MIT LIBRARIES
DUPL
3 9080 02617 7920
lii
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium
Member
Libraries
http://www.archive.org/details/kleptocracydividOOacem
31
:4i5
1
'\
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
Kleptocracy and Divide-and-Rule:
A Model
of Personal Rule
Daron Acemoglu
James Robinson
Thierry Verdier
Working Paper 03-39
July 2003
Room
E52-251
50 Memorial Drive
Cambridge,
This
MA 021 42
paper can be downloaded without charge from the
Social Science Research Network Paper Collection at
http://ssrn.com/absti-act=471 821
A
Kleptocracy and Divide-and-Rule:
Model
of
Personal Rule*
James A.
Daron Acemoglu^
Thierry Verdier^
Robinson-'-
July 2003
Abstract
Many
developing countries have sufFered under the personal rule of "kleptocrats"
implement highly
inefficient
use the proceeds for their
economic
own
policies, expropriate the
glorification or
consumption.
We
argue that the success
Members
operate in order to depose a kleptocrat, yet such cooperation
punitive rates of taxation on any citizen
the benefits to those
be
intensified
who need
made
by threats which remain
of society need to co-
may be
off
it.
Thus the
collective action
the equihbrium path.
buy
off
and
rents
from natural resources provide
redistributing
problem can
In equilibrium,
exploited and no one challenges the kleptocrat. Kleptocratic policies
foreign aid
defused by imposing
who proposes such a move, and
to agree to
who
wealth of their citizens, and
of kleptocrats rests, in part, on their ability to use a "divide-and-rule" strategy,
possible by weakness of the institutions in these societies.
,
Eire
more
all
likely
are
when
rulers with substantial resources to
opponents; when opposition groups are shortsighted; when the average productiv-
ity in the
economy
is
low;
and when there
(because more productive groups are more
JEL
is
greater inequality between producer groups
difficult to
buy
off).
Classification: 012, HOO.
Keywords:
dictatorship, divide-and-rule, economic development, institutions, klep-
tocracy, personal rule, political economy.
in
* Paper presented as the Marshall Lecture at the European Economic Association's Annual meetings
Stockholm, August, 24th, 2003. We thank Silje Aslaksen for pointing out an algebraic mistake in the
first
version and Alexander
Debs
for excellent research assistance.
'^Massachussetts Institute of Technology, Department of Economics, E52-380, 50 Memorial Drive,
Cambriudge MA02142. e-mail: daron@mit.edu.
* Department of Political Science and Department of Economics, University
210 Barrows Hall, Berkeley CA94720. e-mail: jamesar@socratcs.berkeley.edu
§DELTA-ENS, 48 Boulevard Jourdan,
of California at Berkeley,
Paris 7.5014, France, e-mail: tv@java.ens.fr.
1
Introduction
1
Many
countries in Africa
the state
their
is
controlled
and the Caribbean
and run
for the benefit of
power to transfer a large fraction of
under "kleptocratic" regimes, where
suffer
an individual, or a small group, who use
society's resources to themselves.
Congo
kleptocratic regimes include the Democratic Republic of the
(Zaire)
Examples
of
under Mobutu
Sese Seko, the Dominican Republic under Rafael Trujillo, Haiti under the Duvaliers,
Nicaragua under the Somozas, Uganda under
and the Philippines under Ferdinand Marcos.
appear to have been disastrous
for
Amin, Liberia under Charles Taylor,
Idi
In
all
these cases, kleptocratic regimes
economic performance and caused the impoverishment
of the citizens.
A
study of the political economy of such regimes must depart from the standard
presumptions of most research
rulers
make
in
economics and
political science,
which assume that
choices within "strongly-institutionalized polities".^ In these polities, formal
political institutions,
such as the constitution, the structure of the legislature, or electoral
on the beha\aor of
rules, place constraints
politicians
and
political ehtes,
and directly
influence political outcomes. In contrast, kleptocracy emerges in "weakly-institutionalized
polities,"
where formal institutions neither place significant
actions nor
make them accountable
restrictions
on
politicians'
to citizens.
While the academic study of strongly-institutionalized
polities is well
advanced
(e.g.
Shepsle and Weingast, 1995, Cox, 1997, Persson and Tabellini, 2000, 2003), there are
few studies, and
What
less of
a consensus, on the nature of weakly-institutionalized
polities.
determines corruption, rent extraction and bad poUcies when institutions are weak?
Indeed, the qualitative nature of poUtics appears to differ markedly between strongly and
weakly-institutionalized polities:
when
institutions are strong, citizens punish politicians
citizens
who
institutions are strong, politicians vie for the support
and
by voting them out of power; when institutions are weak, pohticians pimish
fail
to support them.
When
endorsement of interest groups; when institutions are weak, pohticians create and control
interest groups.
When
institutions are strong, citizens
demand
rights;
when
institutions
are weak, citizens beg for favors.
The
research
program proposed
in this
in weakly-institutionalized societies,
^We owe
this
paper
is
a systematic study of pohcymaking
and ultimately, a study of the process via which
terminology and the distinction between strongly- and weakly-institutionalized polities
to Robert Powell.
North and Weingast, 1989,
strongly-institutionalized societies emerge (see
account of such a process). To take a
study kleptocratic
politics.
step in this program,
first
a classic
for
we construct a model
to
Perhaps the most puzzling feature of kleptocracies, illustrated
by the examples from the Congo, the Dominican Republic, Nicaragua or
longevity, despite the disastrous policies
pursued by the
Haiti,
This longevity
rulers.^
their
is
is
made
even more paradoxical by the fact that such regimes apparently lacked a political base
(a core constituency) that supported them.
mechanisms
deposing unpopular
Despite the absence of formal institutional
on the behavior of
rulers exist
even
in weakly-institutionalized societies (e.g., the threat of revolution, or competition
from
for
other strongmen).
Why
rulers, constraints
do, then, the heavily-taxed producers or the poverty-stricken
Wliy do they rarely form an
citizens not replace the kleptocrat?
constraining the kleptocrat?
How
effective opposition
can a regime that apparently benefits nobody outside
of the narrowest of cliques survive?
Our
basic answer
is
that this
is
because the absence
of strong institutions allows rulers to adopt political strategies which are highly effective
at defusing
any opposition to
their regime.
The seminal book by Robert
many
clues towards
in effect in
many
the following
Bates, Markets and States in Tropical Africa, provides
an answer. Bates described the web of
parts of Africa, but most notably in
logic:
many
inefficient transfers
and
Ghana and Zambia, and suggested
of these inefficient policies are in place to transfer resources
from the population to the ruling groups, while at the same time ensuring their
survival. In particular, the
nexus of
inefficient
politically mobilized against the rulers
could be punished, while those that remained loyal were rewarded.
Ghanaian government heavily taxed cocoa producers, while
and
The
fertilizers.
at the
With
this logic, the
same time subsidizing
subsidies could be allocated selectively as a
potential reward for not attempting to change the status quo.
rate
political
pohcies appeared to be useful for creating
an environment where any group that became
their inputs of seeds
policies
was kept overvalued because then the government could
Similarly, the
exchange
allocate or withhold valuable
rations of foreign exchange in order to guarantee support.
In this paper,
we suggest a
generalization of this reasoning, which
and-rule" strategy. Divide-and-rule
is
we dub the
"divide-
a method used by kleptocrats to maintain power in
weakly-institutionalized polities while simultaneously pursuing pohcies costly to society.
The
logic of the divide-and-rule strategy is to
^Despite
the
appealing
intuition,
institutionalized societies, that rulers
perhaps derived from many
who reduce
and the Somozas
for
political
analyses
of
strongly-
the utility of citizens ought to be replaced, this appears
not to be the case in weakly-institutionalized societies.
for 32 years, Trujillo for 31,
enable a ruler to bribe pohtically pivotal
Many
42 years.
kleptocrats rule for long periods,
Mobutu
groups
off
the equilibrium path, ensuring that he can remain in power against chaUenges.
To remove a
is
made
ruler
difficult
incentives
from power requires the cooperation of distinct social groups which
by the
problem (Olson, 1965). By providing
collective action
and punishments, the divide-and-rule strategy exploits the
cooperation:
when
selective
fragility of social
faced with the threat of being ousted, the kleptocratic ruler intensifies
the collective action problem and destroys the coalition against him by bribing the pivotal
groups.
More
groups.
exphcitly,
The
we consider a dynamic game between the
and potential foreign
consumption. The two producer groups,
power and establish democracy
this cooperation as follows:
remove the
ruler
(a regime
is
aid from outside donors for his
own
they can cooperate, can remove the ruler from
more favorable
to their interests).
We
model
one of the two groups (the "proposer") makes a proposal to
from power, and
removed and democracy
if
and two producer
and uses the ensuing tax revenue, the
kleptocratic ruler taxes production
rents from natural resources
ruler
if
the other gi^oup (the "proposed") agrees, the ruler
established.
The
is
ruler-friendly political institutions, however,
imply that before the proposed group responds to the proposal, the ruler can make a
counteroffer.
This counteroffer enables him to use a divide-and-rule strategy: following
a challenge, the ruler uses
all his
resources and the tax revenues to bribe the proposed
down
group (and compensate them
for future
keep the ruler in power).
he can do so successfully, he can fight off the challenge,
and anticipating
this,
If
no group
higher taxes
if
they turn
will challenge the ruler.
strategy remains off the equilibrium path,
foUow highly distortionary (kleptocratic)
and
its
policies
the proposal and
Therefore, the divide-and-rule
anticipation imphes that the ruler can
without being challenged. Not only
is
the kleptocrat able to stay in power, but the tlu'eat of divide-and-rule implies that there
wiU be no challenges to remove him from power along the equilibrium
In addition to providing an answer to the question of
vive
and formalizing the idea
of divide-and-rule, the
interesting comparative static results.
political institutions,
These
results
how
model
path.'^
kleptocratic regimes suralso implies a
number
of
stem not from varying the formal
but from differences in the feasibility of divide-and-rule depends
on the socio-economic environment. In
particular, they follow directly from the logic of
divide-and-rule: kleptocracy survives
off the
the pivotal group.
^We
if,
Consequently, kleptocracy
eqmhbrium
is
more
path,
hkel}^:
(a)
it
can successfully bribe
when producer groups
conjecture that these results extend to an environment with uncertainty, where, along the lines
Green and Porter (1984) and Abreu, Pearce and Stacchetti (1990), punishment will
sometimes occur along the equilibrium path. Obviously, most of the examples in the case study literature
feature situations where divide-and-rule is actually observed in practice.
of the analysis in
.
or their political representatives are
future returns under democracy; (b)
he can use to bribe pivotal groups;
more
when
(c)
can be used to bribe pivotal groups;^
(loosely speaking, because this
more
there
when
(d)
is
more
there are greater natural resource rents that
when average
productivity in the economy
makes natural resource
is
rents
number
of African
common
when
there
is less
inequality
about the experiences of a
static results are useful in thinking
and Caribbean countries.
First, kleptocratic
regimes seem to be more
such as the Congo, Liberia, Sierra Leone, and
consistent with our result that, as long as producer groups are not too
is
patient, greater natural resource rents faciUtate kleptocracy.
shows that, other things equal, the presence of
will
and funds from foreign aid
to gain from democracy. If inequality increases,
in natural-resource-rich countries
Nigeria, which
low
the ruler to sustain his kleptocratic policies.
difficult for
These comparative
is
because the main challenge to the ruler comes from the
more productive group, which has more
becomes more
on
foreign aid to the ruler, which
effective instruments to bribe the pivotal group) ;'^ (e)
between producer groups. This
it
shortsighted, because they put less weight
oil
For example, Ross (2001)
increases the probabihty that a country
be a dictatorship.^
Second,
many
accounts of the worst kleptocracies also illustrate
how
the ruler was
able to use foreign aid (as well as rents from natural resources) to stay in power
off challenges.
The
Mobutu, which we discuss
case of
in the next section,
is
and
fight
especially
pertinent here. This negative role of foreign aid on equihbrium poHcies might also explain
the coirmion finding in the empirical growth literature that foreign aid seems to have
positive effect on the rate of economic growth in less developed countries,
might even have a negative
Easterly, Levine
evidence
(e.g.,
^However,
effect
on non-democracies
(e.g.,
and Roodman, 2003). In addition there
Dohar and
is
and
little
in fact,
it
Burnside and Dollar, 2000,
a large amount of anecdotal
Pritchett, 1998, Brautigam, 2000, Easterly, 2001, van der Walle,
this result requires
producer groups not to be too patient.
Otherwise, the prospect of
obtaining some of those natural resource rents in the future, once the ruler
sufficiently attractive to oust the ruler,
making kleptocratic
is
deposed,
may make
it
policies harder to sustain.
^However, as with the comparative static with respect to natural resource rents, the effect of average
if citizens axe very impatient, greater
productivity on policies depends on the discount rate of the citizens:
average productivity
may make
kleptocracy more
likely.
"^Moreover, the case study evidence also suggests that whether or not resources are a curse depends
on institutions (see Ross, 1999, for a survey). Similarly, Moene, Mehlum and Torvik (2001)
show that the negative effects of resources pointed out by Sachs and Warner (1995) are conditional on
institutions
only countries with poor institutions, specifically weak constraints on the executive, suffer
a resource curse. In our model this follows from the fact that once political institutions do not constrain
rulers, natural resources help to facilitate divide- and-rule (in this context, see also Robinson, Torvik and
Verdier, 2002, for a related argument that when institutions are weak, resource booms are a curse because
crucially
—
they intensify incentives for regimes to use inefficient forms of redistribution to stay in power)
2001),
and some tentative econometric evidence (Knack, 2000) that
foreign aid leads to
worse governance.
The comparative
static result
pretation of this result
power between the
is
ruler
with respect to inequality
that kleptocracy
is less
is
also interesting.
likely to arise
and a powerful producer group
when
in society.
there
This
is
is
One
inter-
a balance of
consistent with
the experiences of a number of more successful economies in sub-Saharan Africa, such
as
Botswana and Mauritius, which have both managed
and
refrain
power
of
from the most distortionary
A
policies.
to
become
stable democracies
possible explanation
major producer groups, such as the cattleowmers
in
is
the political
Botswana and the sugar
planters in Mauritius, which has placed real constraints on the behavior of political elites
example, Leith, 2000, and Acemoglu, .Johnson Robinson,
(see, for
and Bowman, 1991, on Mauritius). Our
that economic pohcies were
less
result
is
distortionary in
200.3,
on Botswana
also consistent with Bates' emphasis
Kenya than
in
Ghana, because
large
landowners in Kenya were better able to solve the collective action problem and act as an
important force
in politics.
The
interpretation suggested
by our model
is
that, in
Kenya,
the presence of a social class with sufficient wealth and power (thus greater inequality
among
producers) counterbalanced the power of the ruling party
KANU
(Kenya African
National Union) under .Jomo Kenyatta.'
In addition to these comparative static results,
we should emphasize
that kleptocratic
rule can be sustained because of two features of our political game: (1) the ability of the
ruler to
make a
counteroffer after the proposal of one of the groups; (2) the ability of
the ruler to charge different
The
first
feature
exploit their
context of
is
tcix rates
less
differential transfers to the
two groups.
plausible in a weakly-institutionalized society where politicians can
power to consolidate
many
and make
their rule.
The second
feature
is
also plausible in the
developed countries where ethnic, geographic and economic divisions
enable policies with very different distributional consequences for various groups to be
implemented.
In the next section,
were important
This paper
in the survival of
is
related to a
we provide a
some
brief discussion of
economy
models are similar to ours
number
literature
in that
these divisions
kleptocratic regimes.
of different hteratures.
institutionalized polities have received little attention
stance, the political
how
by
on dictatorships
As we
noted, weakly-
political economists.
is
For
very underdeveloped.
in-
Most
they formalize dictatorship as the maximization of an
individual's or group's utihty function subject to the constraint of staying in power (e.g.
^In this context,
it is
also interesting that in the
most extreme examples of kleptocracy
in the
Americas,
the Dominican Republic, Haiti and Nicaragua, revolution and U.S. intervention destroyed traditional ehtes
that might have blocked the power of kleptocrats.
North, 1981, Levi, 1988, Grossman, 1991, Grossman and Noh, 1994, McGuire and Olson,
1996, Robinson, 1998, Wintrobe, 1998,
Bueno de Mesquita
Bates, 2001,
Acemoglu and Robinson, 2000b, La Ferrara and
et al., 2003,
do not capture the notion that the nature of
than in strongly- institutionalized
this
polities.
Acemoglu, 2003).^ However, these models
politics
under weak institutions
Our model develops
can lead to the divide-and-rule strategy
this idea
is
different
and shows how
in weakly-institutionalized societies,
and how
divide-and-rule leads to very different comparative static results.
Political scientists often distinguish
between
different types of "autocracies" or dicta-
and Stepan (1996) posits the existence
torships. For example, the seminal study of Linz
and
of four types, totalitarian, post-totalitarian, authoritarian
which
is
close to
by pohtical
what we mean by kleptocracy,
sultanistic.
also equivalent to other concepts used
is
Scholars working in African politics, for instance,
scientists.
Sultanism,
I'efer
to this as
"neopatrimonialism", "personal rule", or "prebendalism". In the words of Bratton and
van der Walle (1997,
p.
person rather than to an
individual
its laws.
and
despite the official existence of a written constitution.
life,
One
dominates the state apparatus and stands above
occupy bureaucratic positions
officials
they also enjoy access to various forms of
corruption, which constitute
circle
an entitlement of
...
undermine the effectiveness
for systematic
patronage and
That such neopatrimoniaUsm
wisdom
is
in political science (see
with patrons,
less to
perform public service
it is
illicit
The
modern
rents, prebends,
and
his inner
state administration
by using
chief executive
order to maintain political order."
at the heart of Africa's
economic
crisis is
by the
and
rule
rivals,
is
who
"a system of relations linking rulers
constitute the 'system.' If personal
limits of their personal authority
and power and by the
—
is
'structured'
...
the conventional
Sandbrook, 1985, Herbst, 2000, Bates, 2001). Jackson and
authority and power of patrons, associates, clients, supporters, and
The systems
and petty
clientelist practices in
clients, supporters,
rulers are restrained,
office.
of the nominally
Rosberg (1982, pp.17-19) note that personal
...
ascribed to a
than to acquire personal wealth and status. Although state functionaries receive an
official salary,
it
is
Relationships of loyalty and dependence pervade a formal political and admin-
istrative system,
...
office,
often a president for
...
neopatrimonial regimes
62): "the right to rule in
—
of course
rivals.
not by institutions, but by the politicians themselves."
^There are interesting comparative
static results
from the literature on dictatorships. Existing ideas
suggest that better economic policies are chosen by dictators with long time horizons (Levi, 1988, Gross-
man and
Noh, 1994), who are "encompassing" in the sense that a large fraction of income accrues to them
(McGuire and Olson, 1996), who do not fear losing their political power (Robinson, 1998, Acemoglu and
Robinson, 2000b), or who have to gain the support of a large "winning coalition" (Bueno de Mesquita et
al., 2003). See also Acemoglu and Robinson (2000b, 2004) for a framework to analyze transitions from
dictatorship to democracy.
Our contribution
to this literature
rule), to systematically investigate
and
to develop a
number
regimes and policies.
is
to provide a formal
how a
divide-and-rule strategy supports kleptocracies,
of comparative static results
These are
all
model of kleptocracy (personal
on the likelihood of kleptocratic
prerequisites to a full understanding of politics in
weakly-institutionaHzed societies.
The
rest of the
paper
is
organized as follows. In the next section, we briefly discuss
the experiences of two countries under kleptocratic regimes, with special emphasis on
main
these regimes were sustained. Section 3 presents a n:iodel that formahzes the
how
ideas
discussed above and derives the main comparative static results. Section 4 concludes.
The Kleptocratic Experience: Two Case Studies
2
In this section
Seko
in
we
briefly discuss
two
the Democratic Republic of the
Ti-ujillo in
the
Congo from 1965
Dominican Republic from 1930
characteristic features of these regimes.
their
The
classic kleptocracies.
to 1961.
First,
to 1997
Our
Mobutu
rule of
and the
Sese
rule of Rafael
discussion emphasizes several
both rulers were true kleptocrats in that
prime aim was to accumulate personal wealth
for
themselves and their families.
Second, neither regime, despite their longevity, had an extensive base of support in society.
Third, each ruler maintained control of power by manipulating people and groups along
the lines of divide-and-rule. Finally, the grip on power of both rulers was greatly facilitated
by the nature of
2.1
society, natural resource rents
Mobutu Sese Seko and
Joseph Mobutu seized power
in the
MPR
Congo, which became independent from Belgium in
and quickly established himself as a dictator
(Mouvement Populaire de
doubt that the aim of Mobutu was to use the state
family.
He was a
aid.
Le Mai Zairois
1960, in a military coup in 1965
of a one-party state, the
and foreign
for the
la
Revolution).
at the
There
head
is
no
enrichment of himself and his
true kleptocrat. In the 1970s, 15-20 percent of the operating budget of
the state went directly to Mobutu. In 1977 Mobutu's family took $71 million from the
National
Bank
for personal use
and by the early
at $5 billion (Leslie, 1987, p. 72). Turner
1
980s his personal fortune was estimated
and Young (1985, pp. 178-183) devote
six pages
to listing Mobutu's assets and wealth.
The
social base of the
Mobutu regime was very
narrow.
Mobutu
himself
a small ethnic group, the Ngbandi, in the Equateur province and there
is
some evidence
of systematic favoritism towards fellow Equateurians especially towards the
regime.
For example, in 1990
46%
of the officer corp,
34%
came from
of diplomats
end of
his
and 19% of
MPR
the
who
Central Committee were from Equateur.
constituted the base of support.
These people,
Nevertheless,
called
it
was
by Zairians
relatively few
les
gros legumes
or les barons du regime constituted "an inner core of persons with especially close kin-
known
ship, ethnoregional, or personal ties to the president; variously
as the 'presidential
brotherhood' or the 'untouchables,' they were able to conduct their mercantile affairs in
blatant disregard for normal legal regulations." (Turner and Young, 1985, p. 398)
Mobutu used
alist ideology,
various strategies to maintain power, including the miUtary and nation-
and most notably the philosophy
most important strategy was similar to what we
of
'Mobutuism' after 1974. However, his
refer to as
"
divide- and-rule," creating
environment in which any person or group could be rewarded or punished
example, Leslie (1987,
dependent on him
for selection
and maintenance
of power.
By
ernment posts, Mobutu manages to maintain uncertainty and
the role of big chief
...
For
"Individuals in public office are totally
70) describes this as:
p.
selectively.
an
frequently rotating gov-
vulnerability...
He
plays
bestowing favors on his subjects based on personal discretion."
Turner and Young, in their seminal book on the Mobutu
state, describe the
same
situa-
tion as follows: "Ghent office holders have been constantly reiTiinded of the precariousness
by the frequency of
of tenure
office rotation,
which simultaneously
Zairians anxiously waiting just outside the portals of power.
for
example, was revamped a dozen times in the
of course
many
as
first
membership
lists...
Mobutu
life.
No one
among top
During the
of sanctions.
(except
persons have figured on
Insecurity has been sustained
and party personnel by the frequent application
of the
decade of party
six
hopes of those
MPR Political Bureau,
The
Mobutu) has been continuously a member, and only
as half of the
fuels the
first
state
decade
regime, 29 of the 212 leaders went directly from their posts to prison on
either pohtical or corruption charges.
An
additional 26 were removed on grounds of dis-
loyalty or dishonesty, with penal sanctions... Frequently in presidential speeches reference
is
made
to
unnamed
disloyal or corrupt persons in the leadership ranks... Cumulatively,
these devices constitute a powerful
mechanism
of informal intimidation,
and suggest why
systematic opposition has never arisen within the top organs of the state." (1985, pp.
"As a co-optative resource, a pool of vacant
165-166). Later they continue (pp. 397-398):
high
offices
was sustained,
...
the sanction for not cooperating
corruption, nepotism, or subversion charges, or exile.
agencies depended
as:
upon
"the shifting patterns of
membership
essence of patrimonial politics."
Gould's analysis concurs
(p.
[of
was imprisonment on
Access to high rank in
They sum up
presidential favor."
...
all
state
the essence of Mobutu's state
the Political Bureau]
...
constituted the very
193).
when he
notes that "the frequent cabinet shuffles and trans-
8
from region to region
fers of officials
may
...
be explained as largely reflections of the
him and
president's skill at using people while they can provide assistance to
same
at the
time keeping factions separated from each other, thus preventing autonomous power centers
from developing." (1980,
The
career of
illustrates the
p. 83)
Nguza Karl-i-Bond, who
released
In 1977 he
Mobutu camp and
was accused of
in exile
the ambassador to the United States (Leslie, 1987,
Turner and Young (1985,
severe, a fall
make
book
and became
in favor,
There are many other cases
Belshilva (see Gould, 1980, p.
166) note in this context "though sanctions could be
Repentance and renewed cultivation of the favor of the sovereign
possible a return to full grace."
The impact
and the
p.
his
from grace was not necessarily permanent. Those jailed seldom remained
in prison for very long.
could
p. 70-71).
example Kamitatu-Massamba and Mbujimayi
In 1979 he was
and gained fame by
on the Mobutu regime (Karl-i-Bond, 1982). By 1985 he was back
83).
in the 1960s,
one point became the foreign
at
and sentenced to death.
ti'eason
and made prime minister. By 1981 he was
like this, for
Mobutu
workings of divide-and-rule and the nature of shifting personal alliances.
In the 1970s he was brought into the
minister.
started as an opponent of
of divide-and-rule
was catastrophic
state. Leslie (1987, p. 6) notes that in
bureaucracy
for the efficiency of the
Mobutu's regime: "what
is
considered to
be simply bureaucratic disorganization and economic mismanagement by external actors
such as the [World] Bank and the IMF,
is
pohcy
to Zaire's ruling elite a rational
of
'organized disorganization' designed to maintain the status quo."
Added
much
up
in
to this, the personal wealth accumulation of
of the economy.
The
nationalization of foreign
Mobutu and
owned
firms,
his clique destroyed
most of wliich ended
hands of Mobutu and the regime, under the Zairianization program
in the
November 1973 was
particularly disastrous, destroying
According to Maddison (1995)
GDP
what was
left
initiated
of the economy.
per-capita in 1992 was less than half of what
had
it
been at independence in 1960.
What
the
for
factors facihtated
Congo was
him
to
buy
Mobutu's use of divide-and-rule? During the
clearly very
off potential challengers.
zinc, gold, silver, oil
natural wealth gave
The Congo
is
an analysis of the
15%
Moreover, the Congo
is
it
easy
endowed with enormous
of the world's copper deposits, vast
and many other resources (LesHe, 1993,
Mobutu a constant
also a
Mobutu,
poor and characterized by low producti\'ity, making
natuT'al resource wealth including
diamonds,
rule of
p.
amounts of
3).
This vast
flow of income to help sustain his power.
huge and geographically diverse country (see Herbst, 2000,
"difficult
geography" of the Congo) and
many
for
scholars have seen
it
as having a uniquely
common
complex cultural heritage. The
about 200 distinct ethnic groups
in the
Congo
estimate
(LesHe, 1993, p. 68)
271) argued that "The particular colonial experience has worked to
in the
Congo
that there are
is
and Young (1965,
make
p.
ethnic loyalties
stronger and supra-tribal identification probably weaker than in most other
African states." According to numbers reported in Easterly and Levine (1997), the Congo
is
The fragmented
the second most ethnically diverse country in the world after Tanzania.
nature of the country, combined with the impact of the Belgian colonial state, meant that
there was no large group or socio-economic class that could
and help obstruct
contrast to
state
his personal rule.
As we noted
ofi'set
in the Introduction, this
Botswana and Mauritius where the existence of such
power and limited the potential
identification of different groups
for kleptocracy.
made
it
the power of
is
Mobutu
in stark
classes placed checks
on
Moreover, the regional and ethnic
straightforward for
Mobutu
to engage in the
types of targeted redistribution and punishment emphasized in our model.
Finally, there
enabling
Mobutu
is
general consensus
to stay in
the
Mobutu regime and
was
crucial in propping
power
among
scholars that foreign aid
for so long. Leslie's analysis of the relationship
foreign aid agencies suggests that aid
up the regime. She notes
agencies such as the [World]
was a key
Bank and
and foreign
factor
between
financial support
(1987, p. 135) "the regime
...
views aid
bilateral donors as sources of foreign exchange,
not partners in development," and also emphasizes that
Mobutu and
his clique "sees
development assistance as yet another means to accumulate personal wealth" (1987,
6).
Mobutu made
abysmal
situation]
efforts to
reform the regime and
its
economic
policies only
undermined the confidence of Western governments and
thus flows of foreign exchange, resources vital for
Thus the Mobutu regime
sustain a regime with
is
little, if
a
classic
Mobutu and
example
of
any, popular support.
how
when
p.
"[the
investors,
and
his clique" (1987, p. 142).^
divide-and-rule can be used to
One could ask why Mobutu,
faced
with ruhng such a pohty, could not simply sustain himself through repression, rather than
resort to the relatively
complex strategy we have outlined. One reason
is
that
always kept the military relatively weak because he could never be sure of
its
Mobutu
support.
Divide-and-rule was used with respect to military officers as well as civilian pohtical
When Mobutu
French,
who
needed
effective military forces
he relied instead on the Belgians and the
could not be expected to repress domestic opponents.
Congo during
this awful era also illustrates the
elites.
The experience
of the
importance of the factors highlighted by
the comparative statics of our model.
®The reforms were never properly implemented, however, because "[such] reforms strike at the heart
Mobutu's patrimonial system and will only serve to undermine patron-client networks, efforts to buy
loyalty, and wealth seeking strategies" Leslie (1993, p. 123).
of
10
The Dominican Republic under Rafael
2.2
The Dominican
Republic, the eastern half of the island of Hispaniola, became indepen-
However,
dent from Spain in 1821.
freedom only
Tru.iillo
in 1844 after a
in
1822
was invaded by
it
Haiti,
and regained
its
war of independence. The Haitian occupation destroyed the
traditional ruling elite of large landowners
who hed
the coimtry never to return. In the
subsequent decades, the country was plagued by serious political instability and a
series
of dictatorships. In 1916 the United States took over the country (previously, they
had
controlled the customs since 1906 to recover delicjucnt loans), an occupation that lasted
Dining
until 1924.
this period the country
was run by a mihtary governor. To establish
control of the coimtry, the United States trained a domestic constabulaiy, which
the Nation
1930 used
Army
it
as a
after 1924.
Trujillo
became the head
this constabulary in 1928
power base to help a coup against the government,
himself elected President in a fraudulent election.
became
Throughout
after
and
in
which he had
his rule Trujillo carried
out the facade of elections, regularly being re-elected by 100 percent of the vote and even
on two instances, from 1938 to 1942 and from 1952 to 1961, stepping down from the
Presidency and controUing power from behind the scenes.
As with Mobutu,
there
is
no doubt about the kleptocratic tendencies of TrujiUo.
In-
deed, relative to the size of the economy, Trujiho's regime was probably the most successful
case of kleptocratic accumulation in history.
Trujillo family
was equal to about 100 percent
"controlled ahnost
try's labor force
80%
Trujillo used
depended on
rule
many
the end of his regime, the fortune of the
of
GDP
at current prices
of the country's industrial production.
and another 15% working
tial rivals,
By
liim directly or indirectly,
for the state"
with
(Moya Pons, 1995,
tools to maintain power.
strategy.
As a
About 60%
45% employed
family,
of the counin his firms
p. 398).
He used coercion and
and murdered and tortured many opponents. But as
was a key
and the
in the
force against poten-
Congo, divide-and-
result, Hartljii explains that "Trujillo eventually
became
the single dominant force in the country by combining abuse of state power, threats and
co-optation. Although certain of the country's economic elite maintained
autonomy
How
some individual
there was no possibility for independent organization" (1998, p. 99-100).'°
the Dominican version of divide-and-rule worked
is
explained in Turits (2003,
p.
Those who benefitted from the regime were few outside Trujillo's extended family. A 1953 New
York Times article (quoted in Wiarda, 1968, p. 74) found that L53 relatives were employed by the
government. Trujillo himself was minister of foreign relations, minister of social security, ambassador at
large, and special ambassador to the United States. Family members held the presidency (his brother),
two senatorial posts, six major diplomatic assignments, the positions of commander in chief of the armed
^'^
forces, undersecretary of defense, chief of staff of the air force, inspector general of the army, inspector of
embassies, plus a great number of other posts.
11
"
"Trujillo continually shuffled cabinet
7):
of
members and other important
government posts to prevent their developing an independent power base. He exerted
Upon assumption
tighter control over the national legislature.
still
own
of the National Congress were obUged to sign their
on
and out
officials in
Trujillo's order
of office,
resignations, a
could be handed to them at any moment...
members
document that
In certain cases, these
resignations were reportedly delivered to legislators in the middle of their speeches before
the assembly.
In the 1942 to 1947 period, for example, there were 32 resignations in a
Senate of 19 members, and 139 resignations in a House composed of 42 deputies."
Wiarda's analysis
is
similax: "Trujillo's
was a highly personalistic dictatorship in which
power was not shared, even among a small
man"
(1968, p.
trolling the
26).
He
further demonstrates that "Trujillo's principal
He had
the constitutional authority
government personnel and he used
his
signed but updated resignations for
all
work only to learn that
The technique
jillo's
of shuffling
armed
"Trujillo's technique
had
forces
p. 62).
all
Wiarda
ruthlessness
Hartly (1998,
p.
100) adds:
and co-optation. There was
own
of the
hands."
...
same reason
as Tru-
Anyone who gained
many
(p.
and, corre-
62)
were often pardoned and
63 and Crassweller, 1966, for
p.
combined
"Politically, Trujillo
...
frequently ar-
continues:
jailed or exiled
turned to positions of power (Wiarda, 1968,
of
no potential opposition power centers
authority in his
who were demoted,
for the
and police commands
...
offlcials
file
all
the date and that they had 'resigned.'
filled in
was to fragment the power
spondingly, to concentrate
political
and remove almost
government employees, and
Trujillo
were aUowed to develop" (1968,
examples).
to appoint
of con-
power extensively. In addition, he kept a
a powerful position could expect to be replaced
in Zaire, people
...
government personnel was employed
frequent changing of the
As
method
governmental machinery was the constant shuffling and reshuffling of
offlceholders.
rived at
but concentrated in the hands of one
clique,
many
guile, cynicism,
incredible manipulation of indi\dduals,
found themselves moved and removed from public
office in
re-
who
complex and disconcerting
fashion even as personal rivalries were promoted and tested.
Trujillo also
(2003, pp.
employed the strategy of divide-and-rule to control the rural
82-83, 95) describes this as follows:
assistance, peasants
efficient
manner...
were compelled to farm
The dual nature
in
exchange
for land access
and state
a more sedentary, intensive and land
of the Trujillo state's proteccion, or support, of the
peasantry was evident in the severe methods
The
"in
sector. Turits
state distributed rewards to those
it
it
used to augment peasant production.
deemed 'man
12
of work'
by variously providing
them with
land, irrigation, tools, seeds, credit,
punished with vagrant penalties
those
it
as a gift
Most
considered to be
more than a
idle...
and technical
The
distribution of lands under Trujillo
notably, Turits (2003, p. 113) explains:^^
"the state's mediation of peasants' access
anyone suspected of
The
political disloyalty
...
who wished
from the neighborhood
certificate of 'good conduct'
was represented
dictator's personal power."
to land fostered a high level of political control. All individuals
had to supply a
harshly
it
and forced labor) and withdrawal of usufruct rights
(jail
and thus served to dramatize the
right
assistance, while
to obtain land
authority... Certainly
would be excluded."
divide-and-rule strategy was facilitated by the extraordinary control that Trujillo
According to Wiarda (1968, pp.
accumulated over the economy.
could hire and
fire
whom
he pleased when he pleased.
87, 90-91):
"Ti'ujillo
Since the great majority of the
population was absolutely dependent on him for day-to-day existence, his control over
was assured
political
...
there was so
economic independence that even a bare minimum of
little
independence was impossible... Trujillo frequently employed
to destroy his political opponents.
Banks could and did
...
and
streets
his
economic power
refuse loans
mortgages; government agencies refused export or import permits
services were interrupted;
it
...
and
foreclose
electricity or
and sidewalks infront of selected business
phone
establish-
ments were torn up."
As with divide-and-rule
of the state machinery.
in the
Congo, one
Turits (2003, p.
an almost inevitable by-product of
efTect
was to destroy the
140) notes:
efficiency of
much
"confusion within the state was
system of continually circulating high-level
Trujillo's
functionaries into different positions almost every j'ear for the purposes of maintaining
his control." Nevertheless,
economic performance under
Mobutu. Though there appears to be no
What
all
data
for this period,
the
Dominican Repubhc experienced some gi-owth
rule.
facilitated divide-and-rule in the
the factors suggested by our model
fit
Dominican Repubhc? Unlike the Congo, not
the Dominican experience. For example, though
the country has very rich agricultural lands,
oil,
was not as bad as under
reliable national account
scholarly consensus appears to be that the
during Trujillo's
Trujillo
or other natural resources.
it
does not have large quantities of minerals,
Also the Trujillo regime took place in a period before
the existence of the IMF, World
Bank and much
international development assistance,
so this cannot have played an important role in sustaining the regime.
United States was generally friendly to the regime,
it
Although the
did not provide great quantities
"Turits (2003) interestingly claims that despite the manipulative nature of the relationship between
and the peasantry, the fact that Trujillo did actually give peasants access to land created quite
Ti'ujillo
a large degree of support for his regime.
13
of financial assistance.
power
However, two factors clearly are important.
Dominican Republic was a very poor country and
in 1930 the
feasible for Trujillo to
buy
Trujillo took
made
this
Moreover, although the country
off support.
heterogeneous in the way the Congo
When
is
it
relatively
not ethnically
the impact of the 19th century Haitian occupation
is,
and the United States military intervention, meant that there was no organized traditional
elite to
counterbalance Trujillo's
rise to
A Model
3
We now
how,
model
outline a simple
power and kleptocratic tendencies.
to formalize the ideas discussed above,
a kleptocratic regime sustained by a divide-
in weakly-institutionalized societies,
and-ruled strategy can emerge.
and to iUustrate
Our purpose
not generality, but to construct a simple
is
framework, which can be used to derive new comparative static results on the incidence
and sustainability of kleptocratic regimes.
3.1
We
The Environment
economy
consider a small open
(alternatively,
an economy with linear technology)
producing three goods: a natural resource, Z, and two goods,
the prices of
all
goods to
1,
which
is
without
loss of
differences in the technology of production of the
any
=
two goods. To
is
normalize
we
will allow
start with,
we focus on
constant in
all
periods,
affect political equihbria.
assume that the natural resource rents accrue to the government, and can then be
distributed to the producers or
There are two
We
We
Z.
Natural resources create rents in this economy, which, in turn,
We
q2-
generality, since
the case where the production of the natural resource good Zt
Zt
and
qi
normalize Ui
groups of agents,
(large)
—
n2
=
consumed by the
I-
E
Both groups have
P'u^s
[y^s,
s=t
where
/3
<
1
is
supply at time
t.
rii
ks)
ruler.
that produce qi and n2 that produce
utility at
= J] /?M y,« ^
time
t
given by:
^/,;
'
s=t
(1)
)
-^
the discount factor, yu denotes their after-tax income, and
This specification implies that labor
For each producer of group
i,
is
qit=^zht,
14
lit
supplied wdth elasticity
the production technology
§2-
is
7]
labor
>
0.
is:
(2)
where u^
the productivity of group
is
that group
is
1
more productive,
=
i
>
i.e., tui
Without
1,2.
uj2-
loss of generahty,
To parametrize
we assume
the degree of inequality
between the two groups, we denote:
uji
where, by construction,
uJ is
=
+
u7(l
and
x)
uj2
=
—
oj[1
x),
(.3)
the average productivity of the economy, and
A
.r
6
is
potentially
[0, 1].
greater x corresponds to greater inequality between the two groups.
The only
redistributive tools in the
specific to each group,
specific ta:x;es
y^t
where
r^t
S
[0, 1]
(non- negative)
is
transfers.
for the results,
The
option to use group-
and plausible
in the context
where there are clear geogi'aphic and ethnic distinctions between
The post-tax income
groups. ^^
are a hnear income tax that
and group-specific lump-sum
and transfers are important
of African societies,
producer
economy
=
of the two groups are
{l-T^t)udit+Tit,
the income tax imposed on group
lump-sum
transfer to group
i
(4)
at time
t
and
T^t
G
[0,oo)
is
a
i.
In each period, each producer maximizes his utihty function (1) taking the tax rate
Tjt
as given. Tliis implies a labor supply function:
/,t(T.O
=
[(l-r,,)w,r.
(5)
This equation relates labor supply, and therefore output, to taxes, and illustrates the
distortionary effects of taxation: greater taxes reduce labor supply and output.
Using
(5),
the instantaneous indirect utility of a representative agent in group
is
i
found to be:
t/.(^.t,T,0
= T^k(l-r,,)]'+'' +
l
and tax revenues
R{Tu,
T2t)
+
r,,,
(6)
T]
are:
= Tuqu + T2tq2t = Tn
The government budget
constraint
Tu +
T2t
(1
- TuT
^l'^""
(1
- Toj"
^2^+''.
(7)
is
+
Cki < R{tu,
r2t)
+Z+
^"In practice, differential taxes also include indirect taxes levied
as well as the use of marketing boards to
+ T2t
manipulate agricultural
African countries (see Bates, 1981).
15
Ft,
(8)
on the producer and consumer groups,
prices, a frequently-used policy in
many
where Cut ^
[OjOo)
revenue given by
the consumption of the (kleptocratic) ruler, i?(rif,T2f)
is
and
(7),
G
Ft
the utility function at time
oo)
[0,
is
foreign aid,
if
any.
The
ruler
is
tax
is
assumed to have
t:
oo
s=t
where
<
/3j^
1
is
the discount factor of the ruler, which could differ from those of the
citizens.
The
racy,
political
Our
D.
"kleptocratic"
system
focus
(i.e.,
is
to
either "dictatorship" (controlled
is
by the
ruler),
K,
whether dictatorship can survive and to what extent
what extent the
ruler will
be able to tax producers
democ-
or
be
will
it
own
for his
consumption, while ensuring the survival of the dictatorship). To focus on this question,
we model democracy
in the simplest possible way,
two producer groups are
in
power
and assume that
we assume that only the
More
foreign aid in democracy).
Ft
=
if
St-i
= D and
= F
Ft
=
O).-'"'
ruler receives foreign aid
formally, denoting the political state
if
— K.
St-i
democracy the
thus they set zero taxes, and share the natural
jointly,
resource rents and foreign aid equally (and therefore, set Cxt
In what follows,
in
appears, since any part of foreign aid that
is
This assumption
is
(i.e.,
by
there
no
is
we have
St-i,
not as extreme as
it
perpetual can be included in the natural
resource rents, Z, and therefore, Ft can be interpreted as the additional portion of foreign
aid that the ruler receives because under his rule there
because the ruler pursues a foreign policy
democracy
Finally,
will receive
we can
more
interpret
aid will continue after
F
in line
is
more poverty
with the donors'
foreign aid than the ruler,
we can
interests.
or famine, or
Moreover,
allow this by letting
as the fungible part of foreign aid. In this case, even
democracy
it
will
be allocated to some
if
specific purposes
if
F<
a
0.
foreign
such as
poverty reduction or education for disadvantaged groups. In this case in democracy the
two producer groups
one here with Ft
Given
=
will
not have access to aid, and this situation
when
S't-i
=
is
equivalent to the
D.
this assumption, the instantaneous utilities of the
two groups
in
democracy
are:
In contrast, in kleptocracy, the ruler will maximize his consumption, subject to the
constraint that he keeps power (alternatively, he can be removed from power, and in this
^^This formulation of policies in democracy
adopted to simplify the discussion, and the results are
is more attractive
for both producer groups than the kleptocratic regime. For example, we could allow the two groups to
be of different sizes, and the larger group to impose limited redistributive taxes on the smaller group.
is
robust to more realistic models of decision-making in democracy, as long as democracy
16
case,
democracy
=
and Cki
will result,
constraints facing the ruler in detail,
given by maximizing R{Tit,
=
rl,
and paying
transfers,
=
i.e., T*^
let
which
T2t),
is
r;,
us write the "unconstrained" solution.
This
is
achieved at the tax rates:
=
^
T*
^,
thus setting Cki
0,
Before describing the
for all future periods).
=
(10)
+ Z + F. The
R{t* t*)
,
instanta-
neous utihties of the two groups under these tax rates are given by:
=
f/.
[,.(...
=
...r..
=
fi;i(^)"\
=
o)
The Political Game and Definition of Equilibrium
3.2
The timing of events
a pohtical
in the political
state, either St-\
= D
game
or St-i
economy ever becomes a democracy,
it
are as follows. In each period,
=
K. St~\
= D
remains so forever.
If
St-i
they set the taxes and share the natural resources rents equally.
i.e.,
St-i
= K,
then the following game, Ft (K),
1.
The
2.
Each group
=
ruler
announces tax rates
denotes that group
1
i
=
1
and the other one has
replace the ruler,
3.
If Pt
at St
= K.
4.
If Ft
=
=
0,
we denote
1, i.e., if
budget constraint,
ruler's
5.
Pjt
= K,
=
1
(8).
Group
for
is
i
is
is
^
vector.
with P(
=
a dictatorship,
ruler
from power.
denotes otherwise
them
=
is
TTJ
(if
both
chosen randomly to have
otherwise.
this policy vector satisfies the
the "proposed", gi'oup
(Tif,r2j,T[(,
where
implemented and the pohtical system remains
a switch to democracy,
and
If society is
is
i.e.,
St
=
D.
i,
If
chooses d^
da
=
0,
=
1,
offer of
government
j then responds to the proposal of "proposer"
If
the
one of the two groups makes a proposal to
If
1,
for convenience,
one of the groups, then: the ruler makes a new
Given the poUcy vector, either
Output
=
if
D, the two producer
remove the
to
p^t
proposal, one of
T^J such that
both groups choose labor supply.
6.
a proposal and
0).
by Pj
this
{t\^, T2t, T[^,
new poUcy
replaced and there
remains at St
=
then (Tit,T2t, T^, T2t)
taxes and transfers,
and the
p^t
=
state, so
played:
make a proposal
make a
gi'oups simultaneously choose to
Pit
made
has
society inherits
,
and transfers {Tu,T2t)-
(Ti,,r2i)
decides whether to
i
is
f
an absorbing
is
groups play the simple game described above, denoted by Ft (D)
p,i
(11)
group
the ruler
j
is
the political system
implemented.
(rit,r2t,Tit, 72t) or {T{^,T2t,T[^,T2t), individuals in
•
'
produced, tax revenues are collected and consumption takes place.
17
If
7.
game
=
Pj
switches to Tt+i {D), and otherwise
it is
There are a number of noteworthy features:
a producer group act in cohesion
here, since there are
same
=
and the proposed group plays du
1
no costs of
preferences, so there
no
is
then
I,
next period the stage
in the
Ft+i (K).
first,
game.
in the political
political action,
we assume
and
that
This
is
individuals within
all
a natural assumption
agents within a group have the
all
free-rider problem. Second, there
is
a specific (political)
structure built in the timing of the political game: the ruler can only be replaced
two groups agree to replace him.
political
game, where the party
compete against potential
or
office
in
may have
considerable power, and
by one of the social groups alone. ^"^
An
power needs to receive support from aU
rivals,
would correspond to
alternative
social groups
"political institutions" placing
checks on politicians. However, such strong political institutions are absent in a
number
of less developed countries. In these weakly-institutionalized polities, the implied
of the ruler,
can
offer
strategy,
combined with the
which
(MPE)
An MPE
fact that after the proposal to
remove him from power he
be the focus of our analysis.
will
is
we
analysis,
of the above
will focus
game (though
on the (pure strategy) Markov Perfect Equilib-
this restriction
a mapping from the cmTent state of the
previously in the
same stage game)
which denotes whether the
game (and from
democracy or
either
is
not important for the results).
the actions taken
Here, the only state variable
to strategies.
political state
is
is
St-i,
dictatorship.
Analysis
3.3
The
MPE
wiU be characterized by backward induction. Wlien St-i
interesting actions,
and the
ruler receives zero utility, while the
utilities of:
with
t//^
receive
On
^''In
given by
D, there are no
two groups receive hfetime
(9).
Note
also that
VP
^,
given by (12)
(12)
is
what the proposed group
chooses da
—
1
the other hand,
if,
in response to the reaction of the ruler
if it
=
ttD
yF =
on
power
a different policy vector, gives him the opportunity to use a "divide-and-rule"
To simphfy the
ria
the
This assumption captures the fact that in weakly-
institutionalized societies, those controlling the state
cannot be easily removed from
if
addition, in
distinct groups
some countries the ethnic and
may be
i,
we
ruler
from power.
(r[j,
Tl^}'^ the
social divisions enabling different taxes
proposed
and transfers
absent, thus also undermining the logic of the divide-and-rule strategy.
^^The reaction of the ruler
relevant for group
and removes the
will
will
is
the vector (tij, r^j, Tf^, TjJ, but since only the component {t\^^TI^
use this lower-dimensional vector to simplify notation.
18
is
=
group chooses da
members
0, its
V,^ {Tl,Tl,
where
[/j is
this group.
will receive
rirn =
I
given by (6) and {t^,T[)
The reasoning
MPE
the
is
(t^ T'^)
^^f^,
(13)
tax transfer combination that applies to
for this expression is that in this period, the
and the kleptocrat remains
receives (r[j,r,^),
BU
+
U. {T\,,Tl^
proposed group
in power, so in the future, the play goes
back to the equihbrium pohcy of (r^jT/).^^
In addition, the response of the ruler
T[,
The
+
must
government budget constraint:
satisfy the
<i?K„r^J + Z +
rj,
F.
(14)
and the
"divicle-and-rule" strategy wiU be successful
ruler will
keep power only
\firl„T:,\rlTn>Vf.
It is
T*
useful to distinguish
1.
The
=
1/ (1
set of
ruler will
+
2.
As we
=
The
will see
(10)
tliis
and
T^^
=
for
happens by S*,
i
=
i.e.,
if
and
2
1,
=
a'
for
(77, /3,
rfj
=
We
denote the
Z, F,u}, x)
G S*, then
aU
t.
be able to maintain power
if
he set (T*,r*, 0,0), thus
cr'
^ S*.
below, in this case, the ruler can reduce taxes and increase transfers so as
will necessarily
We
will also see that in this case (tj, r|)
be forced to reduce taxes, and policy
To characterize S*,
replace the ruler,
let
for
i
^j
such that
To analyze how, and when, the
V^
subject to (14).
If
[rl 77]
=
Vf [r|, Tf] < VP
ruler
^
for
can do
max
i
=
or
is,
the ruler
less distortionaxy.
(r[j, 7;^
rf Tf)
,
|
(tjj, tI^, T^^, Tj^J.
< Vf then he
,
(15).
so, let us first define
lA^ (r^, 7^
1
{t*,t*), that
by the ruler by
Vf
be replaced. This shows that the ruler must ensure
<
which group j has proposed to
in
policies initially chosen
(r^^, T[i)
be
will
subgame
us start with the
and denote the
the ruler responds with
will
cases:
(r*,r*, 0,0).
ruler will not
to maintain power.
If
by
parameters such that
(r?„r^„rf„T|,)
(15)
be able to maintain power, with the equilibrium strategy of
as given
ry)
two
if
2,
|
r^ T^)
then group ]
^
(16)
i,
anticipating that
proposal win be accepted, will propose to replace the ruler, and the ruler
-will
be deposed.
"'Since the deviation does not affect any payoff-relevant variable in the continuation game, the
involves return to the
same
strategy,
and since we are focusing on pure strategy
19
its
equilibria, this is
MPE
simply
Therefore, the ruler must guarantee that
Vf
[t\,T^]
> Vf
for
=
i
1
and
2 to
remain
in
power.
Consequently,
we
first
V^
need to find
[r|, T/],
the
maximum
To do
give to the proposed group off the equilibrium path.
utility that the ruler
this,
we need
can
to maximize
(16) subject (14). Straightforward differentiation establishes that
<=
and
-^.
=
r;
^
Therefore, in fighting off a challenge from group
1
j,
+
r;
the ruler will set the revenue-maximizing
tax rate on this group, and set zero taxes on the proposed group
will clearly give the
minimum
possible
the government budget constraint,
amount
of the proposed group, as a function of the
proposal by group
Given
the
is
j,
maximum
In addition, the ruler
to the proposer group, thus
TJ
=
0.
Then
(8), implies:
Using these expressions, we can derive the
This expression
i.
maximum
MPE policy
utility that the ruler
off-thc-equilibrium-path payoff
vector
This
(rfj, r^j, Tfj, T2i)-
can give to group
i,
following a
as a function of the equilibrium tax and transfer rates on group
this analysis, the
problem of finding the
MPE is equivalent
is
i.
to finding a solution
to the following maximization probleixi of the ruler:
subject the constraint set:
V^\r\,Tt\>Vl'ioTi=\,2.
We now
(19)
characterize the solution to this constrained maximization problem.
notice that combining (11), (12)
and
(17), the constraint set, (19),
can be rewritten
-
1
+
7]
\
+ 'r]\\^r])
1-/3 V1 +
as:
—+—
)'\-^\\ +
?/
First,
7/
2
(20)
for
i
=
1,
2
and
j
^
i.
Then
can write the constraint
exploiting the fact that
coi
=
uj{l
+ x)
and
uj2
=
cJ(l
— .t), we
set as
*(ri, Ti,
.x)
>z(/3-^)-{1-P)F,
20
and
(21)
^{t2,T2,-x)>Z{I3-1] -{l~(3)F,
(22)
2
where
(1-/9)0;^+'' (1-2;)
i+v
y
•'/
p^^^H'^+^y^\,
_.l+rJ^
,,
(23)
Moreover, in the case where there
X
=
0,
the constraint set
is
It is
=
1'(r,;,r„a;
why
already possible to see
The
rium.
> Z (^ - ^) -
0)
[l
-
P) F.
(24)
the divide-and-rule strategy can arise in equilib-
constraint set, characterized by (21) and (22), will be satisfied when, ofF-the-
equilibrium path, the ruler can shift enough resources to the proposed group.
proposes to replace the
who
ruler,
ruler, the ruler will bribe the other gi'oup successfully
Recognizing this ofF-the-equilibrium path threat, no group
power.
In other
when each group knows
words, a very inefficient set of policies can be supported
in
when
i.e.,
simply:
=
<^{t„T^
no inequahty between the two groups,
is
that
if it
and remain
will challenge the
then be able to pursue kleptocratic poUcies along the equilibrium path.
will
Equilibrium Without Inequality
3.4
=
Let us start with the case in which x
and there
is
no inequality between the two
groups. First, note that whenever he can, the ruler would like to set the tax rates that
maximize
i.e.,
(18),
(Tjt, T^cT'ujT'lt)
=
(r*,T*, 0,0).
Therefore, the
first
step
is
to char-
acterize the set of parameters E* such that these best tax rates (from the point of view
of the kleptocratic ruler) can be supported as equilibria.
{T'lf,T2t,Tf^,T2^)
ters
such that
=
'I'
(
we immediately obtain
(r*,r*, 0,0),
Y^,
j
> Z
(/3
—
i)
—
—
(1
/?)
Using (24) and substituting
the set E* as the set of parame-
F, or more explicitly:
(25)
lia
=
{ri,l3,Z,F,iu)
happens
^(r^,
will
T"")
if
cr
=Z
e
E*, then the
^ S*. Then,
(/3
—
i)
—
(1
—
MPE
(rft, r^j,
F
/3)
involves (r^t,T|j,Tf(,T|,)
=
rfj,r|j)
(given the
=
(r*,r*, 0,0).
What
[t^t^.T^^T^) wih be chosen such that
symmetry between the two groups, the
ruler
choose the same taxes and transfers for both groups). Moreover, inspection of (23)
establishes that as long as
'I'(r,
ruler will reduce taxes to f
and
T=
sets
0)
= Z
(/3
-
lump-sum
21
-
-
/?)
F
for
transfers (this
is
intuitive, since taxes are
i)
(1
some f G
[0,
r*],
the
The important
distortionary).
by allowing the
make
of political institutions
is
that the ruler can always satisfy (25), and
This highlights the importance of the underlying
therefore remain in power. ^^
institutions in this context:
point to note
political
ruler to use divide-and-rule, the current set
sure that he always remains in power.
Nevertheless, the
extent to which he can transfer rents to himself and distort the allocation of resources
will
depend on parameter values as we
will see below.
This discussion establishes the following proposition:
Proposition
1.
When a
G S*, then the unique
(rft, Tlt,Tft,
2.
Wlien
=
riJ
Then we
MPE
is
{t*,t*, 0, 0) for
^ E*, then the unique
0"
(25).
be given by
1 Let E*
equilibrium policies are
MPE
some f G
Z
(/3
-
I)
-
[0,r*],
-
(1
/?)
F
and
O)
I
^
J
given by (13) and
The
is
remove the
also
i
=
1, 2.
a constrained kleptocratic regime where the
~
('^I'^^OiO) if
= z('/3-0-(l-/3)F,
=
is
Vf
j, i.e., pjt
{t^,TI,
have t%
from power, the ruler
=
t*
.
It
\
=
1, is
rf^Tl)
1
met by
= V^
{T'jf_,T^f)
where
may
will
is
if
respond with
therefore appear that
of Proposition 1
The only
Recall that
there are challenges along the equilibrium path
described in part
=
0,f)
=
V^
=
{t\,,TI,
(t*,0)
\
and
rlTf)
is
given by (12).
the uniqueness of equilibrium.
ruler
(26)
{0,0,f,f} where *(r
discussion above establishes this proposition.
comment
we
is
and
otherwise.
such that
Vf
t
{Tl^,Tl^,T^^,T^^)
In both cases, a challenge from group
{tI„T[^ for
an unconstrained kleptocratic regime where
(t'u, T2i,7'uj7"2^t)
*(f,r =
for
ah
have:
when
not unique. This
may need more
group j makes a proposal to
r^j
we can
o"
part that
=
r*,
and when a G E*,
construct equilibria where
G E*, and thus the equilibrium
is
not the case, however.
Any
^^To see this, suppose that for some parameter values, the kleptocrat is removed from power. Then it
must be the case that Vf \t^,TI] < VP for i = 1 or 2. But in this case, it is credible for the kleptocrat to
reduce t\ and increase Tf and by construction, V^" \t\ = 0,rf = Z/2] > V^ providing a contradiction
and establishing that the kleptocrat will not be removed from power along the equilibrium path. This
argument would not necessarily work when F < 0, however.
In a more general set up with stochastic productivity or costs of removing the ruler, there would also
be parameter values for which dictatorship cannot survive, similar to, and for the same reasons as, the
models of transition to democracy in Acemoglu and Robinson (2000b, 2004).
,
,
22
combination of strategies where
to [t%,Tj^)
Pjt
=
iT*,s) for e
>
ruler, the only
one described
=
cannot be an equilibrium.
1
would be a best response
would then cost group
1
by the
=
pjt
j
an amount
e
>
0.
for
were, a deviation
If it
the ruler, and the strategy of
Since a smaller e
combination of best response strategies
is
when
always preferred
is
^
e
0,
which
the
in the proposition.
This proposition therefore formalizes how the ruler remains in power and
He
transfer resources to himself thanks to the divide-and-rule strategy.
follows:
is
when threatened by
is
able to
achieves this as
the "proposer" group, he can always gain the allegiance of
the other, "proposed" group, by shifting resources to them. Because the proposed group
pivotal, the ruler can
tlais is
remain
in
power
if
he can successfully buy off the proposed group.
is
If
the case, anticipating this outcome, neither group will attempt to remove the ruler
from power, and he
will
be able to establish a kleptocratic regime transferring resources
to himself at the expense of the productive groups in society.
The
proposition also highhghts the notion of "constrained kleptocratic regime," where
the ruler
is
able to pursue kleptocratic policies transferring resources to himself, but in
endeavor he
this
constrained by the threat that the two groups will coordinate and
is
remove him from power. To avoid
(or
sometimes
sets
tax:es
Notice that when
which was discussed
cr
this possibility, the ruler reduces the equilibrium taxes
and mak;es
positive transfers) to the
two groups.
G S*, the equihbrium does not feature the notion
in the introduction.
of "punishment",
According to this notion, kleptocrats are
power
in
because they can threaten to punish challengers and reward loyal groups. In this case,
both on and
the rate r* and receives
transfers.
In contrast,
challenges the ruler, not only will gi'oup
also
group j
group that challenges the ruler
ofT the equilibrium path, the
will
be punished
v.dth
i
^
of those are
Greater
1.
F
makes
budget constraint of the
cr
ruler
<
r*,
and
group j
ruler,
but
intuitive:
greater F,
cr
i.e.,
(26):
^ E*. greater
F
increases taxes.
greater foreign aid, relaxes the
and provides him with more resources to buy
F
off the pivotal
makes the kleptocratic regime
easier
This comparative static result suggests that the foreign aid given to
many
African regimes by the United States and the United Nations during the Cold
may
if
of natural comparative statics in this model.
G E* more hkely, and when
is
r^
be bribed to cooperate with the
off the equilibrium path. Therefore, greater
to sustain.
^ E*,
immediate from the inspection of (25) and
This comparative static
group
cr
taxed at
the tax increased to r*.
Next we turn to a discussion of a number
Most
j
when
is
War
period
have had the unforeseen consequence of consohdating kleptocratic regimes. As we
discussed in the introduction, this comparative static result
23
may help
us understand
why in
the postwar period, foreign aid appears to have had no positive effect on economic growth
on average, and
in fact,
it
may
have had a negative
effect
on the economic outcomes
in
certain non-democratic countries.
Greater
2.
Greater
ruler
/3
fi
makes a eT,*
means that both groups
—greater returns
in
likely that the ruler will
and when a ^
less hkely,
democracy
are
more
—accrues
T.*,
greater
/3
reduces taxes.
patient. Since the benefit of replacing the
in the future, greater patience
be able to maintain
his kleptocratic regime.
more
static suggests that kleptocratic regimes are
likely to
emerge
makes
less
it
This comparative
where
in societies
citizens or their poUtical representatives value the future less.
<
If /?
3.
taxes. If
/5
>
1/2,
then greater
Z makes a
G E* more
hkely,
and when a ^ S*,
it
increases
1/2, the opposite comparative statics apply.
Inspection of (25) shows that greater natural resource rents create two opposing forces.
First, like foreign aid, greater
Z relaxes the budget constraint of the ruler,
to sustain his kleptocratic regime
greater
Z
by buying
off pivotal
The reason why the
When /5 < 1/2, the two groups are sufficiently
dominates. When /3 > 1/2, the second effect dominates.
first effect
relevant threshold
is
1/2
is
that in democracy natural resource rents
will
be divided between the two groups, whereas
pay
all
<
may
off
the equilibrium path, the ruler can
the rents to the proposed group. Casual empiricism suggests that the case with
1/2 appears
more
and suggests that natural-resource-rich countries
relevant here,
be more prone to kleptocratic regimes. -^^ In
rents from natural resources
may
groups when challenged. Second,
increases the value of democracy.
short-sighted that the
(3
and enables him
help us explain
why
in resource-rich countries,
fact,
make kleptocracy more
kleptocratic
the comparative static that greater
likely
(i.e.,
the case vdth
/5
<
1/2)
and neopatrimonial regimes have often emerged
such as the Democratic Republic of the Congo (Zaire), Sierra
Leone, Liberia, and Nigeria.
4.
If /3
>
[r]/ {1
reduces taxes.
If /?
+ r]))^,
<
(ry/ (1
then greater
+ t]))^,
uj
makes a G S*
less likely,
and when a ^ E*,
the opposite comparative statics apply.
This result can be obtained by differentiating (25). The intuition for the result
be better understood by considering condition
of the productivity of the proposed
the proposed group,
likely to survive),
a;,,
it
(20),
which separately shows the
and the proposer groups.
makes the condition
less likely to
may
effect
Higher productivity of
hold (and kleptocracy more
because the proposed group has more to gain from democracy.
On
the
other hand, a greater level of productivity of the proposer group, Uj, makes kleptocracy
more hkely to
survive, because
'^The relevant time period here
is
it
implies greater tax revenues that the ruler can use
not a year, so
/3
24
< 1/2 may not be unreasonable.
proposed group.
to bril3e the
Consequently, higher average productivity
opposing forces. Wlien the discount factor
own
the
dominates and
effect
sufficiently large,
/3 is
[3
makes kleptocracy
gi'eater productivity
comparative static suggests that, as long as
i.e.,
u
>
creates two
(///(I
less likely.
+v)T^
This
-^^
not very low, societies that are otherwise
/5 is
productive are also more likely to suffer from kleptocratic regimes and distortionary
less
policies of rulers.
Equilibrium With Inequality
3.5
Let us
now
return to the case where the two groups do not necessarily have the
productivity,
x G
i.e.,
First, recall that despite the differences in productivity
(0,1].
between the two groups, the analysis
in subsection 3.3
preferred (unconstrained) policy for the ruler
question
when
is
To answer
and
let
E2 the
this policy vector will
parameters such that
make a
the ruler can
members
—
(U,
if
\E'(r*,
T=
=
{T'l^,Tl^,Tl^,T2^)
(r*,r*, 0,0).
The
for the ruler.
77, /3,
0,
above established that the most
Z, F, x)
the vector of parameters,
> Z (/? — i) — (1 —
> Z [3 — \) - (1 — /3) F.
T=
— x)
is
0,
x)
/3)
F and
In other
group 2 makes a proposal to remove the ruler from power,
counteroffer that co-opts gi'oup
1
even when along the
are taxed at the rate r* and receive no transfers.
group
set for
be possible
this question, recall that a'
words, a' G Si impUes that
I's
is still
Si the the set of parameters such that ^(r*,
set of
same
S2
is
MPE
group
the corresponding
2.
Formally, these two sets are defined by
=
Si
,
(1-/3)0;^+'' (1-.t)^+'?
n
+
l+T]
1
zu^+^ (1
/3
i
+
77
(27)
1+7,
xy+''
+v
1
+
>z{P--
(1-/?)F
r/
and
So
=
{i-i3)zj'^^{i
,
'
YVv
07^+" (1
/3
l
Since '^{t* ,T
when
=
+ xy^
0,
x)
is
+
-
1
+
(28)
77
l+T,
x)^+'' r
Tj
/
7]
\l
+
> Z
a decreasing function of
the producers in one of the groups
(1-/?)F
3
V
x,
we have that Si C
S2. In other words,
become more productive they
also
become more
can be verified that (77/ (1 + t]))^ is monotonically decreasing in 77, with (77/ (1 + 77))'' —^e^ :0.37
—» 00. Therefore, for the set of parameters where greater natural resources rents make kleptocratic
'^It
as
77
policies
;0
more
likely,
while simultaneously greater average productivity
€ (0.37,0.5).
25
make them
less likely,
we need
Consequently, the tighter constraint faced by the ruler
willing to oust the ruler.
satisfy the
that the
more productive group
off
more productive group has more
game above
ruler has to
to
the equilibrium path. This result reflects the fact
to gain
from democracy, where
its
members
The
not be taxed (or more generally, where they will be taxed more lightly).
political
is
will
logic of the
therefore implies that, everything else equal, the constraints that the
worry about
is
group 2 making an
and group
offer
the more productive
1,
group, accepting this proposal.
This observation and a similar analysis to the one
in the
previous subsection lead to
the following proposition:
Proposition 2 Let Ei and
1.
When
o'
When
a'
be given by
G Si, then the unique
{t%, T^2^,Tft,T^t)
2.
112
^ Ei but
=
a'
]\/[PE
(27)
and
(28).
Then we
have:
an unconstrained kleptocratic regime where
is
(r*, r*, 0, 0) for all t
and
i
=
1, 2.
MPE is
G E2, then the unique
regime where the equihbrium poUcy combination
a partially constrained kleptocratic
is (r^j,
t^^, T^i, T2t)
=
(ti, t* ,Ti, 0)
with
*(fi,ti,a;)
3.
When
a'
^
E2, then the unique
{'^lu^2t^T^uT^t)
=
MPE
(^1,-^2,^1,2^2)
*(fi,fi,x)
=
^z(p-^^-{l-P)F.
a fully constrained kleptocratic regime where
is
with
*(f2,f2, -x)
This proposition extends Proposition
the productivities of the two groups.
(29)
1
= z(l3-^]-{l-P)F.
(30)
to a situation with potential heterogeneity in
It also
introduces the notion of partially- and fully-
constrained kleptocratic regimes: as before, an unconstrained kleptocratic regime pursues
the policy most preferred by the ruler. Wlien
it is
partially constrained, the ruler has to
reduce the tax rate on the more productive group, but can tax the
as heavily as he wishes.
Wlien the regime
is
less
productive group
tax rates on both
fully constrained, the
groups are constrained.
The comparative
in this
static results discussed in the previous subsection continue to
extended model. The new result here
between the two producer groups.
groups
A
is
greater x
with respect to
the degree of inequality
—greater inequahty between the producer
—makes the unconstrained kleptocratic regime
that a' G Ei). Intuitively, the
x,
less
hkely
(i.e., it
more binding constraint from the point
26
apply
makes
it
less likely
of view of the ruler
"
is
to satisfy the
more productive
democracy becomes more attractive
becomes more
difficult for
room
less likely to
emerge.
buy them
when
to maneuver,
there
group becomes even more productive,
this
producers in this gi'oup, and therefore,
for the
the ruler to
static captures the notion that
ruler has less
when
gi'oup:
is
off
when
challenged. This comparative
a strong producer group
in society, the
and therefore the unconstrained kleptocratic regime
is
Loosely speaking, we can say that a highly productive producer
group creates a "balance of power", and
this balance
between one of the major producer
groups and the ruler prevents the most egregious kleptocratic
static result
it
policies.
might help us understand why kleptocracies are rare
This comparative
in African countries
with
powerful producer groups, such as the cattleowners in Botswana or the sugar planters in
Mauritius. -°
Concluding Remarks
4
economy research
Most current
political
where formal
political institutions place effective constraints
political strategies.
focuses on the study of politics in en\ironments
This approach does not seem sufficient to understand the nature of
politics in weakly-institutionalized societies
public, or the
both on politicians and on
such as Liberia, Haiti, the Dominican Re-
Democratic Republic of the Congo. In these
political institutions
formal
polities, differences in
cannot be the main determinant of differences in
policies.
For ex-
ample, even though the Congo had several constitutions during the rule of Mobutu,
appears that this had
little
impact on
his behavior.
For example, Gould (1980,
p.
it
63)
shows that the 1974 Zairian Constitution put Mobutu: "above the constitution, giving
him the
right
...
to
run
he wishes, to modify the conditions
for president as often as
governing elections, to change the number of seats on the Political Bureau [the governing
body
of Mobutu's pohtical pai'ty
ical
commissioners as he would
fit. "
Indeed,
Mpinga Kasenda, a
and
like,
effectively of the state], to dismiss as
many
and indeed to abrogate the constitution
poUt-
as he sees
professor of law and subsequently Zairian prime minister
noted in 1975 before an audience at the Institut Royal des Relacions Internacionales
Brussels that (quoted in Turner and Young, 1985,
to Montesquieu to determine
Such a situation
is
what form
of
p.
70):
in
"Zairians have no need to refer
government they need.
perhaps extreme. Zimbabwe currently has a much more conven-
^°Note, however, that a different timing of events might lead to the opposite comparative static with
For example, if after the two groups agree, the ruler can make a counteroffer to
one of the groups, then he will always choose the "weakest link", the group more likely to agree, which
will be the less productive group. In that Ccise, greater inequality may make kleptocracy more likely. We
respect to inequahty.
believe that the timing of
moves
in the political
game
27
in the text
is
more
plausible.
tional constitution, written during the Lancaster
House negotiations
of 1979 with the
British government before the end of the Rhodesian state. Nevertheless,
the
Mugabe regime
it is
clear that
acts in flagrant disregard for the formal structure of the constitution.
Motivated by these observations, and as a
step on the road to building a framework
first
for analysis of weakly-institutionalized polities,
we developed a model
and incomes of
rulers that expropriate the wealth
their citizens
how
of
remain
kleptocratic
power vidthout
in
a significant base of support in society. The success of kleptocrats rests on their abihty
to use a particular type of political strategy
—
divide- and-rule".
"
Members
of society
The
kleptocrat
(producer groups) need to cooperate in order to depose a kleptocrat.
may undermine
such cooperation, however, by using the tax revenues, the rents from
natural resources and the funds from foreign aid to bribe other groups to maintain his
position.
Thus kleptocrats can
remain
the equilibrium path. In equilibrium,
off
intensify the collective action
all
kleptocrat because of the threat of divide-and-rule.
good
stylistic description of the
behavior of
many
problem by threats which
are exploited
and no one challenges the
We argue that such a model provides
kleptocratic regimes, including
a
Mobutu
Sese Seko in the Democratic Republic of the Congo (Zaire) and Rafael Trujillo in the
Dominican Republic.
Our model not only
formalizes divide-and-rule, but yields a
static results useful in thinking
We
tocracy.
number
of comparative
about the emergence, longevity and implications of klep-
showed that foreign aid and rents from natural resources typically help
kleptocratic rulers by providing
tocratic policies are also
more
resentatives are shortsighted
them with
greater resources to
likely to arise
when producer groups
and when the average productivity
Also interestingly, greater inequality between producer groups
policies because
buy
more productive groups
are
more
difficult to
these comparative static results are useful in interpreting a
off
opponents. Klep-
or their political rep-
in the
may
buy
number
economy
is
low.
constrain kleptocratic
off.
We
discussed
how
of cases of kleptocratic
rule.
We
view this paper as only a
first
tentative step towards a systematic analysis of
equilibrium politics in weakly-institutionahzed polities.
omy
of less developed societies,
institutionalized societies.
political behavior that
lyze
how
We
we need a theory
To understand the
of behavior
political econ-
and change
in
weakly-
developed a simple model to study a specific aspect of
seems to be endemic
weakly-institutionalized societies
a society possess weak institutions
in
such
polities,
may develop
in the first place.
also exciting areas for future research.
28
but did not attempt to ana-
their institutions,
and what makes
These are not only important, but
References
5
Abreu, Dilip, David G. Peaice and Emiio Stacchetti (1990) "Toward a Theory
of Discounted Repeated
Games with Imperfect
Acemoglu, Daron (2003)
Monitoring," Econometrica, 58, 1041-1063.
Not a Pohtical Coase Theorem?" forthcoming
""Wliy
in
the Journal of Comparative Economics.
Acemoglu, Daron, Simon Johnson and James A. Robinson (2001)
Origins of Comparative Development:
An
Empirical Investigation,"
"Colonial
Amencan Economic
Review, 91, pp. 1369-1401.
Acemoglu, Daron, Simon Johnson and James A. Robinson (2003) "An
African Success Story: Botswana.
"
In In Search of Prosperity:
Analytic Narraiives on
Economic Growth edited by Dani Rodrik, Princeton: Princeton University
Acemoglu, Daron and James A. Robinson (2000a) "Wliy
Press.
did the West Extend
the Franchise? Democracy, Inequality and Growth in Historical Perspective," Quarterly
Journal of Economics, 115, pp. 1167-1199.
Acemoglu, Dairon, and James A. Robinson (2000b)
"Political Losers as Barriers
to Economic Development", American Economic Review, 90, pp. 126-130.
Acemoglu, Daron and James A. Robinson (2004) Economic
torship
Origins of Dicta-
and Democracy, book manuscript, forthcoming.
Bates, Robert H. (1981) Markets and States in Tropical Africa, Berkeley: University of California Press.
Bates, Robert H. (1989) Beyond the Miracle of the Market,
New
York: Cambridge
University Press.
Bates, Robert E. (2001) Prosperity and Violence: The
New York: W.W. Norton & Co.
Bowman, Larry W. (1991) MauriMus: Democracy and
Political
Economy
of De-
velopment,
Development
in the Indian
Ocean, Boulder: Westview Press.
Bratton, Michael and Nicolas van de Walle (1997) Democratic Experiments in
Africa:
Regime Transitions
in Comparative Perspective,
New York: Cambridge
University
Press.
Brautigcun,
Deborah (2000) Aid Dependence and
Govem.ance, Stckholm: Almqvist
and Wicksell.
Bueno de Mesquita, Bruce
D.,
James D. Morrow, Randolph M. Siverson
and Alastair Smith (2003) The Logic of Political Survival, Cambridge: MIT Press.
Burnside, Craig and David Dollar (2000) "Aid, Policies and Growth," American
Economic Review,
90, pp. 847-68.
29
Cox, Gary
W.
(1997) Making Votes Count,
Crassweller, Robert D. (1966)
tator,
New
The Life and Times of a Caribbean Dic-
Truyillo:
York: Macmillan.
Dollcir,
Doesn't,
New York: Cambridge University Press.
David and
and Why,
Easterly,
New
What Works, What
York: Oxford University Press.
William (2001) The
and Misadventures in
Easterly,
Lctnt Pritchett (1998) Assessing Aid;
Elusive Quest for Crowth: Economists' Adventures
the Tropics, Cambridge:
MIT
William and Ross Levine (1997)
Press.
"Africa's
and Ethnic Divisions," Quarterly Journal of Economics,
Ross Levine and David
Easterly, William,
Growth Tragedy:
112, pp.
Roodman
Policies
1203-1250.
(2003) "New Data,
New
Doubts: Revisiting 'Aid, Pohcies, and Growth"', Center for Global Development Working
Paper
#
http://www.cgdev.org/wp/cgd_wp026.pdf
26.
Gould, David
J.
(1980) Bureaucratic Corruption and Underdevelopment in the
Third World: The Case of Zaire,
Green, Edward
New
Pergamon
York:
and Robert H. Porter (1984) "Noncooperative
J.
under Imperfect Price Information," Econometrica,
Grossman, Herschel
I.
American Economic Review,
Grossman, Herschel
I.
J.
52, pp. 87-100.
81, pp. 912-921.
and
Noh
S.J.
Jonathan (1998) "The
Chehabi, H.E. and Juan
Collusion
(1991) "A General EquiHbrium Model of Insurrections,"
(1994) "Proprietary Public Finance and
Economic Welfare," Journal of Public Economics,
Hcirtlyn,
Press.
Linz eds.
Trujillo
53, pp. 187-204.
Regime
in the
Dominican Republic,"
Sultanistic Regimes, Baltimore:
in
Johns Hopkins
University Press.
Herbst, Jeffrey
I.
(2000) States and Poiuer in Africa:
Comparative Lessons in
Authority and Control, Princeton: Princeton University Press.
Jackson, Robert H. and Carl G. Rosberg (1982) Personal Rule
m Black Africa,
Berkeley: University of California Press.
Karl-i-Bond,
Rex
Nguza
(1982) Mobutu,
ou,
L'lncamation du Mai Zairois, London:
Collins.
Knack, Stephen (2000) "Aid Dependence and the Quality
of Governance,"
World
Bank, Policy Research Working Paper #2396.
La Ferrara, Eliana and Robert H. Bates (2001)
States,"
Economics and
Weak
Politics, 13, pp. 159-184.
Leith, J. Clark (2000)
Department
"PoUtical Competition in
Why Botswana
of Economics, University of
Prospered, Unpublished
Western Ontario.
30
Book Manuscript,
Winsome
Leslie,
J.
(1989) The World Bank and Structural Transformation in
Developing Countries: The Case of Zaire, Boulder: Lynne Rienner Publishers.
Leslie,
Winsome J.
State, Boulder:
(1993) Zaire: Continuity and
Westview
Political
Change in an Oppressive
Press.
Levi, McU'garet (1988) Of Rule and Revenue, Berkeley:
University of California
Press.
Juan
Linz,
J.
and Alfred Stepan (1996) Problems
of Dem.ocratic Transition
and
Consolidation: Southern Europe, South America, and post- communist Europe, Baltimore:
Johns Hopkins University Press.
Maddison, Angus (1995) Monitoring
World Economy 1820-1992,
the
Paris:
OECD
Development Centre.
McGuire, Martin C. and Mancur C. Olson (1996) "The Economics
of Autoc-
racy and Majority Rule," Jowmal of Economic Literature, 34. pp. 72-96.
Mehlum, Halvar, Karl Ove Moene and Ragnar Torvik (2001)
and the Resource Curse," Department of Economics, University
"Institutions
of Oslo,
http://econwpa.wustl.edu/eps/ge/papers/02 1 0/02 10004 pdf
.
Moya Pons, Frank
(1995) The Dominican Republic:
A
National History, Princeton:
Markus Weiner.
North, Douglass C. (1981) Stnicture and Cfiange
W.W. Norton &
in
Economic History,
New York:
Co.
North, Douglass C. and Bcirry R. Weingast (1989) "Constitutions and Commitment: The Evolution of Institutions Governing Public Choice
England" Jowm.al of Economic History,
Mancur
Olson,
in
Seventeenth-Century
49, pp. 803-832.
C. (1965) The Logic of
Collective Action: Public
Goods and
the
Theory of Groups, Cambridge: Harvard University Press.
Mancur
Olson,
C. (2000) Power and Prosperity,
Persson, Torsten and Guido Tabellini (2000)
Economic
Policy, Cambridge:
MIT
New
York: Basic Books.
Political
Economics: Explaining
Press.
Persson, Torsten and Guido Tabellini (2003) The Economic Effects of Constitutions:
Wiat Do
the
Data Say? Cambridge:
Robinson, James A. (1998) "Theories
3,
MIT
Press.
of 'Bad Policy'," Journal of Policy Refonn,
pp. 1-46.
Robinson, James A., R^ignar Torvik and Thierry Verdier (2002)
Foundations of the Resource Curse,"
CEPR Discussion
"Political
Paper #3422.
Ross, Michael (1999) "The Pohtical Economy of the Resource Curse," World Pol-
31
itics,
51, pp. 297-322.
Ross, Michael (2001) "Does Oil Hinder Democracy?" World
Politics, 53, pp. 325-
361.
Sachs, Jeffery D. and
Andrew Warner
and Economic Growth," Harvard Institute
(1995) "Natural Resource Abundance
for International
Development, Development
Discussion Paper no. 517.
Scindbrook, Richard (1985) The
Politics of Africa's
Economic Stagnation,
New
York: Cambridge University Press.
Kenneth A. and Barry R. Weingcist (1995) Positive
Institutions, Ann Arbor: University of Michigan Press.
Shepsle,
gressional
Turits,
Richard Lee (2003) Foundations
of Despotism:
Theories of Con-
Peasants,
The
Trujillo
Regime, and Modernity in Dominican History, Stanford: Stanford University Press.
Turner,
Thomas and
Cra^vford
Young
(1985) The Rise and Decline of the
Zairian State, Madison: University of Wisconsin Press.
Van de Walle, Nicolas (2001) African Economies and the Politics of Permanent
Crisis, 1979-1999, New York: Cambridge University Press.
Wiarda, Howard J (1968) Dictatorship and Development: The Method's of Control
in Trujillo's
Dominican Republic,
Gainesville: University of Florida Press.
Wintrobe, Ronald (1998) The
Political
Economy
of Dictatorship,
New York: Cam-
bridge University Press.
Young, Crawford (1965)
Politics in the Congo, Princeton:
Press.
3837 n06
32
Princeton University
Date Due
Lib-26-67
MIT LIBRARIES
3 9080 02617 7920
iiiiiiii
If