence issues, such as economic performance
and corruption, that cut across distributive cleavages.
In other cases, purely intra-elite conflicts cause reversions
. The military—or factions within it—might stage
a coup against incumbent office holders, or competing
economic elites might mobilize military, militia, or
6 Results available on request from the authors.
other armed forces against democratic rule. We called
these nondistributive conflict cases “cross-class” and
“intra-elite” reversions, respectively.7
Table 4 reports the incidence of distributive conflict
and nondistributive conflict reversions by level of inequality.
The distributive conflict column aggregates
both elite-reaction and populist reversions; the nondistributive
conflict column aggregates both cross-class
and intra-elite reversions. Table 5 shows the types of
reversion, Polity scores of the deposed regimes, and
economic circumstances surrounding the change. Although
there is surprisingly little overlap between the
ranking of cases on the three measures of inequality,
reversions do cluster at the middle and high levels of
inequality; relatively few took place at the lowest levels.
However, we find only a minority of cases that conform
with the distributive conflict model. In the CGV
dataset, four cases (Bolivia 1980, Burundi 1996, Fiji
2000, and Turkey 1980) or 21% of the cases are elitereaction
reversions. Three cases (16% of the sample)—
Ecuador (2000), Ghana (1981), and Suriname (1980)—
are populist reversions. A substantial majority (63%)
of the reversions are nondistributive.
In the 20 reversions identified in the Polity measure
(not shown here), 8 of the cases—40%—are classified
as elite-reaction reversions,8
and there are 2 populist
reversions.9
Half the cases, however, are classified as
nondistributive. Missing data play more of a constraint
in allocating the Polity cases across levels of inequality,
but they are somewhat less concentrated at higher levels
of inequality, and there is no evidence that higher
inequality cases are more likely to be distributive. Fourteen
Polity reversions fall into the high-inequality tercile
on one or more of the three measures of inequality;
if each case is counted only once, only five are distributive
conflict reversions.10
To elaborate the implications of these findings, we
focus on the high-inequality cases using the capital
share measure; as the distribution of cases across different
inequality terciles suggests, very similar results
would be obtained by using different income inequality
measures.11
According to distributive conflict models,
these cases are most likely to revert as a result of elite
reactions to distributive demands from below. Given
the low correlation between measures of inequality,
alternative measures would show a different set of
high-inequality cases. However, as can be seen from
Table 4 no measure of inequality generates a distribution
of reversions that conforms with theoretical
expectations for a clustering of distributive conflict
reversions among high-inequality cases; selection of
7 Precise coding rules are available in Haggard, Kaufman, and Teo
(2012).
8 Armenia (1995), Dominican Republic (1994), Fiji (1987 and 2000),
Haiti (1991), Turkey (1991), Ukraine (1993), and Zambia (1996).
9 Ghana (1981) and Haiti (1999).
10 Ghana (1981) was a populist reversion; Armenia (1995), the Dominican
Republic (1994), the Ukraine (1993) and Zambia (1991)
were elite-reaction reversions.
11 One case, Sierra Leone, is identified as high inequality on the
capshare measure, but was not included by Houle (2009) in the
regressions because of a subsequent change in coding of the case.
508
AmericanPoliticalScienceReviewVol.106,No.3
TABLE 4. Distributive and Nondistributive Reversions by Level of Inequality, 1980–2000
Inequality Measures
Capital Share of Income in
Manufacturing Sector (capshare)
Gini coefficient
(Texas Inequality dataset) Share of Family Farms (Vanhanen)
Level of
Inequality
Distributive
(Elite/Populist) Nondistributive
Distributive
(Elite/Populist) Nondistributive
Distributive
(Elite/Populist) Non-distributive
High Bolivia (1980) E Nigeria (1983) Bolivia (1980) E Congo (1997) Bolivia (1980) E Guatemala
(1982)
Burundi (1996) E Peru (1990) Burundi (1996) E Guatemala (1982) Peru (1990)
Ghana (1981) P Sierra Leone
(1997)
Ghana (1981) P Sierra Leone
(1997)
Thailand (1991)
Medium Ecuador (2000) P Guatemala (1982) Ecuador (2000) P Pakistan (1999) Ecuador (2000) P Comoros (1995)
Turkey (1980) E Pakistan (1999) Fiji (2000) E Peru (1990) Fiji (2000) E Congo (1997)
Sudan (1989) Turkey(1980) E Suriname (1990) Pakistan (1999)
Suriname (1980) P Thailand (1991) Sudan (1989)
Uganda (1985)
Low Fiji (2000) E Niger (1996) Nigeria (1983) Burundi (1996) E Niger (1996)
Uganda (1985) Ghana (1981) P Nigeria (1983)
Turkey (1980) E Sierra Leone
(1997)
Thailand (1991)
Uganda(1985)
Missing Suriname (1980) P Comoros (1995) Comoros (1995) Suriname (1980) P Suriname (1990)
Data Congo (1997) Niger (1996)
Suriname (1990) Sudan (1989)
Sources: Reversions: Cheibub, Ghandi, and Vreeland (2010); reversion types: Haggard, Kaufman, and Teo (2012); capital share: Houle (2009); Gini: University
of Texas Inequality Project (2008); family farms: Vanhanen (2003).
509
Inequality and Regime Change August 2012
TABLE 5. Distributive and Nondistributive CGV Reversions, Polity Scores, Prior Coups,
Per Capita GDP, and GDP Growth, 1980–2000
Country/Year
Distributive
Reversions
Nondistributive
Reversions Polity Score Prior Coups GDP/ capita ($) Growth
Bolivia (1980) E −4 3 1070 −1.4
Burundi (1996) E 0 2 113 −8.0
Comoros (1995) X 4 1 386 3.6
Congo (1997) X 5 0 104 −5.6
Ecuador (2000) P 9 0 1295 2.8
Fiji (2000) E 6 1 2075 −1.7
Ghana (1981) P 6 4 224 −3.5
Guatemala (1982) X −5 1 1556 −3.5
Niger (1996) X 8 1 168 3.4
Nigeria (1983) X 7 2 319 −5.3
Pakistan (1999) X 7 0 526 3.7
Peru (1990) X 7 0 1657 −5.1
Sierra Leone (1997) X 4 3 168 −16.7
Sudan (1989) X 7 2 282 8.9
Suriname (1980) P − - 2536 −5.3
Suriname (1990) X − - 2049 −0.5
Thailand (1991) X 3 1 1500 8.6
Turkey (1980) E 9 1 2427 −2.4
Uganda (1985) X 3 2 170 −3.3
% or average 36.8% 63.2% 4.5 1.4 980 −1.6
Notes and sources: E, elite reversion; P, populist reversion. Polity scores are the country’s score the year preceding the
reversion (Marshall, Gurr, and Jaggers 2010). Prior coups is the number of coups (excluding attempted coups or plots) in
the 10 years prior to the reversion (Marshall and Marshall 2010; McGowan 2007). GDP/capita and GDP growth refer to the
values of these variables in the year of the reversion (World Bank 2010). Reversion types are from Haggard, Kaufman, and
Teo (2012).
cases based on a different inequality indicator would
therefore yield similar results.
Distributive Conflict I:
Elite Reactions in Bolivia and Burundi
Bolivia and Burundi are the only high-inequality cases
to revert to authoritarian rule through the causal process
stipulated by the theory. In Bolivia, a right-wing
military faction led by General Luis Garcia Meza
deposed acting President Lidia Gueiler on July 17,
1980, following the victory of leftist Hernan Siles in
an election held earlier that year. The coup occurred
in the context of severe, ongoing conflicts between
militant miners’ unions and more conservative political
and economic forces after the breakdown of the
long-standing Banzer dictatorship in 1978. The Meza
dictatorship was in turn ousted only two years later
by working-class protests that forced new elections.
Significantly, severe distributive conflicts continued to
threaten the stability of the new democratic regime
and ended only in 1985, when the elected government
harshly repressed union opposition and implemented
an aggressive structural adjustment program.
In Burundi, inequality is by no means correctly captured
by the capshare or other inequality measures;
much more significant are the deep ethnic cleavages
that divide the country. A Tutsi minority (about 15% of
the population) had long dominated the military, civil
service, and the economy. Hutus constituted a large
and clearly less well-off majority, producing a highly
fraught political environment (Lemarchand 1996). Between
1966 and 1996, the country experienced no fewer
than 11 coups and attempted coups (McGowan 2007),
with periodic episodes of wider violence. The deposed
democratic government was led by moderate Hutu
politician Melchior Ndadaye, but was extremely fragile;
the coding of the transition to democracy is 1993
is dubious. Ndadave died in an unsuccessful coup attempt
in 1994, and his successor, Cyprien Ntaryamira,
was killed in a suspicious plane crash in the same year.
After a massacre of more than three hundred Tutsis by
radical Hutu rebels in 1996, a military coup by former
president Pierre Beyoya restored the Tutsis to power.
Distributive Conflict II:
Populist Reversion in Ghana
Jerry Rawlings’ coup in Ghana constitutes a clear example
of a populist reversion, although once in office
his military government shifted sharply to the right. In
1981, Rawlings overthrew the feckless constitutional
government of Hilla Limann with the backing of militant
student organizations, unions, and left social movements.
By the time of the coup, the economy had deteriorated
badly, and the Limann government faced strikes
and confrontations with workers over back pay and a
tough austerity program. On seizing power, Rawlings
actively solicited the support of these forces by placing
representatives of radical left organizations on the
510
American Political Science Review Vol. 106, No. 3
military’s Provisional National Defense Council and
creating a raft of populist consultative organizations
(Graham 1985; Hutchful 1997). Rawlings’ populism
only aggravated Ghana’s economic problems, and the
military regime ultimately reversed course entirely and
vigorously embraced the “Washington consensus.” Yet
the initial overthrow of the democratic regime clearly
appealed to, and mobilized support from, populist and
leftist social forces.
The Null Cases: Nondistributive Reversions
The other high-inequality reversions are Peru, Nigeria,
Thailand, and Sierra Leone. Some of these involved
broad appeals that cut across class lines, whereas others
resulted primarily from conflicts within the elite
itself. However, in none of the cases were redistributive
cleavages between elites and masses central to
the reversion, and in several the specific political pressures
stipulated by the theory—redistributive democratic
governments or social movements—were altogether
absent.
Peru. Alberto Fujimori’s decision to close congress
and rule by decree in April 1992 drew support from a
broad cross-section of Peruvian society. Military backing
was, of course, essential and was motivated in part
by the desire for a free hand to confront the Shining
Path, an insurgency that had pretenses of representing
disadvantaged peasants in some highland areas of the
country. Yet in other important ways, the case does not
correspond with the theory. First, the “self-coup” initially
met opposition from international and some local
business sectors—in short, from economic elites—who
were concerned that an outright dictatorship would
have adverse economic consequences. Although these
sectors eventually warmed to the regime after Fujimori
agreed to a fac¸ade of constitutionalism, they were by
no means drivers or even supporters of the coup.
At the same time, Fujimori enjoyed surprisingly wide
popular support, visible in his overwhelming victory in
an early referendum on a new constitution that would
cement his hold on power. The unions and the political
left did oppose the coup, but their organizations had
been decimated by the hyperinflation and economic
collapse of the late 1980s, and they themselves enjoyed
little popular support. The large majority of the Peruvian
poor were attracted by a leader who promised to
deal with a strong hand with the economic crisis and the
insurgency. One 1992 survey showed that almost 76%
of low-income people supported Fujimori’s plan for
constitutional reform (Rubio 1992, 7; cited in Weyland
1996, fn 16). While undertaking economic reforms, Fujimori
also strengthened his electoral base through the
expansion of clientelistic antipoverty programs (Weyland
1996). In the late 1990s, as the economy once
again slowed and corruption scandals surfaced, Fujimori’s
popularity waned, and he was eventually forced
to withdraw from power. Until that time, however, his
government rested on a surprisingly broad cross-class
coalition.
Nigeria. As in Peru, the 1983 coup in Nigeria occurred
in the context of severe economic deterioration
and a widespread loss of public confidence in the
government. The leader of the coup, Major General
Muhaamadu Buhari, was—like his predecessors—tied
closely to the Muslim north and had held a high position
within the deposed government. Yet there are
no indications that the takeover was motivated by
class or ethnic demands on the state, nor by the significant
involvement of civil society. Nor is there evidence
that factional rivalries within the military were
connected with broader social conflicts that could be
modeled in elite-mass terms, whether engaging class,
ethnic, or regional interests. The most consequential
divisions were within the elites, most notably, the military,
clientelistic politicians, and the business class.
When oil revenues collapsed, the ruling coalition fragmented
under competing claims for patronage. The
inability of the hegemonic party to reconcile these
conflicting interests, argues Augustine Udo (1985, 337),
came to a head in a blatantly corrupt election in 1983
that exposed “unprecedented corruption, intimidation,
and flagrant abuse of electoral privilege by all parties.”
The coup was a response to these democratic
failures.
Thailand 1991. The 1991 coup in Thailand was undertaken
by a military faction that bridled under both
the existing military leadership and the efforts of the
elected assembly to exercise greater control over military
spending and prerogatives (Baker and Phongpaichit
2002). Elected officials were concerned, among
other things, with channeling patronage resources to
disadvantaged parts of the country, but they were
linked closely to upcountry business interests. Although
the distribution of income had deteriorated in
Thailand during the economic reforms of the 1980s, left
parties remained confined to the fringes of political life,
and a long-standing rural insurgency had long since petered
out. The coup had the effect of galvanizing mass
opposition, including groups explicitly representing the
poor, and this opposition subsequently played a role
in the transition back to democratic rule. Yet there is
no evidence that the coup either responded to popular
pressures for redistribution or reflected populistauthoritarian
dissatisfaction with democracy’s failure
to redress redistributive grievances.
Weak Democracy Syndrome
We do not seek to elaborate an alternative theory of
democratic instability during the third wave, but our
analysis suggests a “weak democracy” syndrome that
comports with a growing body of literature on democratic
vulnerability (Diamond 2008; Levitsky and Way
2010). Before turning to this issue, however, we should
underscore that at least some of the reversions may be
artifacts of coding rules governing these two influential
datasets. Table 5 shows that 8 of the 19 cases coded as
reversions in the CGV dataset did not rise above the
standard Polity cutoff score of 6 in the year preceding
their collapse (Bolivia, Burundi, Comoros, Congo,
511
Inequality and Regime Change August 2012
Guatemala, Sierra Leone, Thailand, and Uganda). Another
six cases (Fiji, Ghana, Nigeria, Pakistan, Peru,
and Sudan) barely make that threshold with scores of
either 6 or 7. The average Polity score for all of the
CGV reversion countries in the year preceding the
collapse of democratic rule is only 4.5. “Reversions”
are occurring against democracies that are marginally
democratic at best.
Yet the weakness of the distributive conflict theory
of regime change is not simply an artifact of the coding
rules; the causal mechanisms stipulated in the theory
do not appear to operate either. Electoral competition
in Thailand, Nigeria, Pakistan, Honduras, Ecuador,
Ghana, and Guatemala was dominated by patronage
parties with close ties to economic elites or the military
establishment. In none of these cases do we see a significant
presence of parties, interest groups, or social
movements representing the interests of the poor that
could serve as the basis for distributive conflict that
would in turn trigger elite intervention.
Rather, conflicts within the political elite—between
ins and outs—was more likely to pose a challenge to
democratic rule, with the military playing a pivotal
role. In the 11 cases in which distributive conflicts
were implicated in the collapse of democratic rule,
the military could plausibly be seen as an agent of either
elites (elite-reaction reversions) or excluded social
forces (populist reversions). However, in many of the
other cases, the military entered politics largely on its
own behalf. Such intervention was more likely to occur
where prior military intervention had established
a precedent. Cross-national quantitative work on both
Latin America and Africa finds that the likelihood of
a military coup is strongly affected by the previous history
of coups (Collier and Hoeffler 2005; Lehoucq and
Perez-Linan 2009). The data presented in Table 5 are
consistent with these findings. Thirteen of 19 reversions
came in countries that had already experienced at least
one prior coup, and in 7 of these cases, the military was
a repeat offender.
The data in Table 4 also highlight the poverty and
poor economic performance of the countries experiencing
reversion. As Londregan and Poole (1990) and
Przeworski et al. (2000) have shown convincingly, the
probability that democratic governments will survive is
strongly affected by the level of development. Average
GDP per capita for the reversion cases at the time of the
collapse of democratic rule was only $980, way below
the thresholds for consolidated democracies. Among
the non-African cases, only Thailand, Ecuador, and
Peru are middle-income countries.
The relationship between short-run economic performance
and reversions has also been explored in
some detail (Gasiorowsksi 1995; Haggard and Kaufman
1995; Kricheli and Livne 2011; Teorell 2010). Przeworski
et al. (2000) show that the odds of democratic
survival decrease substantially after three consecutive
years of negative economic growth. On average, the
economies of the reversion countries declined by 1.6%
in the year of the reversion, and a number were in the
midst of full-blown economic crises (Table 5). Both low
per capita income and slow growth provided openings
for challengers to act with the acquiescence or even
support from disaffected publics.
In sum, a close examination of the causal mechanisms
driving reversal during the third wave suggests
a more deep-seated syndrome in which distributive
conflict plays a surprisingly minor role. Structural
constraints such as low per capita income and
weak institutions combined with short-run crises seem
to be major factors in the breakdown of these weak
democracies.
CONCLUSION
Viewed over the long run, the emergence of democracy
in the advanced industrial states resulted in part from
fundamental changes in class structures. Demands on
the state from new social classes—first the emergent
bourgeoisie and then the urban working class—played
a role in the gradual extension of the franchise. These
stylized facts played an important role in the new distributive
conflict models of regime change.
Yet these models do not appear to travel well to
the very different international, political, and socioeconomic
conditions that prevailed during the third wave
of democratization. Standard panel designs have found
at best limited evidence for the inequality-transition
logic of the distributive conflict models, and the causal
process observations reported here show that it does
not appear to operate even in cases in which it should.
Although more refined measures of inequality may ultimately
capture ethnic or regional inequalities that
we are underestimating, our causal process observations
are designed to capture at least the overt political
manifestations of a wide array of different distributive
cleavages. It therefore seems likely that the problems
lie with theory as well as measurement.
How should we respond to such findings? Distributive
conflict theories may simply be weaker than their
proponents suggest, and we later highlight several alternative
approaches to regime change. However, the
core insight of distributive conflict theories is intuitively
appealing, and we are inclined to look for avenues for
refinement. One avenue would be to consider whether
inequality influences the stability of democratic rule
through channels other than those postulated by the
distributive conflict theorists. High inequality may be
a determinant of the “weak democracy” syndrome, for
example by contributing to low growth and poverty
(Persson and Tabellini 1994), which are in turn related
to political instability and weak, ineffective states
(Londregan and Poole 1990). This causal path may
well help explain an important class of low-income
cases, as we argued in the conclusion to our discussion
of reversions; we return to this group of countries
later.
Yet the collective protest of citizens against elites is a
core causal mechanism in distributive conflict theories,
and such an approach would abandon that insight altogether.
The incentives and capacity to mobilize such
protest are central to the theory, yet are either assumed
to be a function of levels of inequality or ignored
512
American Political Science Review Vol. 106, No. 3
altogether.12
The free-rider problem highlighted by Olson
decades ago (1965) problematizes the assumption
that shared interests in redistribution will enable large
groups to overcome barriers to collective action. The
question of how to solve this problem has become
the cornerstone of the literature not only on regime
change but also on revolution, collective violence, and
contentious politics.
In the absence of a capacity to overcome barriers to
collective action, transitions both to and from democratic
rule are more likely to reflect narrow, intra-elite
conflicts. Although such conflicts certainly have a distributive
component—and indeed a highly conflictual
one—it is harder to root them in the class-conflict logic
of the underlying Meltzer-Richard model. Although
we found a surprising number of distributive conflict
transitions in the high-inequality cases, the concentration
of income and assets in such settings may also
empower elites to shape the course of regime change;
the effects of transitions on the distribution of income
could as well be regressive as progressive.
We suspect that distributive conflict theories may
ultimately prove to be conditional in form; that is,
they are dependent on incentives and capacities for
collective action that are not in fact given by the level
of inequality. What are these additional factors that
might enable subaltern groups to overcome barriers
to collective action? We identify at least three lines of
research, each of which is potentially complementary
to the political economy approaches discussed in this
article but may also represent competing approaches
to regime change.
It may not be necessary to reach beyond a politicaleconomy
framework to clarify conditions in which distributive
conflict becomes more likely to affect regime
change. One such condition is economic development.
At various points, both Boix (2003) and Acemoglu
and Robinson (2006) suggest that capacities for collective
action are likely to be greater in relatively developed
countries where industrialization and urbanization
provide a social basis for organization. Our
case studies also suggest a contrast between middleincome
countries with substantial concentrations of industrial
labor and poorer countries where low-income
groups are concentrated in the agricultural and urban
informal sectors and face greater barriers to collective
action. In relatively industrialized countries such
as Argentina, Brazil, Poland, South Africa, and South
Korea, distributive conflict transitions involved or were
even led by workers’ movements with a relatively long
history of political mobilization and collective action
(Collier 1999; Drake 1998). Conversely, in a number of
the poorer African transitions we examined, political
parties and civil society groups representing the poor
were often too weak to check the predatory tendencies
of state elites and of other more privileged social
forces; as a result regime change was better understood
in terms of intra-elite processes (Bratton and van de
12 See Green and Shapiro (1994) for a general critique of rational
choice theory and its inability to deal persuasively with collective
action problems.
Walle 1997). Both transitions and reversion in these
cases often came at best in response to generalized
protest against poor economic conditions waged by
relatively better-off urban forces, many with close ties
to the state apparatus.
Institutional approaches represent another point of
departure for explaining collective action. Prior experience
with democracy or institutionalized opportunities
for collective action in semi-authoritarian regimes may
be important for understanding how collective challenges
are subsequently mobilized. In Latin America,
corporatist unions, which had initially been financed
and sponsored by the state (Collier and Collier 1991;
Schmitter 1974), subsequently formed a core component
of protests against authoritarian incumbents in
Peru, Bolivia, and Argentina, as did a number of laborbased
political parties with close links to the state.
At a more general level, differences in authoritarian
institutions might shape both the actors and cleavages
that lead to the establishment or reversal of democracy.
Collier (1999) has shown this empirically with earlier
democratic transitions in Europe and Latin America,
and an exploding literature on varieties of authoritarian
rule raises the possibility for the postwar period
as well (Geddes 1999; Levitsky and Way 2010;
Magaloni and Kricheli 2010). The effects of the organizational
spaces provided by such regimes on collective
demands for democracy remain a subject of ongoing
research. For instance, controlled competition under
semi-competitive regimes might provide opportunities
for mobilization that subsequently spill over into challenges
to the regime itself. Yet it is also possible that
controlled opening may yield advantages for incumbents
by establishing organized channels for recruitment
of supporters, opportunities for control, and the
revelation of politically useful information, such as the
identity and strength of the opposition.
Finally, the social movement and “contentious politics”
literature provides the starkest alternative to
political-economy approaches. Work in this area emphasizes
the significance of political opportunities, resources,
and cultural framing, typically casting the approach
in opposition both to strictly rationalist explanations
for collective action and theories that stress
underlying structural conditions such as inequality (see
McAdam, Tarrow, and Tilly 2003). As noted earlier,
much of the analysis of nationalist and ethnic movements
in Eastern Europe and the former Soviet Union
focuses on the factors emphasized in the contentious
politics literature. Kubik (1994), for example, provides
an important account of how the Solidarity movement
mobilized around a protest discourse that emerged in
the wake of Pope John Paul II’s return to Poland in
1979. More broadly, Beissinger’s (2002) seminal work
on anti-regime protest in the Soviet Union emphasizes
nationalism and ethnic identities, rather than socioeconomic
grievances, as the principal spur to protest
against Soviet authority. As we argued in our discussion
of the coding, such protest can be viewed as a reaction
against other forms of inequality. Yet it can also be
viewed as an alternative to the structural and rationalist
foundations of distributive conflict approaches.
513
Inequality and Regime Change August 2012
The relationship between these analyses of the
sources of collective action and distributive conflict theories
is not straightforward. Economic development,
political institutions, and even the dynamics of contentious
politics may simply mediate the effects of
inequality emphasized in the distributive conflict approaches.
However, given the agnostic nature of our
findings, they might also prove to be contending explanations
that move away from an emphasis on underlying
inequalities altogether. An important line of
research is to exploit opportunities to distinguish between
competing theories that may appear observationally
equivalent.
A final theoretical note concerns an important conditioning
factor in our own analysis, which is limited to
the third wave era. As noted in the introduction, this
period encompasses important international systemic
changes, and a number of our cases had a significant
international dimension. The winding down of the Cold
War eliminated opportunities for authoritarian rulers
to secure support from great power patrons and thus
altered the resources and calculus of domestic political
actors, including disadvantaged groups. With more
permissive international conditions, democracy could
spread to countries with vastly different class structures
and degrees of inequality.
Boix (2011) has recently explored the mediating
effects of the international order and finds that economic
development is more likely to drive democratization
when the international order is dominated by a
democratic hegemon rather than by rivalries between
democratic and authoritarian powers. Against this argument,
our examination of democratization during
this period emphasizes the spread of democracy—at
least temporarily—to poorer countries as well, partly
as a consequence of pressure and encouragement from
international donors. However, we also found that it
was precisely in the poorer countries where transitions
facilitated by positive international conditions were
most likely to be reversed, suggesting that the longer
run structural forces outlined earlier are likely at work.
Distributive conflict models have formalized and
thus given a sharper edge to class-conflict models of
political change that have long been a part of the
social science canon. They have done so in part by
etching more sharply the causal mechanisms that drive
regime change. Yet in doing so, it is incumbent on these
theories to provide compelling empirical tests of their
claims, not only with respect to the relationship between
inequality and regime change but with respect
to the postulated causal processes as well. Although an
earlier generation of case study scholarship on democratic
transitions faced significant selection problems,
large-N cross-national designs have their own disabilities
related to measurement error and reduced-form
designs that do not actually test for stipulated causal
mechanisms.
The tradeoffs between small- and large-N designs are
real. Yet medium-N designs based on causal process
observations provide an additional means of squaring
the methodological circle, particularly where the
phenomena being explained are relatively rare—such
as regime change, revolution, financial crisis, war, or
famine—and thus amenable to intensive qualitative
scrutiny. Such an approach combines within-case analysis
that is sensitive to context and sequence with tests
of the underlying theories and causal mechanisms that
are often only implicit in larger-N designs. However,
this approach rests on a willingness to open up existing
datasets and recode them in line with theoretical
expectations to maximize inferential leverage.
REFERENCES
Acemoglu, Daron, and James A. Robinson. 2000. “Democratization
or Repression?” European Economics Review 44 (4–6): 683–93.
Acemoglu, Daron, and James A. Robinson. 2001. “A Theory of
Political Transitions.” American Economic Review 91(4): 938–63.
Acemoglu, Daron, and James A. Robinson. 2006. Economic Origins
of Dictatorship and Democracy. New York: Cambridge University
Press.
Ansell, Ben, and David Samuels. 2010. “Inequality and Democratization:
A Contractarian Approach.” Comparative Political Studies
43 (12): 1543–74.
Baker, Chris, and Pasuk Phongpaichit. 2002. Thailand: Economy and
Politics. 2nd ed. Oxford: Oxford University Press.
Bates, Robert H., Avner Greif, Margaret Levi, Jean-Laurent Rosenthal,
and Barry R. Weingast. 1998. Analytic Narratives. Princeton,
NJ: Princeton University Press.
Beck, Nathaniel. 2006. “Is Causal-process Observation an Oxymoron?”
Political Analysis 14 (3): 347–52.
Beck, Nathaniel. 2010. “Causal Process ‘Observation’: Oxymoron or
(Fine) Old Wine.” Political Analysis 18 (4): 499–505.
Beissinger, Mark 2002. Nationalist Mobilization and the Collapse of
the Soviet State. Cambridge: Cambridge University Press.
Boix, Carles. 2003. Democracy and Redistribution. New York: Cambridge
University Press.
Boix, Carles. 2008. “Economic Roots and Civil Wars and Revolutions
in the Contemporary World.” World Politics 60 (3): 390–
437.
Boix, Carles 2011. “Democracy, Development, and the International
System.” American Political Science Review 105 (4): 809–28.
Brady, Henry E., and David Collier. 2004. Rethinking Social Inquiry,
Diverse Tools, Shared Standards. Lanham, MD: Rowman and Lit-
tlefield.
Bratton, Michael, and Nicolas van de Walle. 1997. Democratic Experiments
in Africa: Regime Transitions in Comparative Perspective.
New York: Cambridge University Press.
Burkhart, Ross E. 1997. “Comparative Democracy and Income Distribution:
Shape and Direction of the Causal Arrow.” Journal of
Politics 59 (1): 148–64.
Cheibub, Jose Antonio, and Jennifer Gandhi. 2004. “Classifying Political
Regimes: A Six-fold Measure of Democracies and Dictatorships.”
Presented at the Annual Meeting of the American Political
Science Association, Chicago.
Cheibub, Jose Antonio, Jennifer Gandhi, and James Raymond
Vreeland. 2010. “Democracy and Dictatorship Revisited.” Public
Choice 143 (1/2): 67–101.
Collier, David. 1979. The New Authoritarianism in Latin America.
Princeton, NJ: Princeton University Press.
Collier, David, Henry E. Brady, and Jason Seawright. 2010. “Sources
of Leverage in Causal Inference: Toward an Alternative View of
Methodology.” In Rethinking Social Inquiry: Diverse Tools, Shared
Standards, eds. Henry E. Brady and David Collier. Lanham, MD:
Rowman and Littlefield, 161–200.
Collier, Paul, and Anke Hoeffler. 2005. “Resource Rents, Governance,
and Conflict.” Journal of Conflict Resolution 49 (4): 625–33.
Collier, Ruth Berins. 1999. Paths toward Democracy; Working Class
and Elites in Western Europe and South America. Cambridge:
Cambridge University Press.
Collier, Ruth Berins, and David Collier. 1991. Shaping the Political
Arena: Critical Junctures, the Labor Movement, and Regime
Dynamics in Latin America. Princeton, NJ: Princeton University
Press.
514
American Political Science Review Vol. 106, No. 3
Diamond, Larry. 2008. The Spirit of Democracy: The Struggle to
Build Free Societies. New York: Times Books.
Drake, Paul W. 1998. Labor Movements and Dictatorships: The
Southern Cone in Comparative Perspective. Baltimore: Johns Hopkins
University Press.
Dutt, Pushan, and Devashish Mitra. 2008. “Inequality and the Instability
of Polity and Policy.” Economic Journal 118 (531): 1285–1314.
Epstein, David L., Robert Bates, Jack Goldstone, Ida Kristensen,
and Sharyn O’Halloran. 2006. “Democratic Transitions.” American
Journal of Political Science 50 (3): 551–69.
Falleti, Tulia G., and Julia F. Lynch. 2010. “Context and Causal Mechanisms
in Political Analysis.” Comparative Political Studies 42 (9):
1143–66.
Fearon, James D., and David D. Laitin. 2008. “Integrating Quantitative
and Qualitative Methods.” In Oxford Handbook of Political
Methodology, eds. Janet M. Box-Steffensmeier, Henry E. Brady,
and David Collier. Oxford: Oxford University Press, 756—78.
Freeman, John R., and Dennis P. Quinn. 2012. “The Economic Origins
of Democracy Reconsidered.” American Political Science Review
106 (1): 58–80.
Gasiorowski, Mark J. 1995. “Economic Crisis and Political Regime
Change: An Event History Analysis.” American Political Science
Review 89 (4): 882–97.
Geddes, Barbara 1999. “What Do We Know about Democratic Transitions
after Twenty Years?” Annual Review of Political Science 2:
115–44.
George, Alexander L., and Andrew Bennett. 2005. Case Studies and
Theory Development in the Social Sciences. Cambridge, MA: MIT
Press.
George, Alexander L., and Timothy J. McKeown. 1985. “Case Studies
and Theories of Organizational Decision Making.” Advances
in Information Processing in Organizations 2: 21–58.
Gerring, John. 2006. “Single-outcome Studies: A Methodological
Primer.” International Sociology 21 (5): 707–34.
Gerring, John. 2007a. Case Study Research: Principles and Practices.
New York: Cambridge University Press.
Gerring, John. 2007b. “The Mechanismic Worldview: Thinking inside
the Box.” British Journal of Political Science 38 (1): 161–79.
Gerring, John. 2010. “Causal Mechanisms: Yes, but. . ..” Comparative
Political Studies 42 (3): 327–59.
Gervais, Myriam. 1997. “Niger: Regime Change, Economic Crisis
and the Perpetuation of Privilege.” In Political Reform in Francophone
Africa, eds. John F. Clark and David Gardinier. Boulder,
CO: Westview Press, 86–108.
Graham, Yao. 1985. “The Politics of Crisis in Ghana: Class Struggle
and Organization, 1981–1984.” Review of African Political Economy
12 (34): 54–68.
Green, Donald P., and Ian Shapiro. 1994. Pathologies of Rational
Choice Theory. New Haven, CT: Yale University Press.
Hafner-Burton, Emilie, and James Ron. 2009. “Seeing Double: Human
Rights Impact through Qualitative and Quantitative Eyes?”
World Politics 61 (2): 360–401.
Haggard, Stephan, and Robert R. Kaufman. 1995. The Political
Economy of Democratic Transitions. Princeton, NJ: Princeton
University Press.
Haggard, Stephan, Robert R. Kaufman, and Terence Teo. 2012. “Distributive
Conflict and Regime Change: A Qualitative Dataset.”
http://hdl.handle.net/1902.1/18276 V1 (accessed June 1, 2012).
Hedstrom, Peter, and Petri Ylikoski. 2010. “Causal Mechanisms in
the Social Sciences.” Annual Review of Sociology 36: 49–67.
Houle, Christian 2009. “Inequality and Democracy: Why Inequality
Harms Consolidation but Does Not Affect Democratization,”
World Politics 61 (4): 589–622.
Huntington, Samuel P. 1991. The Third Wave: Democratization in the
Late Twentieth Century. Norman: University of Oklahoma Press.
Hutchful, Eboe. 1997. “Military Policy and Reform in Ghana.” Journal
of Modern African Studies 35 (2): 251–78.
Im, Hyung Baeg. 1987. “The Rise of Bureaucratic Authoritarianism
in South Korea.” World Politics 39 (2): 231–57.
King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing
Social Inquiry: Scientific Inference in Qualitative Research. Princeton,
NJ: Princeton University Press.
Kitschelt, Herbert, and Steven Wilkinson, eds. 2006. Patrons or Policies?
Patterns of Democratic Accountability and Political Competition.
New York: Cambridge University Press.
Kricheli, Ruth, and Yair Livne. 2011. “Mass Revolutions vs.
Elite Coups.” Presented at the Annual Meeting of the American
Political Science Association, Toronto. http://ssrn.com/
abstract=1449852 (accessed June 1, 2012).
Kubik, Jan 1994. The Power of Symbols against the Symbols of Power:
The Rise of Solidarity and the Fall of State Socialism in Poland.
University Park: Pennsylvania University Press.
Lambert, Peter. 2000. “A Decade of Electoral Democracy: Continuity,
Change, and Crisis in Paraguay.” Bulletin of Latin American
Research 19 (3): 379–96.
Lehoucq, Fabrice, and Anibal Perez-Linan. 2009. “Regimes,
Competition, and Military Coups in Latin America.” Presented
at the Annual Meeting of the American Political
Science Association, Toronto. http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=1449035 (accessed June 1, 2012).
Lemarchand, Renee. 1996. Burundi: Ethnic Conflict and Genocide.
Washington, DC: Woodrow Wilson Press.
Levitsky, Steven, and Lucan A. Way. 2010. Competitive Authoritarianism:
Hybrid Regimes after the Cold War. New York: Cambridge
University Press.
Linz, Juan, and Alfred Stepan. 1978. The Breakdown of Democratic
Regimes. Baltimore: Johns Hopkins University Press.
Linz, Juan, and Alfred Stepan. 1996. Problems of Democratic Transition
and Consolidation. Baltimore: Johns Hopkins University
Press.
Lipset, Seymour M. 1960. Political Man: The Social Bases of Politics.
Baltimore: Johns Hopkins University Press.
Londregan, John B., and Keith T. Poole. 1990. “Poverty, the Coup
Trap, and the Seizure of Executive Power.” World Politics 42 (2):
151–83.
Magaloni, Beatriz, and Ruth Kricheli. 2010. “Political Order and
One-party Rule.” Annual Review of Political Science 13: 123–43.
Mainwaring, Scott, Daniel Brinks, and Anibal Perez-Linan. 2001.
“Classifying Political Regimes in Latin America, 1945–1999.”
Studies in Comparative International Development 36 (1): 37–65.
Marshall, Thomas Humphrey. 1963. Class, Citizenship, and Social
Development: Essays. Chicago: University of Chicago Press.
Marshall, Monty G., Ted Robert Gurr, and Keith Jaggers. 2010.
Polity IV Project: Political Regime Characteristics and Transitions,
1800–2009. Vienna, VA: Center for Systemic Peace.
http://www.systemicpeace.org/polity/polity4.htm (accessed June 1,
2012).
Marshall, Monty G., and Donna Ramsey Marshall. 2010. Coup d’´etat
Events, 1946–2009. Vienna, VA: Center for Systemic Peace.
McAdam, Doug, Sidney Tarrow, and Charles Tilly 2003. Dynamics
of Contention. Cambridge: Cambridge University Press.
McGowan, Patrick K. 2007. “African Military Intervention Events,
January 1, 1955, to December 31, 2006.” Arizona State University.
Dataset.
Meltzer, Allan H., and Scott F. Richard. 1981. “A Rational Theory
of the Size of Government.” Journal of Political Economy 89 (5):
917–27.
Moore, Barrington, Jr. 1966. Social Origins of Dictatorship and
Democracy. Boston: Beacon Press.
O’Donnell, Guillermo A. 1973. Modernization and Bureaucraticauthoritarianism:
Studies in South American Politics. Berkeley:
University of California Press.
O’Donnell, Guillermo A., Philippe Schmitter, and Laurence
Whitehead, eds. 1986. Transitions from Authoritarian Rule: Tentative
Conclusions about Uncertain Democracies. Baltimore: Johns
Hopkins University Press.
Olson, Mancur. 1965. The Logic of Collective Action: Public Goods
and the Theory of Groups. Cambridge, MA: Harvard University
Press.
Persson, Torsten, and Guido Tabellini. 1994. “Is Inequality Harmful
for Economic Growth?” American Economic Review 84: 600–21.
Pierson, Paul. 2004. Politics in Time: History, Institutions, and Social
Analysis. Princeton, NJ: Princeton University Press.
Przeworski, Adam. 2009. “Conquered or Granted? A History of
Suffrage Extensions.” British Journal of Political Science 39 (2):
291–321.
Przeworski, Adam, Michael Alvarez, Jose Antonio Cheibub, and
Fernando Limongi. 2000. Democracy and Development: Political
Institutions and Well-being in the World, 1950–1990. Cambridge:
Cambridge University Press.
515
Inequality and Regime Change August 2012
Reenock, Christopher, Michael Bernhard, and David Sobek. 2007.
“Regressive Socioeconomic Distribution and Democratic Survival.”
International Studies Quarterly 51 (3): 677–99.
Roberts, Kevin W. S. 1977. “Voting over Income Tax Schedules.”
Journal of Public Economics 8 (3): 329–40.
Romer, Thomas. 1975. “Individual Welfare, Majority Voting, and the
Properties of a Linear Income Tax.” Journal of Public Economics
4 (2): 163–85.
Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John
Stephens. 1992. Capitalist Development and Democracy. Chicago:
University of Chicago Press.
Schmitter, Philippe C. 1974. “Still the Century of Corporatism?”
Review of Politics 36 (1): 85–131.
Slater, Dan, and Benjamin Smith. 2012. “Economic Origins of Democratic
Breakdown? The Redistributive Model and the Postcolonial
State.” University of Chicago. Unpublished manuscript.
Stepan, Alfred 2001. Arguing Comparative Politics. Oxford: Oxford
University Press.
Teorell, Jan. 2010. Determinants of Democratization: Explaining
Regime Change in the World, 1972–2006. Cambridge: Cambridge
University Press.
Udo, Augustine. 1985. “Class, Party Politics, and the 1983 Coup in
Nigeria.” Africa Spectrum 20 (3): 327–38.
University of Texas Inequality Project. 2008. Estimated Household
Income Inequality (EHII) Dataset. http://utip.gov.utexas.
edu/data.html (accessed June 1, 2012).
Valenzuela, Arturo 1978. The Breakdown of Democratic Regimes:
Chile. Baltimore: Johns Hopkins University Press.
Vanhanen, Tatu. 2003. “Democratization and Power Resources
1850–2000.” FSD1216, version 1.0 (2003-03-10). University
of Tampere, Department of Political Science. Finnish Social
Science Data Archive. http://www.fsd.uta.fi/english/
data/catalogue/FSD1216/meF1216e.html (accessed June 1,
2012).
Weyland, Kurt. 1996. “Neoliberalism and Neopopulism in Latin
America: Unexpected Affinities.” Studies in Comparative International
Development 31 (3): 3–31.
Whitehead, Lawrence, ed. 1996. The International Dimensions of
Democratization: Europe and the Americas. Oxford: Oxford University
Press.
World Bank. 2010. World Development Indicators 2010. Washington
DC: World Bank.
516