+1443 776-2705 panelessays@gmail.com


Same format as last time. Please make sure to use 2 of the readings.


– Williams

– Widner

– Bates

Country Specialization Paper 3 – Due December 7, 2021,


By the beginning of the 1990s, most African countries were in the midst of major
political transitions. In most cases, these transitions involved the liberalization of
previously closed political systems. For your country of expertise, please do the

1) Explain what triggered the transition
2) Describe how the transition unfolded and why
3) Describe and explain the eventual outcome of the transition (in other words, what is
the long term picture)

If your country of expertise did not experience a transition of some type, your paper
should focus on why this was the case. As in other papers, you are expected to
incorporate course material and outside texts. The paper should be 6-8 pages in length.








PRESS / Baltimore and London

7Q rni
,€ lbf

il liiliillniil ilXit ililitiri iliil u rm nfriffi lt m
s21o1 ozzgaz+lj


Preface and Acknowledgments vii

Map of Africa x

Introduction t (
lennifer A. Widner


One: The Impulse to Reform in Africa 13 ,/ zV;q
Robert H. Bates

Two: Through the structural Adjustment Minefield: Politics in an
Era of Economic Liberalization 29

Barbara Grosh


Three: Political Reform in Anglophone and Francophone
African Countries 49

lennifer A. Widner

Four: Structural Adjustment, Rent Seeking, and
Liberalization in Benin 80

Richnrd Westebbe

Five: Economic Crisis and Political Realignment in Zambia 101
Michnel Bratton

@ 1994 The Johns Hopkins University Press
All rights reserved
Printed in the United States of America on acid-free paper

The ]ohns Hopkins University Press
2715 North Charles Street, Baltimore, Maryland 21218-4319
The Johns Hopkins Press Ltd., London

Library of Congress Cataloging in Publication Data will
be found at the end of this book.

A catalog record for this book is available from
the British Library.

(‘{ ,r









[email protected] ,
The Impulse to Reform in Africa

Africa is in the midst of a sweeping movement of political reform.
Incumbent presidents have been voted out of office. others have
retired, mosi notably Julius Nyerere. Still others have been challenged

to broaden their ruling coalitions, oPen uP their political institutions,
and tolerate opposition and dissent. Authoritarians, such as Hastings
Banda, Daniei arap Moi, and Mobutu Sese Seko linger on, But even
they face strong .hull”.,gut. As analyzed by Claude Ake, Richard
Joseph, and otliers, the democratic impulse now runs strong in Af-
rica-a fact long ago discerned by Richard Sklar’1

Africa thus j;id what Samuel Huntington celebrates as the “third
wave,’ of democrattzatton: the political reforms that began in Iberia
in the mid-1970s, then swept through Latin America and Asia, cul-
minating in the late 1980s in the downfall of Leninist parties in Ea_stern
Europe lnd ttre Soviet Union. In analyzing this cuuent of political
reforin, Huntington, like others, notes that democratization tends to-
“break out” in middle-income countries, countries whose per capita
incomes range from $1,000 to $3,000.2 only a handful of corrntries in
black Africa fall in this range: Gabon, South Africa, and possibly
Mauritius and the Congo. In contrast to other parts of the world, the
political impetus to reiorm in Africa thus operates in the midst of

In this respect, the reformist movements in Africa are distinguished
by several paradoxes:


-while demanded domestically, they are often
supplied interna-


-IlVhile calling for new political
beginnings, they are often led by old


-ftre reforms they call for are often initiated by the
very governments

they are attacking.
This chapter seeks to resolve these paradoxes.

The Development Commitment

The “year of democratizafiott” in Africa-l99l-followed by three
decadesihe,,year of independence” -1960. Characteristic

of the drive

for independence was ttre demand for the use of politicalpower for
economilc purposes, a demand that ran parallel to deeply held political
convictions and ideological commitments. ,

The nationalist movement mobilized a wide atTay of. economic in-
terests. one segment was drawn from the white-collar workers who
staffed the bur6aucracies that managed the governments and firms

created by the European occupiers. Articulating their_demand-s in the
language”of political liberalism and racial equality, they -condemned
theluplrior ionditions of service accorded expatriates: higher salaries,
better housing, larger transport allowances, and so forth.3 Their de-
mands drew iupport from ill who aspired to salaried employment:
students in the universities and secondary school and their parents,
who were investing in the education of their children. These shared
economic interests gave impetus to broader political demands for racial

Business and commerce also backed the nationalist movement.

Because of the collusion of the large, expatriate firms in western Africa,

indigenous businessmen were able to attribute inflation in the postwar
peri6d to machinationsby the colonialists’ Condemningprice Eiuging
and monopoly profits by imperial firms, they mobilized popular sen-

timent for nationalist atiackJon foreign economic domination, on the

one hand, and nationalist appeals for public support for indigenous
entrepeneur#tn the other.a

An analogous,-pattern aPPears in eastern Africa. Michael Cowen
and Nicolab#iinson traci-the origins of nationalist politicians in
Kenya to indigenous competitors of colonial firms’ Many entered
poliiics to proiest the administrative barriers that pr.otected foreign
iirms and i^4creasedthe cost of entry for those who, like themselves,
sought weath through commerce. A. L. Epstein notes a similar Pattern

“,in /ambia, where businessmen-the so-called
hawkers-helped or-

The lmpulse to Reform in Africa 15

ganize nationalist political parties in the urban townships.s
The examples could be multiplied. Robert Bates, Lionel Cliffe,

Goran Hyden, and others stress the economic demands of the peas-
antrfinichard Sandbrook, Richard |effries, Epstein, and others itress
those of urban labor.6 The nationalist movement thus introduced a
core political expectation: that in Africa the road to prosperity ran
through politics, a point perhaps best made by Kwame Nkrumah
when he enjoined his contemporaries: “Seek ye first the political king-
dom, and all else shall be added unto you.”7

In the postindependence period, the nations of Africa took different
and COte

d’Ivoire coristructed a


so as to channel resources from activities to which they gave a higher
priority than did the market. The result was a characteristic pattern
of market intervention. This poliry mix provides a mechanism that
helps account for why, in the African case, poor economic perform-
ance and demand for political reform go together.

The Economic and Political Consequences: A Simple Model
This section sketches a simple, idealized,version of this mechanism

and derives from it implications for the economics and politics of
Africa. As seen in figure 1.1, given competition, normally behaved
forces of supply and demand generate a market equilibrium: they
establish a price (PJ at which the quantity demanded (QB) equals the
quantity supplied (Q$). Governments may seek to alter the market
solution, however. For example, for social or political purposes, they
may lower the price to a point below that generated by market forces
(say to P,).

In their attempts at state-led growth, African governments fre-
quently adopt policies that generate such distortions in markets. In
agriculture, for example, they have sought to lower the price of food

promoting the formation of an indigenous (if politically dependent)
bourgeoisie.l0 While differing in their developmental Eaiectories, Af-

nts [email protected] sought to overturn
ffi5fr4-To alter incentives and relative rates of return
nts were


16 ROBERT H. BATES The Impulse to Ret’orm in Africa 17

grams, their costs contribute to the macro level imbalances charac-
teristic of Africa’s economies. They also lead to characteristic political
patterns: clientelism and patronage politics, corruption, and the pri-
vatization of public institutions, resulting in a loss of legitimacy by
the activist governments that captured power at the time of inde-

The mechanics that link the microdistortions to macrolevel effects
can readily be outlined. The new price, P’ is not an equilibrium price.
At the lower price, less efficient producers cannot profitably supply
goods; and consumers will demand more goods than the remaining
producers can supply. The quantity demanded therefore exceeds the
quantity supplied, putting an upward pressure on the price. To render
the price sustainable, the government must therefore subsidize the
costs of production such that the less efficient producers find it prof-
itable to supply the good at the reduced price; the result is fiscal
pressure on the state. Alternatively, the government must import the
good; the result is loss of foreign exchange.

It should be stressed that the interuentionist policies pursued by
governments in Africa have been pursued elsewhere. As revealed in
Indonesia and Thailand, such policies can in fact be compatible with
economic growth, Unlike Indonesia, however, most African nations
have failed to maintain stable macroeconomic policies; they have pur-
sued populist economic programs, using their budgets to purchase
public support, to finance large military establishments, or to construct
elaborate monuments for the sake of hational pride. Unlike Thailand,
most African governments fail to pursue positions of advantage in
global trade; most often, they adopt trade or commercial policies that
separate their economies from international rirarkets. In the context
of such macroeconomic policies, the interventionist policies of African
governments inflict great damage on Africa’s economies, contributing
to Africa’s budgetary deficits and mounting international indebted-
ness. Insofar as the legitimary of any incumbent regime is tied to the
performance of its economy, they also weaken Africa’s governments.

The characteristic pattern of government intervention not only af-
fects Africa’s economic performance, it helps to define Africa’s char-
acteristic patterns of politics, as well. Interventionist policies can trans-
form economic markets into political machines. Interventions help to
create a form of politics that has been termed personalistic: one in
which those who hold power exchange personal benefits for political
loyalty.l3 The logic can be seen in figure 1.1: at the official price (Pr),
the quantity demanded exceeds the quantity supplied; the value of
the commodity then rises to P*-a price that high demanders of the

at” Q””= ab v;

Figure 1.1 A stYlized market

;;ilt” price of inputs for farmers’11 For industty’.-‘.1:I l”:”-i3t*n’
to lower the pricei charged for electricilV u”q PetrgleuT lroducts’
For consumeis, they have subsidized the prices of ess€ntral


*”aiti*, .ooLirrg oil, paraffin, processed foods’ and clothinS’ for
example.l2 In creJit rrruik”tt, they have sought to lower interest


;;;I’;”tl;;;;;;;”‘s” uxct’a”ge, thev-have sought to lower the
;;;;G;;i*h”reuy r:educing fhe costsof

importing from abroad’

Such interv”rrt pr,. are ofteri targeted at a specific market
or in-

a”ffi-1″ p*.geeltn”y are carried oulrr arylii!:]:ii:’il1:it” *
mobilize resources ,o* io futfill developm”nt obiectives. while de-
fined at the level of a specific market or industry’ the

impact of these

intewentions soon refrsters at the macro level’ This
is true by deti-

il;;i*”,rdt intplsed upon the exchangg rte ald.interest rates:
the impact of such measures is economy*idu’ ntrt

it is also true of
“kil#;; i”J”rir’,f*el distortions. Whet, summed over all pro-




good are willing to pay’ The excess demand
therefore creates new

value, but it also .t””iJJ tnu oppottu”ity for discretion’
With demand

exceeding supply, tt o'” *tlo’to**u”d po*”t over the
allocation of

the good grve it to- some and withhold it from others’

intewention ir, ir,a.”iJus and markets can thus
be used to create

organized followings-groop’ of ee,1s91s
who owe their individual

fortunes to discretron ufi o’ p”ttottutitfic acts bY those
in Power’

In Africa, the interv6niio’t’ ut’o help to cr”atu
a form of politics

that has been termed-pitrimoniatrone.in
which those who hold power

derive their incom”, f;;; lh;.o,”.tn In the sftuation
just described,

the political leader gi”es uwuy the scarce
commodity at subsidized

prices and reaps PJ;it”t t”*it” in exchange’ There exisT,-l

il;ibi*)” h;;”iut, ti’t”i”uaer can use his 6mmand
over the market

or industry to r”.ot”-ihe iood 1tfr9
government price’ P” and sell

it at the market .r”””i”j frice, P*’ thereby reaping a
higher income

from his Public office’1s
Intewentions hef create a third form of P”liT”:.ii O.tlt:u:-::” ‘nu’

has been termed ,i’t titt’i”g’ In this interpretation’ the
origins of

sovernmen. progru*’ ;;;t[i” t^e politi&t
or economic ambitions

of leaders, but in the actions of interest grouPs’
Political leaders in-

tervene in markets;;;;;;” io politicaip.esi,rrer by lobbyists

sponse, go.r”rr,*ur,’r”JfiiJ”rt *”i”t”in the Pt9gl3* !y

offlils suu-
sidies that make the ;;ut athactive to poiiticl[y t”Fil:l-“:ial


terests.l6 The result ir’th” t”” of government po*er to
underpin the

economic fortunes “l ”
prJr”gef, segment of civil society-private

groopt that possess influence.over Pullic Programs’
Thus far, I assume that leaders c-ontrot the Pt”gtiT:ll:l.govern

and either give tLe eco;omic benefits
of the ryoSt-u*

to individuals

or groups o, .or,roiil^ii-irtl*t”r”us’-Naturilly’
however’ the bu-

reaucrats who run;r;;”‘;r;t;;, are less than’perfect ag€nts
of the

leaders’will: they il;;;k; can-privatize the benefits of the

prosrams, selling ;;;;;”;”r thie counter at black market

[ffi;;;ftpi?r”Et” public agencies survive in part off the

of these sates. rnis,”f]’lrrh;;i1″”;” of politics
in Africa is bureaucratic

corruption. ^–F.A”
KX,Yr”;A{{“”suepoliciesthatareecon”*t:t}Ij :jiY:l1*l:’.'(Jvcr’ursr r'”


S”ritiJ anJ snortages of foreign – exchange’contributing to publi -:.,^-^^ ^G *^lirinc in nosr-
#;J;;;,qi’i*’:Af ica’sec!nSc-T::’:^H’.:T’:#ilffi::

The Impulse to Reform in Afriu 19
for-and demand for-reform originate in the midst of economic

The Delay of Reform: The Weakness of the Bourgeoisie

– Personalism, patrimonialism, rent seeking, and corruption have
long provoked condemnation within Africa. writers as diverse in style
and orientation as Ayi Kwei Armah, Chinua Achebe, and Ngugi wa
Thiong’o dramatize the relative wealth of those who havJ seized
power in postindependence Affica.l7 Ordinary citizens cheer the mil-
itary coups that toppled civilian governments and then turn against
the military.ds well. Their condemnation of their governments is fueled
by their growing recognition that governmeni policies undermine
their economic well-being, even while promoting the well-being of
those in politics.

Given the breadth, depth, and duration of public opposition to
prevailing political practices, why, then, has reform been so delayed?
There are several reasons, all of which share a common attributel the
lr”k of u”otro*i” ir.u{ttirrur fot irt”t”rtr,

“trd “”p””iullv

– interests, to opPose African governmenti.
Notwiffitanaing numero”s p”Utic staii-ents to the contrary, Af-

rica possesses an active and powerful private sector. The markets and
trading routes of Africa have long histories, strong institutions, and
highly professionalized cadres.l8 These commercial networks thrive
even in the midst of the interventionist policies of the postindepen-
dence period; they provide the framework for the black markets that
have burgeoned even as public institutions decayed.le A/hile a major
force in the real economies of Africa, this market economy has none-
theless failed to provide the focus for a coherent, reformist political

Participants in this informal economyhave failed to act as a coherent
bourgeoisie-failed to provide the political impetus necessary for sus-
tained economic reform. One reason is that, instead of being op-
pressed by the gap between official and market clearing prices, traders
in Africa profit from it. A portion of their profits results fuom arbitrage
between official and unofficial markets, especially in internationar
trade, where traders benefit from smuggling and profit from the ar-
bitrary valuation of national currencies.

Rather than turning to protest, then, “marketeers’, turn to trade
and seek to benefit from the very policies that destroy the business
environment for firms in the formal sector.



lli:t””*1trtr#i;?;*;;r crisis. rhe consequence
is that the need


is that ” in the
4ut Hirschman-s distinction does not quite serve/

The Impulse to Reform in Africa 21,

economic performance of African governments and the distributive
impact of their policies undermine their moral right to govern, there
is little in the way of a coherent middle class to oppose these policies.

impoltant to realize that efhnic rivalries can be manufactured by pol–mfiilbusettr
rder. If a reformer advocates

merit promotions instead of patronage, for example, incumbents point
out the potential for discrimination against those who are historically
deprived.z If reformers call for a reduction in subsidies, incumbents
from poorer regions condemn the proposal. And if reformers call for
a more competitive political regime, incumbents point to the threat
of ethnic violence, even while inciting some violence to lend credibility
to their contention.

Ethnic ties are deeply meaningful to many Africans. To gain social
standing in their communities,-elites often invest their material wealth
in local causes, building schools, clinics, or public works, for example.
They often marry locally, build a second home in the village, and
send their children to live there. Some seek traditional titles and
sacrifice major portions of their incomes from the modern sector for
traditional prestige. Appeals to ethnic loyalty by politicians who feel
threatened by reformist agendas thus can easily divide Africa’s eco-
nomic elite and undermine the potential for forming a coherent op-

The reformist impulse is thus weak in the African bourgeoisie.
Merchant capital depends upon, or even benefits from, government
interventions. Industrial capital is publicly owned. The managerial
bourgeoisie is silenced by public authorities. And the loyalties of the
economic elites are divided by political manipulation.

The Organization of Reform

. From whom, then. dnes fhp reformisf imnulse emanate? One source
is fixed and specific hrrmarlcafifal fhnse oeople who have invegjE:d

_in skills fhaf are hrr+ imFp#ectly transferable elsewhere Among these
are the old guard politicians and those who people the ranks of local,
community-based hierarchies: lawyers, community activists, and
church leaders.

A lawyer in Ghana, for example, especially one at the peak of his
career, is unlikely to find a comparable position in London. Although

._ildustries not voice tn
rica’s econo*i”, 4{@rygons? one r.easo: ll tPt.I*
i” f the caPital embodied in these
;;r#;;;;;pitalis often ownedby foreigners or the government. The
;;il po.u<it of the nontrading bourgeoisie is managerial’ not cap-
itaisf its members receive paynient inihe-form of salaries rather than

;;;;g; from shares.’zl Th; squeezing.of guuti rents from.these in-
vestmJnts and their dissipation by inefficient regulatorypolicies


offend the sensibilities of
‘these tallnted cadres, but they do-no] rlolate


p”i””tu economic interests. The losgeq imPosedsgnqtitute losses
! ir —c^.^^r k^^-,,n, saslrar +hqn fn fheir oersonal fortunes.to the [email protected] PsrDurrqr

As a manag”.ru, “ffi6″se

cadtei Je”[loin-fneif-1me-df’.ft

involves solving production problems under highly adverse

tions. It also inioives giving policy advice to directors of enterprises,

t”””y * *t om are brireauirats oi party personnel who vote but do
not own the shares of government’ ei ihose of us who have done
ilr”ur.r, in Africa trroi, much of the counsel the managers offer
constitutes a stinging inilictment of government,policies

anda plea

ior ttre freeing of iruit”ts’ But the ref6rmist impulse tl T:lI:f.lPt^”
cadres has beln stilled by the governance structure of Atnca’s


ir, *t ict ind”strial authority ofien rests in the hands of political elites’
Bv hirine and firing managerial personnel’ making

judicious use of

;’#?; “”Jtae”ii”g
u.&tt to credit and foreign exchange’ public

Ji*.iorr’ q”icttyitine ine protests of disaffected managers and

to-thgpr€taifi”g ry1y o{

doing business in Africa’

Activist governments in”Afriia intervene in markets and

industries,”contributing to the macroeconomic imbalances

weaken Africur,

and generate demands for reform’ but also

strengthening”the caficity of i”ncumbent regimes to neutralize

ie{ori’,ist opp”osition.’ihe;esult is deadlock, as those who
oppose the

pr5″”ifi”g pttu.ut-“.onomic order are stifled by it’ While the poor

Africa’s economic elite fails to become a coherent
t is easy to com-

prehend how ethnic loyalties provoke

commerce and trade

Urri iutt u, in physical capital? There 1e polts’ “llif::’

establishments, and sopIU.tt







he might join an international agency, his career path is not inter-
nation-al. Unlike finance capital or people with general, as opposed
to community-specific, skills, professionals cannot readily defect
abroad. Therefore, in the face of declining welfare, rather than exit,
they may find it preferable to give voice.

Such people have paid; they have used family savings to gain an
education. After others left school, they remained to train profes-
sionally. In the early stages of their careers they continued to train,
undertaking low-paying apprenticeships. Only later did they begin
to reap the rewards of their sacrifices. @

ve them of a
heads of families, moreover, a Pros-

ihe future prosperity and political stability of their country. These

The Impulse to Reform in Africa 23

government rather than by ascending through its ranks. It is char-
acteristic of African reformist movements that they tend to be led by
old guard politicians. For the reasons given here, it is also not sur-

Thus from a human capital framework, along with the theory of
exit and voice, one can derive the characteristics of those who people
the reform movement in Africa. I now add two refinements.

The decision to join the reformist opposition constitutes a gamble;
the uncertain value of the prospect must exceed the certain value of
the status quo. All else being equal, then, the less attractive the pre-
vailing conditions, the greater the willingness to gamble on reform.
Two implicadons follow. One pertains to the conditions under which
reform will,’take place: we would expect to see reformist political
gambles being taken in countries subject to economic stagnation and
poverty. Another pertains to the-composition of the reforrnist lead-
ership: economic elites have the most to lose and therefore would
choose risky political gambles only when convinced that the “certainty
equivalent” will remain intolerable. For this reason/ too, reformism
is linked to poverty.

The decision to initiate reform movements is also a strategic de-
cision: the value to an individual of the gamble to demand political
reform is determined in large partby the decisions of others. When
protesting politically, it is safer to be a member of a crowd. The
implication is that coordination is essential to the politics of reform.
Knowing that others have acted or will act decreases the perceived
risks of acting. If reformism appears to be sweeping through the
continent or assumes the status of a force of history, then being a
reformer becomes less dangerous. The strategic nature of initiating
reforms thus helps explain the pronounced element of contagion in
reformist politics. It helps explain the diffusion rate of reformist move-
ments and their geographical patterns. It also helps explain the sig-
nificance of external actors. Increasingly, international agencies de-
mand domestic politicat reforms in Africa. In responding to these
external demands, some African governments themselves become
reformist. And the interventions of these agencies encourage indi-
viduals to enter the reformist ranks, both by reducing the threat of
reprisal and by convincing them that, rather than being isolated dis-
sidents, they are part of a concerted movement.

The impetus for refqtro then, rather thaloriginating in the ranlg
of commercial or industrial capital. originates within fhe ranks of .;

people, then, want good government.-
In addition to their willingness and ability to give voice to their

demands for reform, professionals possess another advantage: access
to independent assoCiadons-the church or the law association, for
example. within these associations, they practice the arts of self-
govetllunce, orgarizing activities, developing tactics, canvassing for
irembership, riising funds, making rules, and so forth’ In addition,
they use these assoiiations to develop, debate, and make public an
alternative vision of national governance. Their associations provide
lines of communication, some of which are privileged (the relationship
between lawyers and their clients or between the clergy and their
congregatiotti, for example). These associations afford their members

inteina=tional ties, linking prominent dissidents with professionals in
other nations, while enabling the latter to monitor African govern-
ments and defend the libertybf their African colleagues. Professional
associations thus provide the organizational infrastructure for dissi-
dent movements in Africa.

The human capital of old guard politicians is also geographically
specific. DefeatElin previous political struggles, the opposition waits
ot, th” sidetnes for an opportunity to reenter the fray. Abroad, they
would ioin those-wltU, onie at the center of events, have been thrust
to the potitic;iinargins, swelling the dejected ranks of political ref-
ugees. Unknown iry say, Paris, they remain famous at home’ They
cin transfer neither their reputations nor their political capital: their

” knowledge of.wh6 holffgrudges against the old regime or w_ho believe
thgir ambitions couldbe realiZed more rapidlyby toppling the existing




at home rather

!-atterned: itJ location ii influenced by the extent and duration of
economic decline and its timing by forces that give assurance that, if
reform is initiated, others will join’

The Impulse to Ret’orm in Africa

newly democratic governments, they are-likely to choose the latter.
Friends of Africa, therefore, view political change as a necessary re-

quirement for continued funding for development assistance, and
ihey proa recipient regimes to open up to political reformers oI to
initiate reform themselves.

Domestic reform in Africa thus presents an odd spectacle. A/hile
demanded domestically, it is supplied internationally; it is also sup-
plied by the very governments whose behavior led to the demand for

reform in the first Place.
With governments taking the lead, political change in Africa could

simply yield the recreation of the old order. Features of the dynamics
give hope, howevdr, that change will result in something more than
ihat. PoliUcal reform has been undertaken as part of economic reform.
The reduction of controls and the movement toward market prices
eliminates rents; and indeed, protests over the losses of these rents,
particular$ by civil servants, help fuel political opposition’ But if
go,rertt-ettts succeed at this task and create strong economic bu-
ieaucracies in their efforts at economic reform-bureaucracies able to
resist distributive claims and to minimize economic distortions-then
the beneficiaries of the old order will not be able to recolonize the
new. They will have been disinherited. Possibilities thus exist for
political realignment, rendering a return to old practices less than


Top-Down Reform

This chapter began by noting that the behavior of African govern-
ments in the postindependence period both weakened their econo-
mies and shaped the nature of Africa’s politics. It ends by demon-
strating that the very policies that made reform …



This essay addresses the following four questions: What is
meant by state failure? What causes states to fail? What is the
scope of this phenomenon in contemporary Africa; and how
have insiders and outsiders responded to this process?
Different people find these questions important for different

reasons. For the citizens whose states fail the impacts upon
their daily lives are rarely uniform: they can range from
immense to negligible depending on a wide range of factors,
including how much control the state previously exerted over
its citizens, or how far the inhabitants happened to live from
the capital city and other major urban centres. For Western
governments, on the other hand, state failure in Africa is
commonly viewed as both a moral catastrophe and, especially
after the terrorist attacks on New York and Washington, DC, of
11 September 2001, a security threat. These dual concerns
were neatly elucidated by the then British Secretary of State
for Foreign and Commonwealth Affairs, Jack Straw, in Sep-
tember 2002. When confronted with state failure, Straw
suggested that ‘we cannot but be concerned at the implications
for the human rights and freedoms of those who are forced to
live in such anarchic and chaotic conditions. Yet the events of
11 September 2001 devastatingly illustrated a more particular
and direct reason for our concern. For it dramatically showed
how a state’s disintegration can impact on the lives of people
many thousands of miles away, even at the heart of the most
powerful democracy in the world. In these circumstances
turning a blind eye to the breakdown of order in any part of
the world, however distant, invites direct threats to our
national security and wellbeing. I believe therefore that pre-
venting states from failing and resuscitating those that fail is
one of the strategic imperatives of our times.’
Africa is commonly viewed as a particular cause for concern

because it is here that the phenomenon of state failure is most
widespread and deeply entrenched. In the case of the British
Government, this produced a spate of bureaucratic activity
culminating in a report in 2005 by the Prime Minister’s
Strategy Unit entitled Investing in Prevention. In a similar
vein, after 11 September 2001 the US Government stated it
was ‘now threatened less by conquering states than…by failing
ones.’ As a result, the US National Security Strategy published
in March 2006 acknowledged that ‘our security depends upon
partnering with Africans to strengthen fragile and failing
states and bring ungoverned areas under the control of effec-
tive democracies.’ State failure is thus a serious concern for
both insiders and outsiders and it is on Africa that the
international spotlight has most commonly fallen.
The prominence of such discourses about state failure has

also generated a great deal of controversy. One important line
of criticism has drawn insights from the post-colonial studies
literature and called upon analysts to refrain from using terms
such as ‘failed states’, ‘weak states’ and ‘quasi-states’ on the
grounds that they are based on ethnocentric assumptions that
depict African states as imperfect copies of Western European
and North American states and judge them according to
external standards in order ‘to promote and justify their
political and economic domination by Western states and other
international actors.’1 This essay is not intended to refute the
post-colonial critique for it raises many sensible questions
about the use of terms like ‘failure’, ‘fragility’ and ‘weakness’.
Nor is it intended to advocate the post-colonial critique by
refusing to use such terms. Rather, this essay will provide an
overview of the dominant discourses on state failure in Africa
and will attempt to understand what responses, from both
insiders and outsiders, they have helped facilitate.

Discussions of state failure are essentially about the inter-
relationships between patterns of authority, political control

and institution-building. Put another way, analysing state
failure in Africa requires a keen sense of the shifting config-
urations of power on the continent and beyond. In most of the
literature on the subject, the idea of ‘failure’ is invoked in two
main senses, referred to in this essay as the failure to control
and the failure to promote human flourishing.2

The Failure to Control
In the first sense, failure is understood in terms of the inability
of state institutions to control actors and processes within a
given territory. Robert I. Rotberg maintains that ‘failed states
cannot control their peripheral regions, especially those
regions occupied by out-groups…Plausibly, the extent of a
state’s failure can be measured by the extent of its geographical
expanse genuinely controlled (especially after dark) by the
official government.’ It is important to remember, however,
that control and failure should not been seen as absolutes. A
‘failed’ state in this sense of the term might successfully control
some of its territory but not all of it. Sudan, for example, is
commonly classified as a failed state yet it continues to exert
effective control over large portions of its territory and can
wreak havoc and terror on some of those individuals and
groups who contest its authority in those areas.
This suggests that viewing the phenomenon of state failure

in absolute terms and through solely statist lenses is not
always particularly helpful. Rather analysts need to appreci-
ate the degrees of success and failure that can exist within a
single state and recognize that so-called ‘failed states’ are
usually made up of numerous (and often interconnected) zones
where different sources of authority may dominate the local
governance structures. In any given zone the authority in
question may vary. Indeed, as Rotberg noted, it may differ
considerably within the same zone depending on the time of
day or night. The authority structure could be an organ of the
state’s official government but it may also be, among other
things, an insurgency or guerrilla movement, a clan, a militia,
an extended family, a spiritual leader, an international peace
operation, or even a transnational corporation or a non-gover-
mental organization.
To give one example, the collapse of the Somali central state

did not automatically exclude the possibility that zones of
alternative forms of governance and authority existed within
Somalia’s officially recognized international borders. As Ken-
neth Menkhaus has observed, since 1991 ‘Somalia has repeat-
edly shown that in some places andat some times communities,
towns, and regions can enjoy relatively high levels of peace,
reconciliation, security and lawfulness despite the absence of
central authority.’ These authority structures have come in
various shapes and sizes. They have included local polities
comprised of coalitions of businessmen, clan elders and Muslim
clergy involved in administering financial services and Shari‘a
courts, and larger-scale structures such as the administrative
centres of the ‘Republic of Somaliland’ (1991–), ‘Puntland’
(1998–), the Rahanwin Resistance Army’s administration of
Bay and Bakool regions (1998–2002) and the Banaadir Regio-
nal Authority (1996).
When analysing state failure in Africa in this first sense,

analysts and practitioners would thus do well to reject a state-
centric ontology in favour of a neo-Gramscian frame of refer-
ence, wherein the world is not simply seen as being made up of
clashing states in an anarchic international system but,
instead, is constituted by the complex inter-relationships
between states, social forces and ideas within specific world
orders.3 Adopting this ontology is far more useful for analysing
state failure because as Timothy Raeymaekers correctly
observed, what we are witnessing in several cases of so-called
‘state failure’ is actually better understood as ‘neopatrimoni-
alism without the state’. That is, systems of patron-client
relations that may or may not be linked to the official institu-
tions of state power. Arguably the closest Western officialdom

www.europaworld.com 1

has come to adopting such a perspective is the US Govern-
ment’s anxiety about what it terms ‘ungoverned spaces’,
‘defined as geographic areas where governments do not exer-
cise effective control.’4 Unfortunately, this misses the crucial
point that just because official governments do not control
these areas it does not necessarily mean that they are com-
pletely lacking other structures of governance.

The Failure to Promote Human Flourishing
Failure is alsocommonly used in a second sense to highlight the
ways in which states, either because of a lack of capacity or a
lack of political will, fail to provide public goods to their entire
population rather than favouring one or other particular
segment of it. The idea that states have a responsibility to
provide their citizens with certain basic rights has long been an
issue of debate within international relations dating back at
least as far as notions of popular sovereignty articulated by
Jean Bodin in the 16th century. Since the publication in late
2001 of a report by the International Commission on Inter-
vention and State Sovereignty, this line of argument is now
commonly known as the ‘responsibility to protect’. Although
African governments have jealously guarded traditional ideas
about sovereignty and non-intervention, the responsibility to
protect idea has made significant headway in recent years and
with it has come a plethora of literature speaking of ‘failure’ in
these terms. Despite these longstanding reservations, in Sep-
tember 2005 African states, along with the rest of the UN
General Assembly, formally accepted the responsibility to
protect idea. As defined in the World Summit Outcome docu-
ment: ‘each individual State has the responsibility to protect its
populations from genocide, war crimes, ethnic cleansing and
crimes against humanity. This responsibility entails the pre-
vention of such crimes, including their incitement, through
appropriate and necessary means. We accept that responsi-
bility and will act in accordance with it.’ When genocide, war
crimes and other atrocities occur states can be said to have
failed in their responsibilities to their citizens.
Understood in these two senses, state failure on the African

continent is a widespread phenomenon. It is important to note,
however, that both these views of failure are based upon a
particular conception of statehood: what Rotberg calls ‘the
fundamental tasks of a nation-state in the modern world’ and
what William Zartman refers to as ‘the basic functions of the
state’. The particular idea of statehood that dominates discus-
sions about state failure was born in Europe and is usually
associated with the Treaties of Westphalia in 1648. That year is
thus commonly understood within mainstream international
relations theory as representing the birth of modern interstate
relations. Yet while this specific date of origin makes for neat
theory, it rests upon a dubious and mythical history. As Benno
Teschke has argued, even in its European birthplace the
practice of Westphalian statehood as opposed to the ideal of
Westphalian statehood did not emerge until well after 1648.
Specifically, Teschke has shown how modern international
relations based on the Westphalian ideal of statehood only
began with the conjunction of the rise of capitalism and modern
state formation in England. Thereafter, the English model
influenced the restructuring of the old regimes of the European
continent, a process that was incremental and highly uneven
and was not completed until the First World War.
The relevant point for this discussion is that the nature of

statehood itself is contested rather than obviously apparent.
Specifically, as Christopher Clapham has argued, the West-
phalian ideal rests on ‘unsure foundations’, not least because in
some parts of the world ‘the essential conditions for statehood
cannot plausibly be met.’ The ‘fundamental tasks’ of statehood
envisaged in this Westphalian ideal revolved around the
provision of security, welfare and representation. In particu-
lar, the defining characteristic of the Westphalian ideal of
statehood has been the right of states to exercise five monopoly
the right to monopolize control of the instruments of violence;
the sole right to tax citizens;
the prerogative of ordering the political allegiances of citizens
and of enlisting their support in war;
the sovereign right to adjudicate in disputes between citizens;

the exclusive right of representation in international society
which has been linked with the authority to bind the whole
community in international law.
Even in Europe, as Teschke observed, the practical acquisi-

tion of these monopoly powers sometimes took centuries of
often violent turmoil and social upheaval. Compared with
Europe and viewed from the perspective of the longue durée,
it is clear that most states in Africa are still mired in the
relatively early stages of state formation. Consequently, it
should come as little surprise that the practical acquisition of
these monopolies has been uneven across the continent.
Although it has been similarly traumatic and drawn out, the

process of state-building has unfolded differently in Africa
than it did in Europe. Unlike in Europe where state borders
were demarcated with reference to their neighbours, in Africa
state power tended to radiate from a focal core (usually the
capital city) that only rarely came into direct confrontation
with its neighbouring governments. As a basic rule of thumb,
the further one travelled from this core, the weaker the state’s
control became. This fact rendered the state borders drawn up
by the European colonial powers in Berlin in 1884–85 rela-
tively meaningless, or at least highly porous, for many prac-
tical aspects of the local inhabitants’ daily existence, including
commerce or communicating with individuals who were offi-
cially ‘foreigners’, but who belonged to the same ethnic or tribal
Understood in these terms, the issue of ‘failed states’ in

Africa is largely about the extent to which the Westphalian
ideal of statehood has taken root in the rather different and in
many ways inhospitable conditions found on the continent. As
the Organization of African Unity’s (OAU) charter made
abundantly clear, the ideal of Westphalian statehood clearly
attracted many advocates among Africa’s first generation of
post-colonial élites. It was also helped by the willingness of the
great powers within international society to grant these states
international recognition. The practical realization of this
ideal, however, has been far more contested and uneven. As
a result, from the outside, African states often looked like the
Westphalian ideal in that they were recognized members of
international society and their representatives sat on the
councils of various international organizations. On the inside,
however, these governments were often considered illegiti-
mate by much of the local population and wielded the institu-
tions of state to subdue political opponents and benefit their
supporters. These were, in Robert Jackson’s famous phrase,
quasi-states: legal fictions that rarely commanded much in the
way of national loyalty or the power to control developments
throughout their designated territory.
What this means for an analysis of state failure is simply that

depending on the local conditions, ‘failure’ is far more likely in
certain parts of the continent than others. More specifically, as
Clapham has argued ‘those areas of Africa that maintained
reasonably settled and effective state structures during the
period prior to colonialism are proving best able to do so as the
institutional legacies of colonialism fade.’ Where these struc-
tures were weak other forms of authority (familial, spiritual,
ethnic, etc.) have filled the vacuum.

There is no simple or single formula for understanding the
causes of state failure in Africa. Nevertheless, the available
literature on the subject often makes at least two relevant
general distinctions. Thefirst distinction is between states that
fail because of a lack of relevant capacities and those that fail to
promote the interests of all their inhabitants through political
choice, often with the intention of benefiting the incumbent
regime and its supporters at the expense of another group
within the state. Robert Mugabe’s ongoing manipulation of
ZANU—PF and state power in Zimbabwe is a paradigmatic
example of a regime choosing to deny basic rights to certain
segments of its population in an attempt to bolster regime
security. The dynamics in this case are somewhat different
from instances where a regime may well want to restore order
to part of its territory but lacks the relevant capacities to do so.
These dynamics are apparent in, for instance, the Ugandan
Government’s inability to quash the Lord’s Resistance Army

GENERAL SURVEY State Failure in Africa

2 www.europaworld.com

and Sudan’s inability to defeat the Sudan People’s Liberation
Movement/Army. Such incapacity may sometimes lead to
political compromises. At other times the result is simply
longstanding stalemates and the de facto partition of a state’s
territory. Outside of the military sphere, a government might
wish to enhance the development prospects of its citizens but
lacksthe necessary resources andinstruments. The incumbent
governments in Liberia and Sierra Leone are cases in point.
A second distinction points to the differences between struc-

tural and contingent causes of state failure. In the structural
category three main arguments are commonly advanced.
Firstly, as noted above, the Westphalian ideal of statehood
has not successfully taken root across all of Africa because local
conditions were inhospitable to state-building and exerting
high levels of state control over local societies. Although
international society helped the process by granting recogni-
tion to Africa’s newly independent states—many of which
became the archetypal examples of Jackson’s quasi-states—
it could not ensure that their inhabitants invested a great deal
of faith in, or commitment to, them. Not long after indepen-
dence, however, Cold War politics meant that the superpowers
often made genuinely national nation-building even more
difficult by stoking the fires of dissent within many African
states in the name of either communism or capitalism.
A second structural argument has revolved around the

challenges posed by political geography, especially resources
and environmental factors. In this case the point is that some
African states that were creations of the European colonial
powers were not endowed with a physical environment con-
ducive to administering an effective state. In particular, states
such as those in the West African savannah suffered from
extremely low densities of people, which made administration
and social control both costly and difficult. The same was true
for much of Africa since large areas of it have ecologies that
cannot easily support high densities of population, not least
because over 50% of the continent suffers from inadequate
rainfall that makes inhospitable environments for both human
settlement and agriculture. Indeed, it is arguably only the
Great Lakes region and the Ethiopian highlands that have
sustained relatively high densities of people.
A third structural argument has applied the concept of the

security dilemma to explain how fear of an ungoverned future
can propel the actors within states to hasten the collapse of
central government once public order begins to erode and a
situation of domestic anarchy seems likely to emerge. Here the
suggestion is that the Hobbesian fear that lies at the heart of
the security dilemma explains why groups begin to think that
their potential rivals will not be restrained by state authority
once the institutions of state have started to disintegrate.
Analytically speaking, the crucial focus becomes understand-
ing the ‘tipping point’ beyond which actors start to behave as if
domestic anarchy exists, even if that is not entirely the case. At
that stage, the dynamics of the domestic security dilemma may
ensure that their conviction that state collapse and anarchy is
imminent becomes a self-fulfilling prophecy.
A fourth argument has suggested that when it comes to state

success in Africa ‘bigger isn’t better’. Specifically, the authors
of a recent edited volume concluded that most of Africa’s large
states have exhibited ‘varying degrees of dysfunctionality
(defined as the lack of provision of welfare and opportunity
to the population) and, excluding to a degree South Africa, a
sustained period of civil unrest, economic decline, state atro-
phy and social corrosion.’5 In a series of case studies, the
authors suggest that in Angola, the Democratic Republic of
the Congo (DRC), Ethiopia, Nigeria and Sudan (all of which are
in the top five African states in terms of population or landmass
or both), ‘the actual size of these states is itself a problem for
governance’ and that rather than being natural leaders, they
‘are actually problems for their associated regions.’
The contingent causes of state failure in Africa are even more

numerous with five main factors commonly cited within the
literature. Crudely summarized, these refer to the influence of
bad leaders, predatory actors such as warlords and so-called
‘spoilers’, bad economic policies, bad environments and bad
First, much of the blame for state failure has been heaped

upon Africa’s leaders, not least Maj.-Gen. Mohammed Siad

Barre (Somalia), Dr Siaka Stevens (Sierra Leone), Mobutu
Sese Seko (Zaire/the DRC), Gen. Samuel Doe and later Charles
Taylor (Liberia) and Robert Mugabe (Zimbabwe). In many
African states such leaders and their political élites have been
criticized for pursuing patrimonial politics that seek to use
external sources of aid and finance to reward their supporters
and weaken their opponents rather than to pursue genuinely
national development strategies. Second, warlords and other
‘spoilers’ have been blamed for inflaming ethnic tensions and
hastening state failure. The motivations of these predatory
actors vary from case to case, but a common claim is that they
have pursued violent strategies in order to accumulate wealth
through the control of formal and/or informal markets. In this
view, weak or failing state institutions provide an environment
from which such warlords and ‘spoilers’ can profit.
The third set of contingent factors concerns the political

economy of state failure, especially the adoption by govern-
ments of ‘bad’ macroeconomic policies resulting in fiscal deficits
and balance of payments crises, and the paradoxical effects of
structural adjustment policies encouraged by a variety of
international donors. As Nicolas van de Walle has argued,
both of these factors encouraged a ‘hollowing out’ of the state,
which, in turn, increased ‘the chances that minor political
incidents and disputes could cause the descent into failure.’
Such political economies did not, however, automatically
produce failed states. Hence, although Zaire/the DRC and
Sierra Leone were both ‘hollowed out’ before failing, states
such as the Central African Republic (CAR), Malawi and Niger
were also weakened by economic failure but did not suffer a
similar fate.
A fourth commonly cited factor relates to the proliferation

and availability of armaments, especially small arms and light
weapons, in many of Africa’s weak and fragile states. An
environment awash with arms makes it difficult for govern-
ments to control all of their territory or protect all of their
citizens because, as Michael Klare suggests, ‘antigovernment
formations can readily assemble sufficient weaponry to mount
a revolution or insurgency.’ A fifth contingent cause of state
failure concerns the role played by actors within neighbouring
states. These have tended to be either incumbent governments
hostile to their neighbouring regimes (e.g. Taylor’s destabili-
zation of Sierra Leone throughout the 1990s), or insurgent
groups which destabilize their target state with or without
support from the government of their (temporary) host state
(e.g. the Rwandan Patriotic Front’s use of Uganda as a base for
its operations before it invaded Rwanda in 1990, or Hutu
genocidaires using eastern Zaire/the DRC to destabilize Paul
Kagame’s regime after the 1994 genocide).
All of these factors can play a role in state failure but the

current state of knowledge remains far too vague accurately to
predict the tipping points in particular cases.

As noted above, failed states raise challenges both for the great
powers concerned about what threats might come out of them
and for the locals who have to endure life inside them. Indeed, it
has been suggested that since ‘the end of the Cold War, weak
and failing states have arguably become the single-most
important problem for international order.’6 Although state
failure is not confined to Africa the problem is arguably more
widespread, deeply rooted and pressing here than in any other
Failed states can spawn a variety of transnational security

problems with terrorism, proliferation of weapons of mass
destruction (WMD), crime, disease, energy insecurity, and
regional instability chief among them. Not all of them, how-
ever, are equally prevalent in Africa’s cases. With the notable
exceptions of actors operating out of Sudan and Somalia,
transnational terrorism has been relatively rare in sub-
Saharan Africa. The same could also be said for WMD pro-
liferation. In contrast, small arms and light weapons prolif-
eration, transnational crime (especially the illicit trade in
drugs, arms, minerals, petroleum, timber, wildlife and human
beings), infectious diseases (including HIV/AIDS, malaria,
tuberculosis, hepatitis B, Ebola, measles, and the West Nile
virus), and political instability in the Horn of Africa and the

GENERAL SURVEY State Failure in Africa

www.europaworld.com 3

Gulf of Guinea oil states do pose significant threats and
challenges to both the locals and the great powers.
It is important to note, however, that these challenges are

unevenly distributed across Africa’s failing states. Trans-
national criminals, for instance, tend not to operate in areas
of complete state collapse (such as Somalia), but instead prefer
areas where a basic degree of physical and financial infra-
structure exists and where bureaucrats and officials are
susceptible to bribery (such as Kenya, Nigeria and South
Africa). Cells of transnational terrorist networks are likely
to require similar levels of infrastructure and at least a degree
of order if they are to use failing states for anything other than
transit routes and temporary bases of operations. The major al-
Qa’ida attacks against US embassies in 1998, for instance, took
place in Kenya and Tanzania but were allegedly orchestrated
from a partially failed state in the form of Sudan and an almost
entirely collapsed state in the shape of Somalia. Viewed from a
longer-term perspective, however, most terrorism in Africa has
been nationally oriented and targeted against white-minority
rule or in specific revolutionary settings, notably Ethiopia and
During the early stages of the 21st century, Africa has

provided many of the usual suspects on the lists of the world’s
failed states including Chad, the CAR, the DRC, Liberia, Sierra
Leone, Sudan and Zimbabwe. In addition, Somalia is usually
classified in a league of its own as having collapsed altogether
rather than simply failed. Africa’s leading status in these
rankings is confirmed by four popular attempts to categorize
different types of state failure, namely, the World Bank’s
Governance Matters data set, the Failed States Index, the
UN Development Programme’s (UNDP) Human Development
Index, and Freedom House’s annual surveys of political free-
The World Bank’s Governance Matters 2008 data set ranks

world-wide governance indicators, covering 212 countries and
territories. It defines governance as ‘the set of traditions and
institutions by which authority in a country is exercised. This
includes (1) the process by which governments are selected,
monitored and replaced, (2) the capacity of the government to
effectively formulate and implement sound policies, and (3) the
respect of citizens and the state for the institutions that govern
economic and social interactions among them.’ The 2008 report
measured six dimensions of governance between 1996 and
2007: voice and accountability, political stability and absence
of violence, government effectiveness, regulatory quality, rule
of law, and control of corruption. The …

Two Leadership Styles and Patterns of Political Liberalization
Author(s): Jennifer Widner
Reviewed work(s):
Source: African Studies Review, Vol. 37, No. 1 (Apr., 1994), pp. 151-174
Published by: African Studies Association
Stable URL: http://www.jstor.org/stable/525117 .
Accessed: 03/11/2011 16:07

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected]

African Studies Association is collaborating with JSTOR to digitize, preserve and extend access to African
Studies Review.


Two Leadership Styles and Patterns of
Political Liberalization

Jennifer Widner

In his novels and essays and most recently in the pages of The
New York Times, Nigerian author Chinua Achebe has repeatedly
claimed that the “trouble with Nigeria” lies with its leaders (1982,
1993)-and few of those who followed the pronouncements of General
Babangida would say that the actions of the Nigerian head of state
were without import for the political future of the country. Achebe
finds an intellectual ally in social scientist Giuseppe de Palma (1992),
who argues that the actions of decision makers at critical junctures in
the process of political opening shape the range of choices and the
kinds of risks they and their successors face subsequently. Although the
conditions that make democratic consolidation more or less difficult
may be a function of levels of socioeconomic development, the presence
or absence of a civic culture, or the existence of a middle class, the de-
cisions of leaders matter. Democracies are crafted, not born whole.

This essay tries to identify the sources of difference in leadership
style during a period of political reform. It proceeds from two observa-
tions. First, tactical and strategic choices of leaders are partly a func-
tion of the incentive structures political and economic institutions offer.
Electoral rules, the amount of central control over important electoral
resources such as the media, participation in regional monetary agree-
ments-all of these institutions shape the options available for build-
ing coalitions.1

How leaders perceive the constraints rules and organizations cre-
ate and the costs, benefits and risks attached to alternative courses of
action affects outcomes too, however. Personal experience, schooling,
the ideas current among a leader’s friends and advisers-these also in-
fluence decision making. This repertoire of skills, styles and preferences
enables some decision makers to learn from their mistakes while others
repeat the errors of the past, and it prompts some to try to change the
rules of the game by re-designing institutions instead of respecting the
limits set by existing structures. In short, character matters in the way
leaders incorporate the features of the political landscape social
scientists so often take as given into their decision making.

The essay probes the interrelationship between institutions and
ideas in styles of governance by comparing the strategic choices of Felix

African Studies Review, Volume 37, Number 1 (April 1994), pp. 151-74.


Houphouiet-Boigny of COte d’Ivoire and Daniel arap Moi of Kenya dur-
ing the period of political liberalization that swept Africa between
1989 and mid-1993. In some important respects these two cases were sim-
ilar. By the end of 1992, the governments of both leaders had conducted
multiparty elections, and both incumbents survived the electoral con-
tests and maintained their positions. In other ways, their strategic
choices in negotiating pressures for political reform diverged dramati-

Under Houphouiet-Boigny, COte d’Ivoire moved to a multiparty
system in May 1990 and held competitive elections several months
later.2 The aging president was not a strong advocate of political liber-
alization, but he nonetheless turned to multiparty competition early in
the wave of political change that swept Africa, beginning in 1989.
Irregularities marked some aspects of the elections his government or-
ganized. For example, so little time elapsed between the legalization
of opposition parties and the conduct of elections that it was difficult
for groups other than the established Parti Democratique de COte
d’Ivoire (PDCI) to mount effective campaigns in rural areas. Loubards,
or thugs, turned up at opposition campaign rallies in some parts of the
country in the early weeks of the campaign period. In urban areas,
people sometimes did not know where polling places were located
(many students assumed irregularities would occur and did not bother to
try to cast ballots), and both the PDCI and opposition parties
attempted to buy votes in some areas. After the elections, the tempo-
rary jailing of the main opposition leader, Laurent Gbagbo, and peri-
odic clashes with students and teachers threatened the new political

On the other side of the balance sheet, the level of electoral vio-
lence was low by comparison with many other countries, except on the
university campus, and the incidence of ballot-tampering was also suf-
ficiently limited that observers judged the election acceptable.
Although the government had jailed the main opposition leader,
Laurent Gbagbo, after the election, on the grounds that his activities
incited violence, by the third quarter of 1992, the government had not
only released Gbagbo, but it also had allowed him to speak at length on
state-run television. It tolerated a flourishing opposition press as well
as criticism in official newspapers and from technocrats who voted
against the PDCI, and it conferred legal status on new, autonomous syn-
dicates. Opposition members of the national assembly worked side by
side with members of the long-established PDCI, their jokes attracting
laughter and their comments taken seriously.

This style was partly promoted by the kinds of institutional con-
straints Houphouet-Boigny faced, as detailed below. It also had prece-
dents in the Ivoirien president’s handling of earlier periods of discon-
tent within his country, however. Felix Houphouet-Boigny generally


Leadership Styles and Political Liberalization

controlled political opposition through co-optation, periodic reasser-
tion of authority over expanding and proliferating patronage networks,
and tight control of the resources necessary to build a credible opposi-
tion. Political violence was a fallback strategy used in comparatively
fewer circumstances than was typical of many other sub-Saharan
African countries, although it was most definitely part of the tactical
repertoire. Most notably, Houphouit-Boigny brought younger genera-
tions into his cabinet and into the party roughly every six years after
independence. He convened national “days of dialogue” during periods
of economic crisis, drawing hundreds of representatives of different so-
cial groups into conversation with officialdom. He expanded political
competition first in 1980, when he sponsored a move from single-party
list voting to multi-candidate, single-party elections, then in 1990,
when, on 3 May, his government sanctioned a move to a multiparty

As French scholar Jean-Franqois Medard has observed (1991),
Daniel arap Moi’s style was different. By contrast with Houphouet-
Boigny, Moi long resisted pressures for expanded political participa-
tion. At a time when many governments in sub-Saharan Africa were
yielding to new popular pressures for participation, he reduced oppor-
tunities for political competition through proscription of several forms
of association, including ethnic welfare societies. He long resisted pres-
sure to legalize opposition parties and continued to jail those who criti-
cized his government. Only at the end of November 1991, when interna-
tional donors suspended aid pending action to clean up corruption and
reduce levels of political repression, did Moi promise elections.

In the campaign period, the Kenyan president differed from
Houphouiet-Boigny in his relatively heavier reliance on security forces
to monitor the opposition and disrupt political participation.3 He used
the Kenya African National Union (KANU) youth wing and the secret
police to monitor opposition, and his close ally, Nicholas Biwott, coor-
dinated use of the employees and trucks of Kenya Power and Lighting to
mount surveillance campaigns during and after the election. Like
Houphouit-Boigny, Moi employed patronage to shatter clientelist
networks that provided potential bases for competitors. He relied more
heavily than his Ivoirien counterpart on violence as a political tactic,
including direct promotion of ethnic clashes and killings in part of the
Rift Valley, and his resort to violence persisted at a higher level after
the elections took place.

In the view of observer teams, most of the standards for fair elec-
tions were not met (IRI 1993). The KANU government rigged the elec-
tions not by stuffing ballot boxes but by manipulating the electoral pro-
cess. Civil servants delayed opposition campaigns by refusing to process
documents in a timely manner. The provincial administration pre-
vented opposition candidates from offering nomination papers in 17 con-



stituencies in the Rift Valley. In some places, it blocked access of par-
ties and church groups to public places.

In the campaign period, KANU used its access to facilities and
capital to great effect. Observer teams noted that vehicles belonging to
government ministries sometimes bore KANU campaign posters and
transported KANU campaign workers and candidates. There were also
allegations that KANU financed its campaign by printing money,
adding several billion new bank notes dated 2 January 1992 into the
country’s money supply and triggering higher inflation (IRI 1993, 26).
Youth for KANU allegedly had access to 1.2 billion Kenya shillings
through the Hospital Insurance Fund.

Election-day irregularities observed by foreign teams included
late delivery of voting materials and late opening of polling places, in-
sufficient voting materials in some polling stations, errors on ballot pa-
pers, registration lists with voters’ names missing, registration lists
with pages missing, no provision for illiterate voters to cast secret bal-
lots, and general confusion. With 36.45% of the vote, Moi and KANU
remained in power, however. One of Moi’s first acts was to suspend
parliament, although he reconvened the legislature in late March

This article argues that the institutions associated with the po-
litical economies of COte d’Ivoire and Kenya accounted for much of the
divergence in the strategies leaders adopted during the period 1989
through 1893 but that a straightforward institutional analysis does not
distinguish a Houphouet-Boigny from an Eyadema in Togo, nor a Moi
from an Ali Hasan Mwinyi in Tanzania. To more completely account for
the variation observed, it is essential to consider the repertoires of
skills and understanding to which heads of state have resorted.
Differences in leadership style are in part a function of the perceptions
the men developed in the early part of their careers, although
institutions shaped the short-term choices they faced. The formative
events in Houphouiet-Boigny’s political career taught lessons that later
made it possible for “Le Vieux” to contemplate preemptive
liberalization. Daniel arap Moi’s early political lessons did not
develop the same kinds of skills in coalition management and
heightened the leader’s uncertainty about his own ability to survive in
competitive politics.

The Comparative Context

The Kenya and C6te d’Ivoire cases can contribute importantly to
understanding the relative strength of ideas and institutions in shaping
leadership styles because their many similarities permit social scien-
tists to discount the significance of a number of possible sources of varia-


Leadership Styles and Political Liberalization

tion. They make a useful paired comparison because they resemble one
another in the character of their resource bases, their independence-era
choices of development strategy, general level of economic performance
until the late 1970s, level of ethnic diversity and security situation.

Although the political economies of the two countries differ in
significant ways, as outlined below, Kenya and C6te d’Ivoire share
some important economic characteristics. The two countries have simi-
lar resource endowments; both draw much of their revenues and foreign
exchange earnings from export agriculture and specifically from sale of
beverage crops-coffee, cocoa (COte d’Ivoire ), and tea (Kenya). In
consequence, both experienced the “beverage boom” of the mid-1970s
and the sudden cash windfalls which that generated as well as
declining terms of trade during the 1980s for main crops. Production
systems used in growing these commodities are also similar. Although
there are important industrial plantations in each country, a
substantial portion of beverage crop production remains in the hands of
smallholders, who rely on a combination of family members, share
tenants and hired workers for farm labor.

Economic strategies pursued at independence also shared impor-
tant characteristics. Unlike many of their neighbors, these two coun-
tries pursued export-oriented policies at the same time they sought to
substitute domestic production for some imports. They offered growers
relatively higher proportions of the world market price for the com-
modities they produced, by comparison with neighboring countries in
Africa, and they limited indirect taxation of export crop producers by
keeping official exchange rates much closer to market rates than did
most other governments south of the Sahara. They invested in rural ac-
cess roads and other infrastructure designed to promote production by
small farmers. They tolerated emergence of private entrepreneurs, al-
though the governments also invested heavily in some sectors. Finally,
they cultivated trade ties with western countries and encouraged for-
eign direct investment, in contrast to Tanzania and Ghana.

In sociological and cultural terms, the two countries also share
some important characteristics. Although majorities are still rural,
rates of urbanization and rates of population growth are extremely
high, compared to other developing countries.

Further, in neither country does one ethnic community predomi-
nate. The largest ethnic group in each encompasses less than 20% of the
population, and coalition building among ethnic communities is essen-
tial to secure a national electoral victory.

It might be argued that because of the strength of the Kikuyu
community in the Kenyan economy, Moi, a Kalenjin, faced an especially
difficult negotiating challenge. It is not abundantly clear than the sit-
uation of Houphouet-Boigny was much different. Certainly the Kikuyu
held especially strong positions in the Kenyan economy during the first



two decades after independence, while commercial and industrial
elites in COte d’Ivoire typically came from a wide variety of communi-
ties. Over the past decade, the Nairobi-based business elite has grown
to include members of other groups too, in large part through Moi’s in-
tervention to break up the dominance of his predecessor’s old support-
ers. By contrast with Jomo Kenyatta, Kenya’s first president, both Moi
and Houphouet-Boigny have had to cultivate cross-ethnic ties to secure
business support.

Finally, the two countries are like one another in that although
both have reputations for high levels of political stability, both have
also been the site of ethnically based insurgencies and coup attempts.
The similarities between the security situations in the two countries
mean that it is not possible to account for the differences in strategy
leaders pursued in the period of political reform. First, Kenya and COte
d’Ivoire stood out from their neighbors in having survived to mid-1993
without a successful coup attempt, although members of the air force, in
league with other groups, mounted a sustained struggle for power in
Kenya in August 1982 and soldiers mutinied in COte d’Ivoire on several
occasions during the early 1990s-at one point seizing the airport. In
the aftermath of the more serious coup attempt in Kenya, the Moi gov-
ernment restructured parts of the military and jailed the alleged perpe-
trators. Although the event must certainly have contributed to a sense
of paranoia on Moi’s part, it did not distinguish his situation from
those of other leaders on the continent or from Houphouiet, especially in
the wake of measures to prevent a recurrence.

Second, both countries had experienced ethnic insurgencies of
which vestiges survived. In neither case did these threaten state secu-
rity, however. In Kenya the uprising of displaced Kikuyu squatters in
the Mau Mau rebellion complicated the task of maintaining political
order in the first years of independence. Moi’s predecessor, Jomo
Kenyatta, had to broker a compromise between the Mau Mau and the
rising Kikuyu commercial class while bringing other communities into
government. In the 1980s a shadowy organization, Mwakenya, re-
asserted the ideals of Mau Mau, but it had little evident grassroots
strength, despite its claims to the contrary.

Indeed, the Moi government masterfully rehabilitated many of
the former Mau Mau leaders, whom the Kenyatta government had kept
at bay, in an effort to undermine the power of Kikuyu politicians and
businessmen in their communities. Under Moi’s patronage, Kariuki
Chotara, Fred Kubai and other former Mau Mau acquired positions of
power for the first time, and others, such as Samuel Thiberi, received
support from the president to compete for the elected positions formerly
held by such prominent members of the Kikuyu community as former
Vice-President Mwai Kibaki. The one ethnic conflict that provoked
widespread violence under Moi’s regime was the clash between al-


Leadership Styles and Political Liberalization

legedly government-sponsored Kalenjin, Moi’s group, and Kikuyu farm-
ers residing in the Rift Valley.

In COte d’Ivoire, there were two brief ethnic insurgencies under
Houphouet’s rule. One was the uprising of the Bete community in and
around Gagnoa in the early 1970s. The second was the longer-term at-
tempt of the Sanwi peoples (an Agni group in the Southeast) to secede,
beginning in the late 1950s and continuing through the 1960s. Both were
resolved by force. The Bete conflict persisted through the 1980s in much
reduced form, and many Ivoiriens argued that the base of support of
opposition leader Laurent Gbagbo and his Front Populaire Ivoirien
(FPI) was Bete and represented a continuation of the Bete struggle for

The security situations were also similar in that by 1989, both
countries had civil wars raging next door. Hunger was a persistent prob-
lem for neighbors through most of the 1980s and both countries had be-
come bases for international relief operations. By 1989, however, civil
war in Liberia had created a refugee problem in the western part of the
COte d’Ivoire and had started to precipitate a back-flow of weaponry
and banditry into some parts of the country. To the north of Kenya, con-
flict was raging in the Sudan and brewing in Somalia, whose border ar-
eas with Kenya had long been closed to most traffic. Both governments
faced difficult security situations, but the challenges did not differ sig-
nificantly between them.

Because leadership styles, the outcomes in this analysis, vary,
while these conditions do not, it is possible to discount explanations
that locate the divergence between the Houphouiet and Moi strategies
in a country’s dependence on primary agricultural commodities, capital-
ist economic orientation, ethnic heterogeneity or underlying security
situation. The sources of difference must lie elsewhere.

Political Economies and Political Liberalization

The institutional legacies of French and British colonial rule
shaped the strategic choices of Houphouit-Boigny and Moi in the 1989-
93 reform period. They influenced the way the burdens of economic cri-
sis were distributed across members of the two societies, the extent to
which independent interest groups could mobilize protests that could
cripple the economy, the strength of political ties between the new op-
position and the grassroots, and the ease with which the head of state
could control critical electoral resources. The distinctive political
economies associated with a Francophone versus Anglophone heritage
created short-run constraints and opportunities for both leaders.

First, the economic crises of the 1980s affected the two countries
differently, creating more intense pressure from key elites for political



change in C6te d’Ivoire than in Kenya. COte d’Ivoire is a member of the
Franc Zone. That means it cannot devalue unilaterally. Further, and
more importantly for this analysis, it also means that the capacity of
the government to run a deficit and to borrow from its central bank is
limited. By international agreement, COte d’Ivoire could only borrow
limited amounts from its central bank, the BCEAO (Banque Centrale
des Etats de l’Afrique de l’Ouest), a bank owned by the several member
nations. It could not print money. In economic hard times, when the gov-
ernment ran a deficit, the leadership had to lay off public employees,
cut pay, or raise taxes sharply. None of these moves was at all popular.
The need to curb the government deficit inspired a renewed effort to
privatize public enterprises as well as thousands of layoffs from cash-
strapped parastatals. Sharply higher taxes and elimination of many
perks, such as free or subsidized housing or medical care, brought elites
into the streets between January and April 1990, precipitating the move
to multiparty rule on 3 May of that year.

Most African countries with Anglophone traditions have no agen-
cies of external restraint such as the Franc Zone. Kenyans felt the in-
creasing economic problems of their country in different ways, in conse-
quence. With greater control over the management of its own currency,
the government in Nairobi could borrow and print money, fueling
higher levels of inflation. By keeping nominal wages constant, the gov-
ernment could allow inflation to erode public sector wages. There was no
serious re-structuring and sale of public enterprises (indeed, the gov-
ernment created some new corporations in its effort to build political
support). The deterioration in purchasing power that occurred provoked
less protest than layoffs did in COte d’Ivoire, although over time it con-
tributed to the sense of malaise among Kenya’s elite. In short, the gov-
ernment counted on money illusion to buy time and to keep discontented
elites, the prime-movers behind political reform in most of Africa, off
the streets and in their offices.

In this way, variations in the economic institutions of the two
countries helped account for the pattern of regime change. The rules of
participation in the Franc Zone meant that the urgency of
liberalization, or of some other method of responding to popular
discontent and channeling conflict, was initially higher in COte
d’Ivoire than in Kenya. The design of monetary policy institutions
distributed the costs of economic crisis differently and altered the
political landscapes of the two countries, as a result.

Economic institutions heightened pressures for political reform in
C6te d’Ivoire, compared to Kenya, in another sense as well. Modeled
explicitly on the arrangements sought by French farmers in the 1940s
and 1950s, a stabilization scheme had long protected the country’s
planters from fluctuations in the world prices of coffee and cocoa-
commodities produced in abundance in the forest zone. The political


Leadership Styles and Political Liberalization

preference for central absorption of market risk, a legacy of the colonial
period, meant that farmers received relatively stable real prices for
their crops. As long as world prices did not drop persistently and as long
as the government had the cash reserves to pay farmers a subsidy when
world market prices dropped, this system could function. During the
late 1980s, the prices of cocoa and coffee tumbled precipitously, how-
ever, and the stablization fund was no longer able to pay farmers the
guaranteed price (or more properly, reimburse traders for paying farm-
ers the official farm-gate producer price). In 1989 the government was
forced to slash prices for the two crops in half. The sums farmers re-
ceived diminished even further thereafter. By 1990 the collapse of the
cocoa- and coffee-based “miracle ivoirien” had produced severe reduc-
tions of standards of living in the countryside, increasing demands of
rural relatives upon urban kin, and a clear sense of dissatisfaction with

By contrast, in Kenya the government had opted for a system that
had long passed a higher portion of market risk onto the farmers. That
meant that when world prices for agricultural exports fell, farmers and
the state divided the risk. In this way, producer prices and world mar-
ket prices bore a stronger relationship to one another, and farmers re-
ceived the price signals necessary to help them allocate resources more
efficiently among alternatives. The collapse in world market prices
was felt not as a sudden reduction in standards of living but as a gradual
change, to which farmers had several years to make adjustments.
Although dissatisfaction with government had grown in Kenya during
the 1980s, there was no such sudden event as the slashing of cocoa and
coffee prices to crystallize concern and provide a focal point for ac-
tivism. Instead, political change in Kenya long remained the preserve
of lawyers, clergy and politicians who suffered the government’s dis-
favor. The popular well-spring of support for change materialized

The institutional contexts of the two countries varied in a third
respect too: in the way parallel associations were organized and in the
amount of autonomy these groups had from the government. In COte
d’Ivoire, before 1990, different producers’ groups or trades unions were
organized through centralized syndicates, along corporatist lines. That
is, interests were structured by single, officially recognized unions, each
with a monopoly over the function of representation within its sphere,
often with government subsidy or sponsorship. The syndicates were part
of a centralized system of peak associations and they met with decision
makers at intervals and in forums set aside for that purpose. In most
cases, the syndicates were tied to the ruling party.

This structure had two main implications for pressure for politi-
cal reform in C~te d’Ivoire. First, it meant that Houphouet encountered
stronger pressure for change in 1989 than did his counterparts in most of



Anglophone Africa, save Zambia, where the country’s mining base had
produced a strong, centralized union structure too. When followers broke
with their old leaders in these associations, the centralized, corpo-
ratist structure lowered the cost of collective action to members. It per-
mitted rapid negotiation among members and helped militants monitor
contributions of members to lobbying efforts, thereby reducing free-
riding problems. It facilitated rapid communication. In short, it made
general strikes easier to carry out.

That is exactly what happened in COte d’Ivoire. If Abidjan’s
elites had greater personal cause for concern, if their pocketbooks were
affected more severely in the 1980s than those of their Kenyan counter-
parts, they could also organize more quickly and effectively. In rapid
succession, teachers, doctors, pharmacists, dentists, vets and bank
workers engaged in devastating strikes between February and mid-
April. Even gas station managers threatened action. New, independent
syndicates took root-usually breakaway sections of the official asso-
ciations the government had tolerated since independence. Some, in-
cluding those representing university professors, bus drivers, govern-
ment health-care workers, the government press and computer pro-
grammers, formed an autonomous federation, the Federation des
Syndicats Autonomes de COte d’Ivoire.

Ironically, it also lowered the costs of negotiating with different
parts of Ivoirian society and reduced the perceived risks that tolerat-
ing parallel associations would make COte d’Ivoire ungovernable. Even
after the dominance of PDCI-linked peak …