Cambridge University Press

978-1-107-15849-8 — States in the Developing World

Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

Unpacking States in the Developing World:

Capacity, Performance, and Politics

Miguel Angel Centeno, Atul Kohli,

and Deborah J. Yashar

Capable states are essential for promoting broad-based development;

states must perform certain roles for any society to function. This prop-
osition is now widely accepted. However, this general claim raises a set of
other critical, complex, and poorly understood issues that deserve further
attention. First, we do not adequately understand either the conceptual
content of state capacity or its causal relationship to state performance.
We need stronger and more plausible hypotheses about what explains
state performance, in general, and why states are more effective in some
parts of the developing world than in others, in particular. Moreover, we
do not fully understand why, even among the more capable states, the
ability to provide some valued goods is often in tension with the ability
to provide other valued goods – resulting in varied levels of state perform-
ance across policy domains; in other words, why state capacity is not
necessarily fungible across issue areas is not well understood. In what
follows we initiate an analysis of these issues, delineating both what we
know and what requires further research.
The contributors to this volume take for granted that states matter,
nay, matter deeply, for the pursuit of a variety of valued outcomes in the
developing world. Thus, we shift our scholarly attention to the origins
and types of states capable of promoting these valued goals. Our starting
point is the fairly obvious observation that states in some regions of the

For valued feedback, we thank our contributors and the many people who took part in our
workshops in Princeton, Cape Town, Delhi, Sao Paulo, and Oxford, with a special thanks to
Dinsha Mistree. Of course, our insightful colleagues are not responsible for any of our
analytical missteps.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

2 Miguel Angel Centeno, Atul Kohli, and Deborah J. Yashar

developing world, say, East Asia, tend to be more effective than states in
other regions, such as sub-Saharan Africa, with a variety of other states
in Latin America and South Asia falling somewhere in between on the
performance continuum; we want to know why. We also begin with
the premise that states pursue many political projects but may not be in
a position simultaneously to achieve all of them successfully. We are
especially interested in the ability of states to provide legitimate order,
facilitate effective economic development, and promote social inclusion.
As important, we ask: Can these goals be successfully pursued simultan-
eously or are there inherent tradeoffs between these important goals?
While our normative commitments lead us to hope that these goals can
be achieved simultaneously, our scholarly commitment is to evaluate this
question empirically and theoretically.
For the purposes of this volume, we understand states as a set of
governing institutions embedded in their respective societies. They are
classically understood as a form of organized domination that delivers
order and public goods – whether we reference Hobbes’s Leviathan,
Marx’s committee for managing the common affairs of the bourgeoisie,
Weber’s legitimate monopoly of the use of force, Tilly’s protection
racket, or Olson’s stationary bandit. Thus, to advance the public good,
states are supposed to rise above the variety of private interests that
constitute any society. At a minimum, this involves centralizing the use
of coercion and extracting resources that can then be used to provide
public goods.
As Tilly (1990) taught us, it took warring European states a few centur-
ies to develop these minimal state capacities.1 In the more demanding
conditions of the contemporary developing world, however, the timeline
for forging these capacities has been compressed. In part this is because
developing countries have pursued the creation of Weberian states in
the context of relatively recent postcolonial regimes (where states were
imposed on them) with limited economic resources, and mobilized societies
that demand full citizenship rights. Thus, while developing countries often
do not command significant fiscal, social, or political resources, they are
held responsible for simultaneously facilitating prosperity, redistribution,
and/or inclusion – a point highlighted so powerfully, if differently, by
Marshall (1963), Huntington (1968), and O’Donnell (1993). The demands
on developing-country states are thus formidable and their respective

1
See Tilly 1975, Olson 1984, and Spruyt 1996 among many others for related arguments.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

Unpacking States in the Developing World 3

capacities to deal with these demands, at best, uneven. The scholarly

challenge is to understand the considerable variation in state capacity
across developing-country regions, within regions, within states, and across
policy domains within states, with the hope of discovering some general,
underlying determinants of state performance.
In the rest of the chapter, we will favor the broadest of concepts having
to do with what states actually do, what we call state performance. This is
what, for good or for ill, the state is able to accomplish. Performance is
actually a more analytically neutral category than simple effectiveness,
as the latter presumes an agenda. We take state capacity to mean the
organizational and bureaucratic ability to implement governing projects.
In this opening chapter, we discuss this organizational-bureaucratic vari-
able as analytically prior to state performance, although we revisit this
linear claim in the conclusion. The relationship between capacity and
performance, however, is not automatic. Politics matters. A political
sensibility requires that one consider the political actors that set agendas
and prioritize among competing goals; that deploy particular state agen-
cies to implement those agendas; that mobilize social forces to support
these agendas; and that confront opposition and conflict. In short, both
state capacity and politics must be studied if we are to explain state
performance – especially in the developing world.
In this introductory chapter, we first discuss the conceptual and meth-
odological challenges that emerge as one pursues the systematic study of
determinants of state capacity and performance. To start, we review the
competing definitions of state capacity that exist in the broader social
science literature. We then disaggregate state performance in terms of
major state goals: order, economic management, and inclusion. We sum-
marize what we already know about the political conditions under which
states might facilitate these valued goals, as well as what we do not
adequately understand and what needs further exploration. This general
discussion sets the stage for substantive chapters that probe these under-
explored issues and that constitute the body of the volume. We return in
the concluding chapter to summarize the key new insights generated by
these chapters and to point to areas where further research is needed.

state capacity as a concept

We have suggested that state capacity is critical (alongside a discussion
of politics) to any explanation of state performance. Since the literature
often conflates capacity and performance, we first discuss how the

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

4 Miguel Angel Centeno, Atul Kohli, and Deborah J. Yashar

literature discusses this essentially contested concept before elaborating

how we use the term (Gallie 1956).
The notion of state capacity has existed in the social science literature
for decades. Yet even after significant time and empirical research there
remain competing definitions, competing hypotheses, and competing
methods of measurement. The debate indicates that there is a broad
consensus regarding the importance of institutional and organizational
competence but, without further theoretical and empirical development,
capacity will cease to be a productive, analytical concept. The political
and sociological literature regularly uses the concept state capacity and
often uses it interchangeably with related terminology and ideas, such as
strength, power, and institutions. The notion of capacity is apparently
self-evident and deceptively simple. The problem comes from attempts to
use it in a systematic manner across a variety of cases. Capacity is in
danger of becoming a classic residual variable called upon to explain
unexpected outcomes given particular combinations of causal factors.
One view of state capacity looks at implementation. That is to say,
scholars have analyzed whether state agencies can fulfill their commonly
accepted mission and mandate in terms of organizational design, training,
cohesion, and reach. Fukuyama (2004, 7), in his definition of capacity as
“the ability of states to plan and execute policies and to enforce laws
cleanly and transparently,” makes clear that coordination, planning, or
coming to a consensus is a part of a state’s capacity. In this regard, capacity
is understood as a function of the organization in question. Implicitly this
focus on the organization is understood relative to its autonomy from
civil society and its ability to pursue and impose outcomes without
societal interference. The range and implementation of state actions are
thus decided internally within the state.
Another view of state capacity might look at the scope and/or content
of what a state pursues. What are the goals of the state and how expansive
are they? Lowi (1964) long ago analyzed state policy functions in terms of
whether they regulate, distribute, and/or redistribute. Mann (1984, 188)
argued that “despotic power” is “the range of actions which the elite is
empowered to undertake without routine, institutionalized negotiation
with civil society groups.” Here, the range of state action is curtailed by
civil society and the notion of capacity is embedded in notions of demo-
cratic rule. Such a scope moves us away from implementation to the
negotiations within the state and between it and other actors regarding
the level, type, and form of intervention in society. An alternative perspec-
tive on scope is exemplified by the World Bank’s “quality-of-governance”

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

Unpacking States in the Developing World 5

approach, which focuses on the process of decision-making, the imple-

mentation of decisions, and the legitimacy thereof (Holmberg, et al. 2009,
137). In other words, state capacity in this regard is concentrated on a
specific mandate, or perhaps even an ideology.2
A relational approach is adopted by those who look at state capacity in
terms of power. Of course, there are many ways to analyze power. But
among the most influential have been those articulated by Dahl (1957) in
comparative politics and Baldwin (1979) drawing on Dahl in inter-
national relations. The assumption in this literature is that power (or state
capacity) refers to the ability to get others to do things that they might not
have done otherwise. This forms the basis of a relational understanding of
state capacity. Kugler and Domke (1986, 39), for example, defined power
in international politics as “the ability of one nation to exercise control
over the behavior or fate of another.” Migdal (2001, xiii) defined capacity
as “the ability of state leaders to use the agencies of the state to get people
in the society to do what they want them to do.” Lukes (1974) expanded
the discussion to include three levels: policy choices, agenda setting, and
discursive hegemony. Skocpol (1979) has argued that state strength
comes from autonomy from civil society and its power holders. Later,
she argued that state capacity is a function of state autonomy, integrity,
bureaucratic refinement, and resources (Skocpol 1985). Geddes (1994)
adds concerns with political consensus and ideology. She argues, “If one
wants to explain a state’s preferences regarding development strategies,
for example, one needs to know who has power and what they want and
believe” (ibid., 6). Mann (1984, 189), moreover, has defined infrastruc-
tural power as “the capacity of the state to actually penetrate civil society
and to implement logistically political decisions throughout the realm.”
Perhaps the most prevalent approach reclaims Max Weber. It analyzes
the organizational competence of the bureaucracy. Evans and Rauch
(1999) are instructive here. They assess the quality of bureaucracy in
select developing countries by a series of reputational surveys. There is
much to applaud about this effort. In contrast to many other studies

2
At present, crossnational scholars measure state capacity either subjectively – by asking
those familiar with a country if the state in that country is more or less efficacious and then
assigning a number to that assessment – or in terms of its impact, say, on the tax revenues a
state may generate. Neither of these approaches is without problems, even serious prob-
lems. For example, some scholars use the World Bank’s “governance measures” to
measure state capacity; doubts about the quality of these measures, however, are growing
(Kurtz and Schrank 2007). For a full discussion of related problems, see Enriquez and
Centeno 2012.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

6 Miguel Angel Centeno, Atul Kohli, and Deborah J. Yashar

(including subsequent efforts by the World Bank Governance Indicators),

the authors are careful to separate capacity from outcomes. They col-
lected scholarly expertise into a measurable set of indicators. This meas-
ure is neither recent nor readily available for a large set of developing
countries, although there have been some recent efforts to update them
(Dahlström, et al. 2012). We find this “quality-of-bureaucracy” approach
compelling and use it as a springboard for our subsequent discussion.
Together, these four competing notions of state capacity – the ability of
the state to achieve its own identified goals (implementation), the ability
of the state to achieve an ideal set of goals usually determined by an
outside party (scope), the ability of a state to impel citizens and other
states to do what they may not have done otherwise (relational power),
and the organizational competence of the civil servants (quality of bur-
eaucracy) – represent the historical contours of scholarly studies on the
subject. A more useful term may be capability, as it implies a potential
use, but we will use the word “capacity” throughout – so as to facilitate
dialogue with the extant literature.
Yet when we juxtapose these approaches, it forces us to consider
whether we all too often elide these various components: conflating the
concept with its causes and consequences. Much of the implementation
and scope literature is explicitly concerned with the outcomes (what we
have called performance). Scholars in these camps assess state capacity by
considering the extent to which the state is providing social development
or achieving growth. Is violence in check? Is there a functioning democ-
racy? Do citizens have greater freedoms than citizens in other states? So,
too, the relational approach also focuses on outcomes. Before making
arguments for the contributions (or dangers) of state capacity, we need to
be analytically clear about what it is that we are after. If capacity has any
meaning in and of itself then it is something that a state should possess
independent of its outcomes. For this reason, we focus on state capacity as
the quality of its bureaucracy, independent of whether it is deployed and
to what end; while normatively jarring, the test might be whether we can
and/or should identify and compare state capacity across politically dis-
tinct kinds of states (democratic, developmental, communist, and maybe
even the apartheid state of South Africa, for example) and issue areas
(immunization, education, and genocide, for example). In this sense, state
capacity as quality-of-bureaucracy is analytically neutral, although it can
be used to achieve normative ends (both desirable and heinous in charac-
ter). The Nazi German, South African apartheid, and Chilean/Pinochet
states have often been characterized as highly capable. Politics, however,

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

Unpacking States in the Developing World 7

determined how that state capacity was deployed and to what end.
Indeed, history has all too clearly demonstrated the ways in which distinct
sets of political actors have deployed these states for both normatively
positive and more strikingly for normatively deplorable ends.
In analyzing state performance, therefore, we must analyze state capacity
alongside the political actors who seek to deploy it. Even Weber was very
much of a skeptic regarding “the state” acting autonomously and with its
own agency. The surviving notes from his last lectures on the state clearly
indicate, however, that Weber purged the state of all collective agency.
Weber implies that the state, and particularly the rule of law, is a façade
covering the reality of relations of power. The state is a form of organized
domination by some over others pursuing a means to some end. According
to Weber, the state is a tool for the purposes of domination. The state is an
“enterprise” or an organization (Betrieb). As an instrument of power, the
state can be used by different groups (including and in particular the state
cadre) for a variety of purposes, but it is imperative to understand that it is an
instrument, not a goal. The state is merely one possible organizational
embodiment of social relations; it represents the institutionalization of rela-
tions of domination. In defining and analyzing state capacity, therefore, it is
also critical to place a state within the appropriate political context.
We contend that the task before us is to discipline our discussion by
disentangling what state capacity is from its causes and consequences.
First, the conceptual challenge is to focus on state capacity as an organiza-
tional feature of bureaucracies. Second, the causal challenge (which
ultimately motivates this discussion) is to articulate a template for analyz-
ing “why” questions: Why do some states develop state capacity and why
do they deploy that capacity toward different ends? Why do some efforts
receive popular support while others do not? Why do we see differential
performance, with some state efforts resulting in success while others
result in failure?

unpacking the state: disentangling capacity from

its origins, deployment, and performance
A common critique of notions of state capacity is the danger of tautology
or the blurring of the line between causes and outcomes. The threat of
tautology creeps in particularly during attempts to operationalize the
concept for empirical work. It is simple enough to conceive of capacity
as a variable, with states having more or less capacity at any given time
in any given area. However, when we seek to assess state capability

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

8 Miguel Angel Centeno, Atul Kohli, and Deborah J. Yashar

empirically – quantitatively or qualitatively – we often use outcome

variables to judge capacity (for example, capacity to collect taxes), poten-
tially introducing circular logic into the heart of our analyses.
How does one separate the conceptualization and measurement of
state performance from the causes and results of capacity? We propose
that the first step toward a better understanding of state performance is
the explicit division between the organizational structures that define our
notion of capacity (Weber’s Betrieb), the conditions that may activate
(or impede) it, and the results that it may produce. In more simple
language, we believe that state capacity in and of itself is best understood
when we exclusively focus on the organizational capability of the state.
If we are to understand a state’s overall performance (results), we have to
recognize the forces that define the contours of state capacity (origins),
how this state capacity is deployed, and how it is received in its political,
social, and economic environment.
Organizational capacity is, then, only one factor that affects a state’s
overall performance. If we are interested in identifying how state capacity
fits into a state’s overall performance, we should recognize that state
performance is a product of the following relationship – although in
reality the relationship between these variables is highly interdependent
and is neither linear nor teleological.3
ORIGINS ! ORGANIZATIONAL ðSTATEÞ
CAPACITY ! POLITICAL DEPLOYMENT ! PERFORMANCE

Origins
We started off this project suggesting that deep historical processes (for
example, patterns of colonialism, types of political economies, socioeco-
nomic inequalities, wars, struggles of national liberation, and ethnic rela-
tions) have something to do with the development of state capacity. By
contrast, some economists have put forward the general proposition that
institutions originate and exist because they help capture gains from
cooperation (North 1990). We do not share this theoretical stance; not
only does such functionalist thinking ignore the deep causal role of
coercion in the creation of states and state institutions (Thelen 2004), but
it also tends to be too general and ahistorical. Rather, we focus on the

3
Although we adopt a linear approach for ease of presentation in this chapter, the Conclu-
sion to the book revisits this assumption by analyzing state capability and performance as
a more dynamic and variegated process that can not only scale up but also deteriorate.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

Unpacking States in the Developing World 9

politics of creating institutions, recognizing the struggles that occur in the

process. As such, we associate ourselves with the view that institutions
originate at historically specifiable “critical junctures” where specific cleav-
ages are politically addressed (R. Collier and Collier 1991). In turn, they
tend to exhibit “path dependence” (Pierson 2004), although on this last
point even North would agree. More specifically, one might propose that
the process of state formation in the developing world has proceeded in a
series of “big bangs,” with formative moments few and far between,
though incremental changes have certainly altered power configurations
within states and, at times, even accumulated to yield basic changes (Thelen
2004). That the latter process is rare is understandable, given that state
formation generally requires a preponderance of force in the hands of some
to impose their preferred design on others for a long enough period that
basic institutions take root. That is why wars are deemed to be so import-
ant an agent of state development, especially in the context of European
countries. By contrast, the historical forces that have molded basic state
forms in the developing world have, we contend, been colonialism, nation-
alist movements including radical revolutions, and other types of forceful
regime changes, especially militaries moving in and out of power. Incre-
mental changes have in turn been pushed by political parties, by social
classes and movements, and, on occasion, by external actors.
In this introduction, we do not seek to identify whether critical junc-
tures matter more or less than incremental changes. Rather we pose the
question of origins as a “placeholder” to flag this crucial issue. Some of
the following chapters analyze the origins of a broad variety of states and
how the historical legacy of founding may yet determine a state’s present
capacity potential.

Organizational (State) Capacity

In its simplest terms, state capacity involves the bureaucratic, managerial,
and organizational ability to process information, implement policies,
and maintain governing systems. State capacity is thus a function of the
organizational skills and institutions required for carrying out relevant
tasks. In particular, we identify organizational capacity in terms of the
following indicators:4

4
Organizational capacity may be indicated by “outcome” variables, but is not determined
by them. For instance, one might suggest that a state’s resources depend on its ability to
collect revenues but, without a competent and rule-following bureaucracy with high
organizational capacity, these revenues will be plundered by venal government officials.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-15849-8 — States in the Developing World
Edited by Miguel A. Centeno , Atul Kohli , Deborah J. Yashar , Assisted by Dinsha Mistree
Excerpt
More Information

10 Miguel Angel Centeno, Atul Kohli, and Deborah J. Yashar

1. Resources: To what degree do states have the resources required to

fulfill their stated mandates? While resources do not equal outcomes
(indeed, we know that inflated budgets can reduce efficiency), there
is a resource floor below which institutions cannot function. A state
pursuing public education, for example, lacks organizational cap-
acity if it cannot pay for teachers, books, or buildings.
2. Presence of the state: To what degree do states penetrate their
societies? Mann (1993) highlighted the importance of the infra-
structural reach of the state. Mann’s critical insight came from
recognizing that infrastructural power comes from increasing the
level and quality of contact citizens have with the state. Organiza-
tional capacity is partly conditioned by productive interactions that
take place throughout a country’s territory.
3. Mandarins: To what degree do states have a trained and profes-
sional civil service? This entails an understanding not only of the
level of education and training for street-level bureaucrats, but also
the issue-area expertise of those who head these institutions. The
question here is not only of the presence of trained personnel (from
basic literacy to advanced degrees), but also that these personnel be
stable and entrusted with appropriate responsibilities.
4. Coherence: Organizational capacity is also a function of institu-
tional coherence, defined as inter- and intra-institutional communi-
cation and oversight. This entails both the coherence of mandates
across and within institutions (meaning that institutions mandated
to implement land reform are all committed to doing so) and the
oversight to assure that civil servants pursue that mandate (mean-
ing that corrupt, captured, or ineffective civil servants will be
penalized if they do not pursue institutionally identified goals, while
meritorious civil servants are recognized with rewards and promo-
tions). It involves the system of controls and incentives created
within the state bureaucracy in order to assure communication of
directives, enforcement of rules, performance feedback, and general
oversight. Lack of coherence is perhaps the most visible failure for
many civil services. Our fourth component of organizational cap-
acity may be the least analyzed and yet most important.

It is worth emphasizing that this conceptual understanding of organiza-

tional capacity is divorced from goals, mandates, and ideology. Yet if we
fail to understand why and how organizational capability is deployed in
certain ways and not others, we cannot fully understand the causes and

© in this web service Cambridge University Press www.cambridge.org