Economic Liberalization and Domestic Instability

Wiesehomeier, Nina

Economic Liberalization and Domestic Instability

A Comparative Study

von Nina Wiesehomeier (Autor:in)

Politik - Internationale Politik - Region: Afrika

Zusammenfassung

Inhaltsangabe:Abstract:
During the nineties globalization has become one of the most disputed issues in the social sciences and a lively debate about its possible detrimental or beneficial consequences is going on. Among economists there exists the general consensus that global economic integration enhances welfare through market allocations that are assumed to be efficient. The majority of papers hereby concentrate on explaining what the causes for the increasing economic liberalization are, the feasibility of these reforms, limiting the debate to the question of correct sequencing of the distinct measures, or to the causes of either their success or failure. Another phenomena of the last decade has been the increasing amount of armed intrastate conflicts around the world which reached its peak in the early 1990s. Although economic aspects gained in interest as roots of internal war, economic liberalization per se was not considered as a causal factor.
This study therefore examines the question of whether economic liberalization is likely to fan the flames of domestic violent conflict, thereby distinguishing between long-term and short-term consequences of this process. I assume that in the long-term economic liberalization has a pacifying effect as the abolition of trade barriers fosters export-led growth. On the contrary, in the short-term, after an economic reform has been introduced, distributional effects will have a major impact on society, causing winners and losers. This situation is likely to trigger domestic instability and violent conflict if the winners are not able or willing to compensate the losers for their economic losses. To explain cleavages arising in society, this study adds an institutional perspective to Rogowskis model based on the Stolper-Samuelson Theorem and highlights it with an exemplary application with the Rubinstein Model. I illustrate the propositions derived from this model by presenting the case of Guinea-Bissau. Testing statistically the causal mechanism between economic liberalization and domestic conflict and intervening factors of political, sociological, and economical kinds, it furthermore engages in comparative work in a dual sense: the study moves across different levels of analysis and compares the three statistical tools ordinary logit, random effects logit and general estimating equation. It first starts with a sample comprising 90 developing countries for the time period of 1978 to 1998 and focuses […]

Leseprobe

Inhaltsverzeichnis

ID 9379

Wiesehomeier, Nina:

Economic Liberalization and Domestic Instability - A Comparative Study

Druck Diplomica GmbH, Hamburg, 2006

Zugl.: Universität Konstanz, Diplomarbeit, 2003

Dieses Werk ist urheberrechtlich geschützt. Die dadurch begründeten Rechte,

insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von

Abbildungen und Tabellen, der Funksendung, der Mikroverfilmung oder der

Vervielfältigung auf anderen Wegen und der Speicherung in Datenverarbeitungsanlagen,

bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Eine Vervielfältigung

dieses Werkes oder von Teilen dieses Werkes ist auch im Einzelfall nur in den Grenzen

der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik

Deutschland in der jeweils geltenden Fassung zulässig. Sie ist grundsätzlich

vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des

Urheberrechtes.

Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in

diesem Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme,

dass solche Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei

zu betrachten wären und daher von jedermann benutzt werden dürften.

Die Informationen in diesem Werk wurden mit Sorgfalt erarbeitet. Dennoch können

Fehler nicht vollständig ausgeschlossen werden, und die Diplomarbeiten Agentur, die

Autoren oder Übersetzer übernehmen keine juristische Verantwortung oder irgendeine

Haftung für evtl. verbliebene fehlerhafte Angaben und deren Folgen.

Diplomica GmbH

http://www.diplom.de, Hamburg 2006

Printed in Germany

Table of Contents

Chapter 1: Introduction ... 2

Chapter 2: Theoretical Implications of Economic Liberalization and Instability ... 9

2.1. A Review of Relevant Economic and Conflict Literature ... 10

2.2. A Panorama of the State Structure in Developing Countries... 11

2.3. The Long-Term Consequences of Economic Liberalization ... 14

2.4. The Distributional Consequences in the Short-Run ... 16

2.4.1. A Market Perspective ... 16

2.4.2. An Institutional Perspective ... 17

2.4.3. Conflicting Interests ... 20

2.5. A Comprehensive Argument and its Exemplary Application with the Rubinstein

Model... 21

2.6. Intervening Factors... 26

Chapter 3: Methodology... 33

3.1. Data and Operationalization... 33

3.2. Discussion of Statistical Models ... 39

Chapter 4: Quantitative Analyses... 44

4.1. Descriptive Statistics ... 44

4.2. Regression Analyses... 50

4.2.1. The Full Sample ... 51

4.2.2. The African Sub-Sample ... 57

4.3. Discussion ... 64

4.4. Regression Diagnostics ... 68

Chapter 5: The Case of Guinea-Bissau ... 71

5.1. The Course of Events ... 71

5.2. Analyzing the Events... 76

Chapter 6: Conclusion ... 79

Appendix ... 81

A.1. Correlation Matrix for the Ordinary Logit Models ... 82

1.1. Full Sample... 82

1.2. African Sub-Sample ... 85

A.2. Correlations for Development, Urban Population and Secondary School Enrolment88

A.3. Likelihood Ratio Tests for Splines in the Full Sample... 88

A.4. Wald Tests performed after Regression for the Full Sample ... 88

A.5. Likelihood Ratio Tests for Splines in the African Sub-Sample ... 90

A.6. Wald Tests performed after Regression for the African Sub-Sample ... 91

A.7. Regression Diagnostic ... 93

7.1. ROC Curves for Ordinary Logit Models... 93

7.2. Leverage and Discrepancy ... 94

7.3. Scatterplots of Pearson Residuals against Predicted Probabilities... 96

References ... 99

List of Figures and Tables

Figure 1: Literature on Globalization... 4

Figure 2: Trade Flows and Trade Policy, 1973-1995... 5

Figure 3: Mean Trade Openness ... 45

Figure 4: Incidences of Internal Conflict ... 47

Table 1: Description of the Full Sample ... 38

Table 2: Description of the African Sub-Sample ... 38

Table 3: Comparative Mean Growth Rates... 44

Table 4: Internal Conflict and Trade Openness... 48

Table 5: Internal Conflict and the Opening Process... 49

Table 6: Expected Influences of the Independent Variables on Conflict... 51

Table 7: Regression Analyses for the Full Sample ... 52

Table 8: Regression Analyses for the Full Sample with Secondary School Enrolment ... 56

Table 9: Regression Analyses for the African Sub-Sample ... 58

Table 10: Regression Analyses for the African Sub-Sample with Secondary School

Enrolment ... 62

Table 11: Synopsis of the Results ... 66

Table 12: Regression Analyses Excluding Influential Cases... 69

Chapter 1:Introduction

Chapter 1: Introduction

Globalization is on everyone's lips. The German parliament was the first to set up a

commission of enquiry dealing with questions related to globalization which recently

published its final report (Deutscher Bundestag 2002). During the nineties globalization has

become one of the most disputed issues in the social sciences and a lively debate about its

possible detrimental or beneficial consequences is going on. In his article Guillén

summarizes the debates about the main fundamental questions and notes that the "most

bewildering feature of the literature [...] is [...] the remarkable diversity of authors that have

contributed to it (Guillén 2001: 240). Figure 1 shows the considerable increase in literature

on globalization.

Figure 1: Literature on Globalization

200

400

600

800

1000

1200

1980

1985

1990

1995

1998

Sociological Abstracts

Economic Literature

Politics & Int. Relations

Historical Abstracts

Anthropoligical Abstracts

Books in Print

Source: Author's presentation using data from Guillén 2001. Literature refers to articles or books with the

words "global" or "globalization" in the title, subject heading or abstract.

Whereas historians and even more so anthropologists have just recently begun to deal

with the topic, scholars of sociology, economy and international relations yet at the

beginning of the 1980s have been aware of this phenomena. Globalization has a myriad of

connotations (Guillén 2001). In my paper I will merely deal with the economic dimension of

this process and leave cultural concerns like the emergence of cultural homogeneity, with

which sociologists tend to deal, aside.

Chapter 1:Introduction

The increase in cross-border economic exchanges shown in Figure 2 has been

facilitated through the reduction of transport costs, communication costs and the consequent

abolition of barriers of political as well as economic kind (Frieden and Rogowski 1997). It

becomes obvious that the crucial development in reduction of trade tariffs took place at the

end of the 1980s, which is the time when most structural adjustment programs led by the

international financial institutions - the Bretton Woods Institutions - were either

implemented or began to take effect.

Figure 2: Trade Flows and Trade Policy, 1973-1995

ource: Garrett, G. (2000). The Causes of Globalization. http://www.yale.edu/leitner/pdf/2000-02.pdf.

mong scholars of the economy in general there exists the consensus that global economic

integration contributes to welfare enhancement through market allocations that are assumed

to be efficient (Rodrik 1994, Krugman/Obstfeld 2000). The majority of papers hereby

concentrate on explaining what causes the increasing economic liberalization, the feasibility

of these reforms, limiting the debate to the question of correct sequencing of the distinct

measures, or to the causes of either their success or failure (for instance, Frey and

Eichenberger 1994; Geddes 1994; Grindle, 2001; Rodrik 1994 and 2002; Stern 1991; World

Chapter 1:Introduction

Bank 1989; World Bank 2000). Although concerns that states may lack the adequate

structures to deal with such a process exist as well (Rodrik 1997), the possibility that this

process might lead to domestic instability has been neglected so far.

Another phenomena puzzling scholars had been the increasing amount of armed

intrastate conflicts around the world which reached its peak in the early 1990s (Gleditsch et

al. 2002). After the end of the cold war, interstate war vanished more and more from the

popular research agendas and attention was directed towards internal conflicts. Whereas

previously scholars concentrated their explanations for internal wars mainly on deprivation

theory, thereby modeling social cleavages, or on political structures offering the opportunity

for mobilizing rebellious upheavals, economic aspects gained in interest as well, but

economic liberalization per se was not considered as a causal factor.

I will therefore bridge the gap between these research agendas and draw the attention

to the question of whether economic liberalization is likely to fan the flames of domestic

violent conflict. Trade is seen as mutually beneficial for the trading partners and fosters

export-led growth. Theoretically, free trade could lessen the likelihood of violent conflict in

two different ways. On the one hand, foreign trade enhances the welfare of society, thus

reducing economic discontents of the population. On the other hand Kant (1795 [1987])

claimed that trading democracies do not engage in war with each other, as war interrupts

trade flows, thus being too costly for both the state and its citizens. But these benefits will

only take effect in the long-term, leaving a risk of instability in the short run. After an

economic reform has been introduced, distributional effects will have a major impact on

society, causing winners and losers. This situation is conflict prone if the winners are not

able or willing to compensate the losers for their economic losses.

Yet in 1989 Rogowski presented a model to explain cleavages arising in society

through the implementation of economic liberalization, which have the potential to increase

domestic instability. For his explanation he draws on the Stolper-Samuelson Theorem that

states that trade will benefit the abundant factor in each nation and harm the scarce factor as

trade in goods compensates for national scarcities in factor supplies. But Rogowski ignores

institutional arrangements and stays within his market perspectives.

So in addition to the examination of the causal relationship between economic

liberalization and domestic conflict, I will also bridge the micro-macro gap. I assume that

institutional arrangements such as the protection of property rights, a competitive electoral

system, the rule of law, and representative political institutions are important for both the

Chapter 1:Introduction

sustainability of economic reform and the prevention of violent conflict. Dealing with

developing countries this aspect is even of special importance as the majority of those

countries lacks such institutions completely or are inconsistent political systems which can

be easily controlled by elite groups having personal interests at stake. A weak institutional

state structure offers rent seeking groups with organizational advantages the possibility for

exploiting the wealth of a nation and distribute it among supporters. A reform like economic

liberalization is likely to alter and change institutions, threatening such a system. I therefore

will present various possible scenarios for the interaction between economic liberalization

and domestic conflict.

Obviously, the struggles about the distribution of the costs and the limitation of the

access to the country's wealth made possible by the lack of regulatory institutions will be

characterized by bargain processes between the contending groups. The Rubinstein Model

(Rubinstein 1982) is helpful to model such non-cooperative bargaining processes and to give

a better idea of the interaction of violence and bargaining. As the focus of my paper lies on

the statistical analysis, I will only give a succinct description of the Rubinstein bargaining

model. This model would be appropriate to investigate such bargaining processes in detail

when concentrating on a thorough case study.

Testing statistically the causal mechanism between economic liberalization and

domestic conflict and intervening factors, I will engage in comparative work in a dual sense:

I will move across different levels of analysis and compare the three statistical models

ordinary logit, random effects logit and general estimating equation. I will start with a

sample comprising 90 developing countries covering the period of 1978 to 1998 and will

focus afterwards on African countries for the same time period. The African continent has

puzzled scholars for its outlier position as it displays considerably poor growth and recently

a relatively high conflict incidence. I will also include intervening factors of political,

sociological, and economical kinds such as, for instance, regime type, ethno-linguistic

fractionalization, growth and development, and population size. Presenting the case of

Guinea-Bissau will finally give me the possibility to examine the propositions derived from

Rubinstein Model.

Although some of the control variables do not perform as predicted and even do so in

a statistically significant way, the outcomes of the statistical analyses in general do support

my question of interest and do show the expected relationship between economic

liberalization and domestic conflict. Furthermore, the results do not differ essentially across

Chapter 1:Introduction

the models presented and the relative effects maintain stable, which strengthens the

confidence in the postulated relationship.

My paper is organized as follows: in the next chapter I will lay the theoretical basis

and explain how economic liberalization may lead to domestic instability and violent

conflict. Moreover I will present a succinct Rubinstein Model, discuss the intervening factors

and derive testable hypothesis. Chapter 3 is dedicated to the methodology. After having

presented and described the data set I will specify the three statistical models. In Chapter 4,

first of all the data set will be expounded in terms of descriptive statistics which gives me the

opportunity to give full particulars about the outlier position of the African continent.

Afterwards, the regression analyses will be carried out separately for the full sample and the

sub-sample of African countries and the results discussed. The regression diagnostics

conclude the statistical analyses. Chapter 5 will incorporate the propositions derived from

the Rubinstein Model and highlight them through the case of Guinea-Bissau. The last

chapter gives the conclusion of my investigation.

Chapter 2: Theoretical Implications of Economic Liberalization

Chapter 2: Theoretical Implications of Economic Liberalization

and Instability

The purpose of this paper is to establish a causal relationship between economic

liberalization

and instability

. Thus far, the literature on trade openness focused on the

reasons why an increasing number of states liberalize or liberalized their economies, the

feasibility of those reforms and their subsequent economic effects. Those scholars dealing

with internal conflict mainly concentrated on deprivation theory, thereby modeling social

cleavages, or on political structures offering the opportunity for mobilizing rebellious

upheavals. Recently, economic aspects have also been taken into account, but generally as

resources providing lootable income to rebels.

This paper intends to bridge the gap between these research agendas. In this chapter,

I will outline the theoretical basis for the question of interest, namely whether the prevalence

of trade openness is causing conflict.

In the following sections, I will first

review the main strands of the existing literature

in the economic and the conflict research area. Then I will describe the prevailing state

structure in developing countries in terms of economic, political, and societal features, since

this is an important aspect of understanding the impact and consequences a trade reform

might have. Afterwards I will clarify briefly the long-term effects of economic liberalization.

The short-term consequences and their relation to conflict will be specified more precisely,

above all the connection between trade, conflict and institutions. Subsequently, I will present

a succinct bargaining model for domestic conflict and put forward some propositions that

will guide the short case study about Guinea-Bissau. Consequently, I present some factors

which intervene in the postulated association between trade openness and domestic conflict,

and thus have to be taken into account. While presenting the theoretical foundations, in this

chapter I will also derive my hypothesis that will be examined with the statistical tool of

regression analysis.

I do not discriminate between the terms economic liberalization and trade openness and use them

interchangeable. Both terms refer to a situation or process in which the necessary changes in institutions

guaranteeing free trade, like the abolition of tariffs, quotas or import/export restrictions, are being made. I thus

neglect movements of services and capital.

Instability refers to a situation threatening the political system. In the following I will use internal conflict,

domestic conflict and violence interchangeable.

Chapter 2: Theoretical Implications of Economic Liberalization

2.1. A Review of Relevant Economic and Conflict Literature

In general, among international economic organizations and scholars of economic

theory, a consensus exists that global economic integration contributes to welfare

enhancement through market allocations that are assumed to be efficient (Rodrik 1994,

Krugman/Obstfeld 2000). Although it is equally widely accepted that trade policy is

redistributive in nature and that the gains and costs of trade liberalization may be very

unevenly distributed, thus causing winners and losers in society, little effort has been made

in the literature of international political economy to further investigate the possibility that

this distributional effect might be a source of violent domestic conflicts. Most papers

concentrate on explaining the phenomenon that an increasing number of developing

countries liberalized their economies during the last two decades, a trend which Rodrik

denominated as "the rush to free trade" (Rodrik 1994: 61). Likewise, when dealing with the

implementation of reform processes, questions discussed refer to the feasibility of these

reforms, limiting the debate to the question of correct sequencing of the distinct measures

(for instance, Frey and Eichenberger 1994; Stern 1991; World Bank 1989; World Bank

2000), or to the causes of either their success or failure (for instance Geddes 1994; Grindle,

2001; Rodrik 2002).

Two areas can be distinguished in conflict research: some scholars concentrate on

interstate conflicts, others focus on internal conflicts, mainly civil war. In the research area

of interstate war a vast literature exists on the liberal peace thesis, which Beck, Katz and

Tucker (1998) subdivide into a political and an economic element. The political component

deals with the proposition that democracies do not fight each other, or at least are less likely

to engage in military conflict with one another. The latter mentioned economic part, is

concerned with the statement that trade has a pacifying effect on the likelihood of interstate

conflict as trading partners do not engage in war with one another. Recently, Hegre et al.

(2001) applied this research agenda to the field of civil war, posing the question of whether

the observed increasing democratization during the last decade will lead to a democratic civil

peace.

But by and large, scholarship investigating the causes of civil war used to focus on

social cleavages, and used to work either in the tradition of the deprivation theory (Gurr

1970) or modeled violent conflict within the political opportunity framework (Tilly 1978). In

the former mentioned theory, the concept of relative deprivation plays an important role. The

Chapter 2: Theoretical Implications of Economic Liberalization

degree of felt discrimination does not correspond with an objective discrimination, but

depends on a comparison of the actual situation with the expectation of an individual or

collective actor on the basis of a reference group. Habitually these discontents experienced

have economic sources. These studies frequently examine topics like nationalism, ideology,

ethnicity, or conflicts as a consequence of environmental degradation causing lateral

pressure (for instance, Fearon and Laitin 2000; Laitin 1995; Ron 2001; Homer-Dixon 1999).

This is believed to lead to grievance-motivated civil wars and people engage in violence in

order to seek justice. The latter mentioned, also rather traditional framework treats political

structures and processes as determinants of political violence. Structures are more likely to

change over time, thus providing opportunities to mobilize rebels and oppose state policies

(Tarrow 1989). In this approach, grievances are believed to stay constant and do not play a

significant role.

On the other hand, it is possible to witness an increasing number of investigations in

the research agenda of "The political economy of civil war" linking the causes of civil war to

more selfish behavior. Rebels pursue the aim of making the most of the opportunity and

organize violence to extract personal rents. By emphasizing the importance of economic

factors in civil wars, these studies turn the attention to the argument that greed instead of

grievance is the motive behind violence (for instance, Addison et al. 2001; Berdal and

Malone 2000; Collier and Hoeffler 2001). Although these studies consider economic aspects

like primary commodity dependence, trade liberalization policy per se is not treated

explicitly as having an impact on society that possibly leads to conflict and violence.

2.2. A Panorama of the State Structure in Developing Countries

The aim of this section is not to explain why governments in the developing countries

adopted the policy choices that led to the economic environment we are faced with today,

but to offer a succinct description of the prevailing structures that any reform strategy will

have to cope with when implementing economic liberalization. Some of these features fit

more for Sub-Saharan African countries than for Asian or Latin American ones, but the

general picture is the same.

The strategy that developing countries adopted for industrialization after having

gained independence was to build up import substituting industries (hereinafter ISI). When

doing so, they generally neglected some fundamental points that proponents of this measure

Chapter 2: Theoretical Implications of Economic Liberalization

emphasize (Lofchie 1994). Instead of choosing a small number of industries that would use

local inputs, the number and range of these firms were immense and they often relied on

imported raw materials. Moreover, these enterprises were predominantly state-owned

resulting in deficit budgets at the national level at the end of the 1970s when the failure of

this strategy became evident.

Closely related to the promotion of ISI is the topic of protectionism. Protection

implies the use of tariffs, quantitative restrictions, and import licenses, which have the effect

of a tax on imports. All these measures raise the price of the importable good for consumers.

They aim at protecting the incomes of the import sector from competition and imply a

welfare loss for society as a whole. The majority of developing countries adopted a mix of

trade limitations which could be used quite arbitrarily, thus giving the power of personal

enrichment to the bureaucratic elites responsible for the allocation of, for instance, licenses

or contingents. Furthermore, tariffs and related measures are important to the government for

fiscal and budgeting reasons. In general, the tax capacity in developing countries is very

weak and states depend heavily on these sources of revenue to perform their tasks.

distorting the macro-economy the narrow governing elites generate short-term benefits

(Humphreys and Bates 2002). So in addition to the fiscal problem involved, the abolition of

such a system will also encounter resistance from those who benefited from it. The most

cited argument for protection, the reference to infant industries, implies just temporary

protection. But if these measures are not tied to conditions for the benefiting enterprises, in

the face of the problems mentioned, this argument is not easy to maintain.

Two more features stem from pursuing an ISI strategy. On the one hand, fixed

exchange rates, that led to overvaluation of currencies. The effect of this measure is the same

as subsidizing the ISI, as the imports of the required raw materials are cheapened. On the

other hand, industrial development was financed with earnings from the export-oriented

agriculture, a sector controlled by a state monopsony (i.e. a lot of suppliers face only one

buyer). Via marketing boards, taxes were imposed on producers, which on the one hand

served as government revenue and on the other hand provided cheap food for the urban

population. These two measures created a heavy anti-export bias.

In summary, this manner of implementing ISI had two consequences: Firstly, it led to

profound problems in the balance of payment and subsequently to a deep crisis in the 1980s

(Rodrik 1996). Secondly, it favored wage-earning groups and gave rise to a special structure

For instance, in 1986 Argentina 13.31% of total government revenue, and in Ghana 40.9% came from these

tariffs (Markusen 1995).

Chapter 2: Theoretical Implications of Economic Liberalization

of political economy in the developing societies that proved to be self-enforcing and very

persistent (Lofchie 1994).

Krueger (1974) refers to this as a "rent-seeking society". The underlying idea is, that

"when a society imposes a limitation on trade, as in the form of quantitative restrictions on

imports, it is creating artificial scarcity. The opportunity to ameliorate that scarcity, as in the

form of an import license, is a valuable commodity." (Lofchie 1994: 174). These artificially

generated benefits have the form of rents and are created when the state restricts market

processes. The resulting activity of rent-seeking is unproductive for the society as a whole

and involves economic and welfare losses, costs that are also denominated as "deadweight

losses". This activity does not increase the value or the quantity of goods or service, but only

engages in the fight for privileges, such as licenses, and can be legal (e.g. lobbying) or illegal

(e.g. bribery). In the environment described above - high government engagement in the

economy - the illegal variant is dominant.

Moreover, rent-seeking prevails in countries with a large public sector, an element

which is quite characteristic for developing countries (Hutchcroft 1997).

Inside the public

sector, corruption is the order of the day. In some countries, this situation is aggravated by

clientelism and neopatrimonial structures, sometimes along ethnic lines. This system of

loyalty and dependence is a form of social organization and serves to maintain the power and

to secure the political survival of the president, the dictator or the ruling clique (Bratton and

Van de Walle 1994). The incumbent(s) distribute(s) jobs to loyal followers and confidents,

for which state-owned enterprises and the bureaucracy are welcomed instruments, and gives

these privileged supporters access to public resources. This leads to inefficiencies, as job

candidates are not chosen for their abilities and capabilities, but for reasons of ethnic ties or

kinship ties. Additionally, very often this system redistributes services, like education, within

and among these groups, which creates solidarity links between them (Azam 2001a).

The last characteristic of developing countries is the so-called urban-land bias, which

simply describes the fact that the most potent and demanding groups are to be found in the

major cities. The ISI strategy gave rise to this development which gets reinforced by it

(Lofchie 1994). Urban-land bias and the rent-seeking activities of political and bureaucratic

elites explain the persistence of economic policy. Remember that this policy, by a variety of

measures, favored wage-earning groups, which will be found in urban agglomerations.

Geddes (1994: 204) mentions that " in Senegal, 45.6 % of those employed outside the agriculture were

employed by the public sector; in Zambia, the figure was 81% in 1980; in Argentina, fairly typical for Latin

America, 22.7% in 1981."

Chapter 2: Theoretical Implications of Economic Liberalization

Normally, the government is highly centralized and in contrast to the rural population, which

on average is dispersed and far off from the decision-making center, the urban population is

able to influence policy choices through informal relationships. If the leaders of urban

interest fail in their "quiet lobbying" (Lofchie 1994), they can threaten the incumbent(s) with

mass protest, strikes or more violent forms of expressing disapproval. So the economic

liberalization implies a policy that may not be politically desirable.

This short depiction gives an idea of how economic liberalization in developing

countries and conflict and violence may interact. Now I will go on to give details on the

consequences this reform may have and will connect these findings to the probability of

violent conflict.

2.3. The Long-term Consequences of Economic Liberalization

In general, trade is seen as a positive sum game, which is mutually beneficial for the trading

partners and increases growth rates for the domestic economy. The basic assumption is that

the best products and the best production methods will catch on. Free price fixing through

supply and demand will lead to the optimal balance between producer and consumer

interests. Countries exploit their comparative advantages in free trade, thus being enabled to

produce and consume more goods than without trade through increasing specialization.

Comparative advantages together with trade are believed to enhance the welfare of the

nations. Dynamic gains from trade, like the acquisition of new knowledge or changes in

institutions for instance, have even stronger long-term benefits and effects than static gains

which are inherent in the theory of comparative advantages (Thirlwall 2003).

This export-led growth has been subject of various studies. Sachs and Warner (1995)

find that open countries grow faster after having liberalized their economy and are also able

to more easily prevent crisis of balance of payment. Harrison (1995) uses different measures

of economic liberalization and finds a general positive association between growth and

openness. And Yusuf and Stiglitz (2001: 229) summarize in their review of present and

coming development issues that "...more open economies showed greater resilience in the

face of shocks, attracted larger flows of FDI, enhanced their growth rates through the link

between exports and domestic investment, and realized technological progress through

export competition, as well as increased access to more technologically sophisticated

imports".

Chapter 2: Theoretical Implications of Economic Liberalization

Theoretically, free trade could lessen the likelihood of violent conflict in two

different ways. On the one hand, foreign trade enhances the welfare of society, thus reducing

influencing factors, which according to deprivation theory, above all through the concept of

relative deprivation outlined before (see section 2.1), could lead to instability. The Stolper-

Samuelson Theorem shows that the abundant factor in each country will gain from trade,

resulting in a higher demand for that factor. Since in developing countries this typically will

be the greatest part of society, namely unskilled work, it is assumed that the population as a

whole will be better off. But evidence about the distribution nature of free trade in terms of

the equality of income is not coherent. Whereas Eusufzai (1996) finds that enhanced income

per capita in open economies is overwhelmingly connected with an increasing income

inequality and worse Human Development Index rankings, Dollar and Kray (2000)

emphasize that growth benefits the poor as much as the rich and relative distribution,

measured with the Gini ratio, stays the same. Edwards (1997) comes to the same conclusion,

but also hints at measurement problems and at the importance of studies about trade

experience of specific countries. If foreign trade really contributes to an equal distribution of

income, it could greatly reduce the likelihood of conflict.

On the other hand, it is possible to rely on the argument of the liberal peace which

goes back to Kant (1795 [1987]). He claimed that democratic states that trade do not engage

in war with each other, as war interrupts trade flows, thus being too costly for both the state

and its citizens. In addition, trade strengthens liberal norms which disapprove violent

conflict. For interstate relations, this argument has been confirmed in myriad empirical

studies (for instance, Maoz and Russett 1993; Oneal and Russett 1997). The logic is

transferable from the interstate context to the topic of domestic conflicts. Likewise, domestic

instability hinders trade and reduces gains, even more, as both phenomena involve domestic

groups as actors. Collier (1999) estimates that GDP per capita declines at an annual rate of

2.2 percent during civil wars compared to the counterfactual. In his study on civil wars and

poverty, Elbadawi (1999) concludes that a higher incidence of civil war leads to lower

growth. So if the actors realize the losses and take them into account to their cost-benefit-

analysis, free trade can have a pacifying effect on domestic conflicts not only through its

institutional settings and norms, that normally accompany an economic liberalization.

Summarizing these arguments, it seems plausible that opened countries benefit from

foreign trade in the long run, showing higher levels of development, growth and stability. So

I expect a higher risk for conflict, also violent conflict, in the short run and will now outline

Chapter 2: Theoretical Implications of Economic Liberalization

the reasons why.

2.4. The Distributional Consequences in the Short Run

2.4.1. A Market Perspective

In the short run, the distributional effects of liberalization will prevail. It is obvious that these

effects will not vaporize in the end, but the impact is most severe in the short-term. In the

long-term the process of economic liberalization is believed to be concluded and the positive

outcomes outweigh the costs via direct or indirect compensation of the losers and a general

adjustment to the new situation. In relation with Compensation Theory, Hicks (1939)

assumes a "balance of interests" in the long run. But it is undisputed that any policy altering

economic structures within a country has the potential to raise cleavages or to make existing

ones more salient.

Rogowski (1989) presents three general assumptions about the relation between

polity and economic changes. Firstly, potential winners of a reform (or change) will try to

accelerate it whereas potential losers will try to retard or stop it. Secondly, winners will have

the benefit of an increase in wealth that will enable them to gain political influence. Thirdly,

as the interest in the reform increases, the resultant pressure on political entrepreneurs will

lead them to attempt to overcome collective action problems. Rogowski draws on the

Stolper-Samuelson Theorem to explain cleavages in society that are caused by economic

liberalization and possibly lead to domestic instability. The Stolper-Samuelson Theorem

states that trade will benefit the abundant factor in each nation and harm the scarce factor as

trade in goods compensates for national scarcities in factor supplies.

Rogowski's model is based on the three factors land, capital, and labor. A country's

endowment of these factors is defined by the land-labor ratio, assuming that no single

country can be rich in both land and labor. An advanced economy is automatically denoted

as capital abundant. Depending on the kind of economy, Rogowski either predicts from his

political model combined with the Stolper-Samuelson Theorem an urban-rural conflict or a

class conflict accompanying economic liberalization in developing countries. In the first

case, land is the abundant factor which would benefit. Capital and labor, normally based in

an urban environment, are scarce and would seek protection. In the latter case, the economy

is endowed in labor, and capital and land are scarce, thus leading to a coalition of

landowners and capitalists supporting protectionism. Midford (1993) criticizes this

Chapter 2: Theoretical Implications of Economic Liberalization

approach as incomplete as it cannot explain coalition building and cleavages in more

developed countries. However, he emphasizes that Rogowski's model is fruitful for

underdeveloped countries or for a historical perspective on today's advanced industrialized

economies.

Frieden (1991) offers a similar perspective as Rogowski, but draws his argument

from the Ricardo-Viner model, which implies cleavages along economic sectors and not

along factor endowments. The pressures to engage in or refrain from economic liberalization

will depend on the specificity and the relative competitiveness of the sector. Industries which

share the same characteristics are expected to cohere as industries in either lobbying for or

opposing changes in trade policies.

Frieden and Rogowski (1997) join their arguments, modeling the pressures for

change via exogenous alterations of relative prices like decrease in costs (for instance

transportation costs) or increase of rewards of such change. In addition to the two concepts

explained above, they introduce the perspective of economies of scale, where conflicts will

be expected between firms and industries experienced in international trade and those that

need to make adjustments in order to be competitive in the global market. For developing

countries, this view seems less suitable. However, all three approaches agree that economic

liberalization has the potential of leading to domestic conflict.

2.4.2. An Institutional Perspective

Regulatory institutions, for example those protecting property rights, are generally

considered to be important to foster economic development (for instance Aron 2000; Greif

1992; Knack and Keefer 1997; Rodrik 2000a; Rodrik 2000b, Rodrik, Subramanian and

Trebbi 2002). Africa's poor economic performance is attributed to the weak institutional

arrangements that dominate the continent, of which more later (see section 4.1.). Olson

(1996) maintains that because of by and large weak institutions in developing states these

countries have not yet taken advantage of their full growth potential. Also, institutions play a

major role in conflict prevention and resolution as they influence the way in which

contending groups deal with conflict (democratic institutions for instance Benson and Kugler

1998, and Gates et al. 2001; for social capital see Varshney 2001). Organizing conflict

suffers from a collective action problem. People will only join upheavals and the like if the

Rogowski himself addresses various possible objections to his model and hints at the possibility that it maybe

has only historical relevance.

Chapter 2: Theoretical Implications of Economic Liberalization

gains outweigh the participation costs. Institutions can channel conflict and reduce such

gains noticeably, thus preventing violence. Easterly (2000) shows that high quality

institutions, e.g. rule of law, bureaucratic quality, can mitigate negative effects of ethnic

diversity and competition on growth, war, and genocides. So in general, scholars in

economic literature as well as conflict researchers agree that institutions are important

features of policy making, but it is difficult to identify the factors that may predispose a state

to develop the "right" institutions and overall, of what kind the right ones are.

A problem arises not only when attempting to put forward a definition of those

institutions, but also when trying to specify their concrete form there is a wide range of

institutions that matter and a wide range of institutional possibilities. Roughly it is possible

to split them up into economic, political, and social institutions, where the former relates to

market and the latter two non-market institutions. In a decade of research, the attention has

been directed to tradable and enforceable property rights, rules preventing fraud, anti-

competitive behavior and moral hazard, electoral systems, the concept of social capital

including norms and values fostering trust and social cooperation, and the rule of law. Often

good institutions are just equated with the democratic political system.

Just as confusing as the variety of classifications is the high number of quantitative

measures of institutions.

A shortcoming of these measures is their limited time-span, often

covering only one decade, and their insufficient geographical coverage, especially the

paucity of data points for developing countries. Moreover, the evaluation of the quality of

institutions, e.g. their performance, a feature that seems of particular interest and logically

affects the policy outcomes most, often relies only on subjective rankings (Aron 2000).

Rodrik (2000) emphasizes that it is important to allow and accept institutional variety and

their local derivatives and argues that institutions develop best on the local level via

experimentation and incorporation of local knowledge.

This makes it plausible to investigate the effects of institutions in a case study

framework where it is possible to provide the necessary depth that their analysis requires.

Apart from the democracy variables included in my data, I will therefore refrain from

including them explicitly in my data analysis.

Some studies hint at participatory and decentralized political systems, for instance Rodrik (2000) and Grindle

(2001), which of course refers to coherent democracies as suitable political systems.

Hall and Jones (1999) use the term "social infrastructure" which describes nicely the idea of horizontal social

networks.

Aron (2000) lists 28 different measures that have been used during the last years only in the empirical growth

literature. She subdivides them into the five categories "Institutional quality measures", "Social capital

measures", "Social characteristics", "Political characteristics" and "Political instability".

Chapter 2: Theoretical Implications of Economic Liberalization

But obviously, institutions matter, since regulatory measures like a trade reform, are

often the result of bargaining processes and therefore represent outcomes negotiated in

institutional arrangements. By and large, a reform process such as economic liberalization

consists of an interaction of need for reform, signaling for credibility, the costs of reform and

the impact of uncertainty on planning and implementation. Politicians, voters, and

bureaucrats interact through political institutions and have at stake different personal

interests with a different order of preference, such as staying in office, a certain public

policy, or career promotion. And of course, a reform such as economic liberalization is likely

to alter and change institutions and societal arrangements.

Moreover, institutions are important with regard to the aspect of compensation.

Rodrik (2000) stresses that institutions like the rule of law, representative political

institutions, free elections, social insurance systems, or institutionalized representation of

minorities offer incentives of cooperation as they constrain predatory behavior of the

winners and provide security to the losers. In this way, such institutions make a commitment

to compensation more credible and mitigate the impact of uncertainty.

But institutions, overall in developing countries are burdened with conflict proneness,

because "institutional arrangements of a society are often the outcome of strategic

distributive conflicts among different social groups, and inequality in the distribution of

power and resources can sometimes block the rearrangement of these institutions in ways

that would have been conducive to overall development" (Bardhan 2001: 284). Yet

Huntington (1968) noted that the process of modernization of political institutions is marked

by violence, a view which is perfectly adoptable to developing countries. And North (1990),

one of the first researchers providing an institutional framework, observed that institutions

do not serve the purpose of social efficiency, but rather the interests of those with the

bargaining power to set the rules of the game. Summarizing his view about the struggles of

the contending groups in connection with economic liberalization, Rogowski (1989: 20)

concludes his introductory chapter by stating that "[v]ictory or defeat depends, so far as I can

see, both on the relative size of the various groups and on those institutional and cultural

factors that this perspective [his model] so resolutely ignores."

In the following section I will briefly describe how conflict may arise. Whereas the

intention of the statistical analysis is to highlight the causal relationship between economic

liberalization and internal conflict, the case study on Guinea-Bissau will pay more attention

to the struggles between the opposing groups.

Chapter 2: Theoretical Implications of Economic Liberalization

2.4.3. Conflicting Interests

As was seen before, the political economy of developing countries is based in principle on

import-substituting industries, which are often run by the state administration, and lead to the

creation of a rent-seeking society (see section 2.2). Additionally, the importance of

employment in the public sector has to be taken into account, since in Africa we find an

inflated civil service that is often entangled with neopatrimonial structures, kin groups,

ethnical ties and corruption. Public sector employment often plays an important role in

(re)distributing resources from the urban elite to rural regions and, as shown in section 2.2. is

used to secure support. This fact concerns directly the organizational capacity of the different

groups affected by economic liberalization and their respective influence on policy

outcomes.

Remember that the Stolper-Samuelson Theorem predicts that trade liberalization

tends to benefit the abundant factor of an economy. So potential losers have an

organizational advantage, as losses tend to be concentrated and apparent, while benefits are

diffuse and not obvious for an individual. Furthermore, because of the existing organization

costs, we expect that the bigger the affected group the less probable becomes its

organization, in this case in favor of liberalization (Olson 1965). In addition to this well-

known collective action problem, Geddes (1994) identifies the problem of what she calls

"external vs. internal constituencies". She falls back on the principal-agent problem to point

out that not all groups in society have the same access to information and can monitor their

representatives in the same way. This asymmetry implies, above all, a disadvantage for the

voters while the political and bureaucratic elite can transfer their informational advantage

into an influential advantage.

Thus economic liberalization not only shakes such an economy severely, but it also

endangers the stability of the entire political system. This is so since reform steps threaten

the loyalty system and the elite's beneficial positions. On the one hand, the rent seeking

system, on which the stability of the regime relies, is at risk through the abolition of licenses,

monopoles, and tariffs. On the other hand, privatization of state led firms and downsizing of

the public service sectors are core elements of adjustment and restructuring programs and

will cause extensive layoffs of state employees.

In this case, the well-organized elite, being

a privileged group, is likely to oppose such reform steps beforehand to maintain the status

quo. This is especially the case, where those loyal to the incumbent leaders staff the

Uganda reduced its staff from 320 000 in 1989 to around 148 000 in 1997. In Argentina, in 1989 the number

of employees reached 1.9 million. 180 000 employees changed from the public to the private sector through

privatisation between 1990 and 1992 (Schlavo-Campo et al. 1997).

Chapter 2: Theoretical Implications of Economic Liberalization

bureaucracy (Geddes 1994) and impose high costs on the politicians, who are considered to

be rather support than welfare maximizers and are therefore likely to oppose change.

Another possibility for conflict arises after the reform has been implemented. Firstly,

as seen in Rogowski's model, the winners - the export-orientated industries - will receive an

increase in wealth. Politically the owners of those industries played a more secondary role

before the reform, which they will possibly try to change now that they have gained

economic importance. They therefore may challenge the incumbent(s). In a democracy this

would be feasible through elections. In a repressive regime this may lead to violent conflicts,

if the incumbent(s) is(are) not willing to change the political distribution of power through

further reforms. This highlights once more the importance of institutions. Secondly, it can be

assumed that the losers will organize protest against the measures implemented, if the

government does not abide by its promise of compensation.

This also might lead losers to

organize violence or join rebel groups in order to gain lootable income, which provides

short-term economic relief.

The theoretical arguments put forward so far can be summarized thus: I assume that opened

countries benefit from foreign trade in the long run, showing higher levels of development,

growth and stability. So I expect a higher risk for conflict, also violent one, in the short run.

(Core) Hypothesis 1: Economic liberalization reduces the likelihood for violence in the long

run, but increases it in the short run.

2.5. A Comprehensive Argument and its Exemplary Application with the

Rubinstein Model

Uniting what I have expounded so far, I can get to the heart of the outlined and present my

underlying argument: Economic liberalization will trigger conflict through the alteration of

the existing societal structure of (re)distributing urban rent-seeking groups, a structure that

constitutes a kind of social system that in modern states is provided by the state, if a

This apparently will include urban labour force as well, such as worker unions, as they will suffer from

adjustment measures in the short run. Moreover, Geddes (1994) observes that the working class failed to

monitor their representatives and thus were not able to make credible threats of punishment. This emphasizes

the points made about organizational capacities (due to different access to information, etc.) and the unequal

distribution of bargaining power that I will highlight in the Rubinstein Model.

Chapter 2: Theoretical Implications of Economic Liberalization

substitution of this system is disregarded and an institutional vacuum created.

An economic

reform should not underestimate the difficulties that may arise when this horizontal state

legitimacy is to be replaced with a vertical one and inter-group cooperation problems

involved.

The incompleteness of the replacement of such structures is reflected by what Gates

et al. (2001) call institutional inconsistency. They argue that "political institutions cannot be

mixed" (Gates et al. 2001: 2) as the lack of a self-enforcing mechanism will lead to

instability. Neither is authority sufficiently diffuse nor highly concentrated to provide

stability. Thus, consistent democracies and consistent autocracies represent the most stable

systems. Remember the prevailing state structure in developing countries described in

section 2.2. States lacking institutions like the protection of property rights, the rule of law,

or institutions for the representation of minorities or that are inconsistent political systems,

can be easily controlled by elite groups having personal interests at stake. A weak

institutional state structure offers rent seeking groups with organizational advantages the

possibility for exploiting the wealth of a nation and distributing it among supporters.

My argument is not invalidated by the hint at the fast growing Asian Tigers, often

cited as an example of the economic benefit that liberalizing dictatorships may have.

Overland et al. (2000) emphasize that dictatorships pursue a "rapid-growth-at-all-costs"

strategy that reflects the incumbent's interest of stabilizing his power as this strategy enables

him to distribute rents to his supporters. This accords with a view expressed in McGuire and

Olson (1996) of the "encompassing interest" the stationary bandit has in the economic

performance of his subordinates. Furthermore Overland and her colleagues argue that

normally dictatorships grow at a faster rate than optimal, and they do so on the expansion of

social costs. This will lead to problems, echoed in the findings of Hegre et al. (2001) that the

median life for consistent autocracies is still less/lower than for consistent democracies.

Obviously,

the

struggles

arising from economic liberalization will be primarily

constrained by economic and institutional factors that will define the rules of the bargaining

game and influence the subsequent outcomes that are supposed to be solutions to the

problem of the distribution of the costs economic liberalization is imposing on the society.

Additionally, it is possible that the threat or the actual use of violence will be used as a

strategic instrument during the bargaining process.

Rodrik (2000: 2) argues that in Latin America economic "reforms have paid too little attention to

mechanisms of social insurance and safety nets."

Chapter 2: Theoretical Implications of Economic Liberalization

The Rubinstein Model (Rubinstein 1982) is helpful to model such non-cooperative

bargaining processes and to give a better idea of the interaction of violence and bargaining.

As the focus of my paper lies on the statistical analysis, I will only give a succinct

description of the Rubinstein bargaining model. This model would be appropriate to

investigate such bargaining processes in detail when concentrating on a thorough case study.

The basic idea is that two players - opposing groups or individuals - can distribute or

divide an exogenously given pie

among them, if they are able to agree upon the shares. The

players alternate offers, so that the first bargaining round consists in one offer made by

player 1 and the corresponding response possibilities of accepting the offer or refusing it by

player 2. If player 2 rejects the partition, she will place the offer in the consecutive round, if

she accepts it, the game will end. Furthermore, as real life would suggest, bargaining

imposes costs on the negotiation partners. Additional to fixed costs, time represents a major

problem to the players. Parting from the idea that "time is money", a discount factor

diminishes the value of the pie with every round of bargaining. As timeframe goes to

infinity, the pie converges to zero. This is so, because individuals assign a certain payoff

today a higher value as to a future payoff, for example, tomorrow. This means that the utility

is higher the bigger the share of the pie and the shorter the bargaining process (Holler and

Illing 2000).

For the bargaining situation of the players, various characteristics are important:

Firstly, the initial endowments of power and resources are of special interest as they

affect the discount factor. Negotiators with a low discount factor will be more impatient

and keen on ending the game.

Secondly, placing the offer first is an advantage as the second player must wait to the

second round if she rejects the proposal and bears the consequence of a discounted

counteroffer (Morrow 1994). This gives more power to player 1 who on her part is able

to offer less to player 2 than the latter would demand in the second round.

Thirdly, there may exist situations where the threat of rejection is not credible to the

offering part as the utility of rejection will be lower than the utility of acceptance (Holler

and Illing 2000).

Fourthly, as the fixed costs arise in each bargaining round, at some time a round is

reached where the payoff turns negative (Holler and Illing 2000).

It is amusing to note that political scientists prefer to divide pies whereas economists seem to have a

preference for cakes!

In our case, the opposing parties negotiate over costs, so obviously the pie has a negative value. But the

underlying idea becomes clear. Recall that a lot of trade reforms take place when the country yet is in a crisis.

Details

Seiten
Erscheinungsform: Originalausgabe
Jahr: 2003
ISBN (eBook): 9783832493790
ISBN (Paperback): 9783838693798
Dateigröße: 1 MB
Sprache: Englisch
Institution / Hochschule: Universität Konstanz – Politik- und Verwaltungswissenschaft
Note: 1,3
Schlagworte: globalization civil africa heckscher-ohlin guinea-bissau

Autor

Nina Wiesehomeier (Autor:in)