Is fiscal decentralization in help for internal conflict risks mitigation? Sub-Saharan African countries focus

doi:10.21203/rs.3.rs-3607420/v1

Download PDF

Research Article

Is fiscal decentralization in help for internal conflict risks mitigation? Sub-Saharan African countries focus

https://doi.org/10.21203/rs.3.rs-3607420/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Decentralization has emerged as an important tool for conflict resolution around the world in recent years. This article examines the impact of fiscal decentralization on the risk of internal conflict in Sub-Saharan Africa, where more than two decades of experimentation with decentralization contrasts with the rise of multifaceted crises. Using a new dataset of 21 countries from 2002 to 2019, the results of several econometric methods show that fiscal decentralization has a statistically significant effect on internal conflicts risk in Sub-Saharan Africa. However, this effect is contingent on how subnational governments are funded. While policies that rely on intergovernmental transfers and expenditure decentralization increase the risk of internal conflict, policies that rely on financial empowerment of decentralized communities and resource decentralization reduce the risk of internal conflict. This means that increasing regional resources autonomy can lead to secessionist tendencies. Governments should increase intergovernmental transfers and local government spending in accordance with the principle of subsidiarity.

JEL classification : H71, H72, H77, Q34.

fiscal decentralization

intergovernmental transfer

internal conflicts

Sub-Saharan Africa

According to the 2020 report by researchers at the University of Hamburg in Germany on the causes of war in the world (AKUF 2020), Africa is the region most affected by internal conflicts with 12 of the 29 cases recorded worldwide. While sub-Saharan Africa records ten war situations (Ethiopia, Burundi, Cameroon, Mali, Mozambique, Nigeria, Somalia, South Sudan, Congo-Kinshasa, Central African Republic), North Africa records two (Libya, Egypt) (AKUF 2020). Most of these conflicts are between states, or between a state and another party such as one or more rebel groups, or between terrorist organizations. Several causes are at the origin of these conflicts and vary according to the country. In the Sub-Saharan African region, Mali, Burkina Faso, Niger, Nigeria, Chad, and Cameroon are affected by armed conflicts, which often have a communal source (AKUF 2020)¹. Other sources of conflict are political quarrels over the control of power (South Sudan, Burundi), the availability of firearms when there is already an armed conflict in the vicinity (case of Cameroon with Nigeria), among others. The multiplication of coups d'état in this part of the continent is a perfect illustration. In just two years (between 2021 and 2023), there have been six putsches in sub-Saharan Africa (Sudan, Chad, Guinea, Mali, Burkina Faso and Niger).

The concern of States today is to find solutions to these various conflicts in order to guarantee peace, the security of goods and individuals, the unity of the country and favourable economic conditions. These objectives, which are the responsibility of the State in crisis, also concern the States not affected. The decentralization policy that most of these countries have adopted since the end of the 1980s may offer a way forward. It has to be said that, over and above the problems of inefficiency of highly centralized governments in the management of resources and the provision of public goods and services (Bird and Vaillancourt 1998), the adoption of decentralization in Africa has been motivated above all by the need to maintain or guarantee national unity (World Bank 1999). If we define decentralization with Kee (2004) as the transfer from central to local government of specific functions with sufficient administrative and financial resources to fulfil these functions, we can find elements in the facts that explain the resurgence of crises. Indeed, decentralization processes in Africa are characterized by the weakness of local government financial resources and the inadequacy of financial arrangements with the central state (Yatta 2009).

Across Africa as a whole, local spending accounts for an average 9% of total public spending, compared with 60% in European countries. As with expenditure, very few cities in Africa are financially autonomous. The autonomy indicator suggests that the African average is very low (around 3%) and that the most decentralized countries in terms of the autonomy of their local finances are Ethiopia (11.4%), South Africa (9%), Zimbabwe (6.9%) and Zambia (4.8%) (Mbassi 2014). The lack of financial resources in sub-Saharan African states is the stumbling block to decentralization policies in most developing countries. This difficulty is all the more damaging in that it can be a source of conflict within a country. As evidence of this, in 2015 the Catalan government sent a new warning message to the Spanish executive about its lack of cash, which depends on regular transfers from the latter, since most taxes are collected by services based in Madrid (Sorens 2016). Furthermore, the widespread political instability in the Sub-Saharan African context would be another manifestation of this incomplete decentralization, as according to Frey & Luechinger (2004), decentralized countries are generally characterized by a high degree of political and administrative stability. However, this weakness can also be explained by the fear of possible secessionist tendencies, which increases with the financial autonomy of certain regions (Sorenz 2016).

There is an extensive literature on the relationship between decentralization and internal conflict. Despite the burgeoning literature, there is no consensus among political economists on the relationship between these two phenomena (Green 2008). Thus, according to several authors, fiscal decentralization increases government efficiency in the provision of public services and ensures social cohesion by curbing secessionist tendencies (Alesina and al., 1995; Oates 1972). But this view is not shared by all specialists in the field. Other authors argue that decentralization can increase the risk of secession (Sorens 2016; Bardhan 2002; Tanzi 2002 et autres). However, it is impossible to generalize the relationship between decentralization and conflict (Treisman, 2006), since fiscal decentralization plays a role in fiscal appeasement by the federal government (Sorenz 2016). This controversy in the literature calls for further investigation. Our study is part of this and is motivated on the one hand by the fact that most studies linking decentralization to conflict, do not explicitly investigate the effect of fiscal decentralization. On the other hand, this study is motivated by the fact that, in all likelihood, no study linking the two variables has yet used panel data for a sample of sub-Saharan African countries whose experimentation with decentralization over more than two decades contrasts with the rise of multifaceted crises. In addition, this study is innovative in that it uses a new database compared with the existing literature.

The objective of this paper is then to analyse the effect of fiscal decentralization on internal conflict risk in sub-Saharan Africa. By applying several estimation technics, on a panel of 21 countries over a period from 2002 to 2019, we obtain results that show that revenue and tax fiscal decentralization increase the risk of conflict while expenditure fiscal decentralization and are intergovernmental transfers contribute to limit the risk of within countries wars.

The rest of this article is structured around four sections. The first presents some stylized facts. The second provides a selective review of the literature. The third section presents the model, defines the variables and data sources, and the fourth section analyses and discusses the results. We conclude with a few remarks in terms of economic policy implications.

Here, some stylized facts about fiscal decentralization and internal conflict were produced.

1.1. Fiscal decentralization in Sub-Saharan Africa remains constant

Fiscal decentralization in Sub-Saharan Africa is highly dependent on transfers, which is about over 50% of local revenues and more often conditional than unconditional. The figure below shows the average evolution of the share of local government revenues, taxes and expenditures respectively in total general government revenues, taxes and expenditures. It also presents the proportion of transfers or grants in total local government revenues between 2002 and 2019. On observation, it can be seen that during this period the level of fiscal decentralization has remained globally constant. Decentralised resources and taxes, for their part, have undergone a U-shaped evolution, with a period of constant decline in the level of the indicators between 2002 and 2011 for the resources index and from 2002 to 2009 for the taxes index, and a period of constant growth from 2011 to 2019 and from 2009 to 2019 respectively. However, as it can be seen, both indicators remain low, with the revenue indicator fluctuating between 9.26% and 11.63% and the tax indicator between 1.91% and 8.09%. As for the expenditure indicator, the graph shows a downward trend from 13.62% in 2002 to 8.06% in 2019.

1.2. Internal conflicts in sub-Saharan Africa have increased over the time

Figure 2 shows on the left the average conflict score in the countries covered by our study and on the right its average evolution between 2002 and 2019. Two observations can be made. On the one hand, the average conflict score among these countries varies between 4.76 (higher or lower risk of conflict) and 10.56 (lower or higher risk of conflict). Countries with a high risk of conflict include Somalia (4.7), Nigeria (6.12), DRC (6.3), Uganda (6.7), Ethiopia (7), Côte d'Ivoire (7.4), Mali and Malawi (7.8). On the other hand, since 2004, the average internal conflict score has evolved considerably from 8.55 to 7.8 in 2019 as this average tends more towards critical (0) than towards the appeasement score (12).

The stylised facts above have shown that fiscal decentralisation needs to be improved in sub-Saharan African countries. Beside this, the internal risk conflict is growing in this part of the world. Does fiscal decentralisation improvement in this region constitute a potential conflict resolution tool? The following section analyses the relationship between the two variables in the literature.

The relationship between fiscal decentralization and internal conflict is widely discussed in the literature. In order to account for this, we present the theoretical foundations of the effect of fiscal decentralization on internal conflicts on the one hand, and on the other, we summarize the empirical work on this relationship.

2.1. The theoretical foundations for the effect of fiscal decentralization on internal conflicts

The relationship between decentralization and internal conflict can be identified through the first and second generation theories of fiscal federalism (Brennan and Buchanan 1980; Oates 1972; Tiebout 1956). Without reviewing the history of these theories, we can show that fiscal decentralization is a factor of social cohesion but it can also increase the risk of conflict by accentuating inter-individual and group inequalities.

Fiscal decentralization can constitute a factor of social cohesion. In most conflict countries, where a large segment of the population, often undereducated and generally deprived of basic social services, is more likely to be manipulated for terrorist purposes and tribal hatred, fiscal decentralization can ensure social cohesion through better resource allocation. Indeed, both first and second-generation theories agree that fiscal decentralization increases the efficiency of government in the provision of public goods and services. This efficiency argument is based on the principles of proximity between governors and governed, correspondence between needs and services offered, and competition between jurisdictions. Since all public goods have not the same spatial characteristics, the first two principles are based on the argument that for a relatively large country (area and/or population) it is difficult to provide public goods and services in a way that satisfies the desires and preferences of every individual, since there is no homogeneity of preferences among individuals. Thus, the proximity between local decision-makers and citizens can make it easier for individuals to reveal their preferences (Musgrave & Musgrave 1989; Tiebout 1956) so that the public services offered best correspond to them (Oates 1972). In addition to better knowledge of needs, proximity and correspondence also allow for a reduction in supply chains, the costs of providing public services since local provision is more likely to correspond to the needs of the needy, and the risks associated with the loss of redistributive power by the central government (Ezcurra and Pascual 2008). Moreover, proximity allows for better management of inter-individual conflicts, since it offers timely and inexpensive solutions at the local level. Thus, the proximity and correspondence offered by decentralization guaranties the social cohesion.

The advantages offered by these two principles are reinforced by the principle of interjurisdictional competition. This last principle, based on the hypothesis of inter-jurisdictional mobility of individuals, indicates that the latter are stratified and ranked according to their preferences for the public good. Thus, in the context of a sub-optimal supply of public goods and services in one locality, individuals can vote with their feet and move to another locality where the supply of public goods and services meets their preferences (Tiebout 1956). In this way, fiscal decentralization offers the possibility for populations with different community origins to cohabit without discrimination. Moreover, this individual mobility is accompanied by fiscal mobility, since the jurisdiction of origin of the migration loses part of its tax base to the host jurisdiction. Fiscal mobility in turn creates horizontal (between jurisdictions) and vertical (between the center and the periphery) fiscal competition that limits the behaviour of the tax revenue maximizing Leviathan (Brennan & Buchanan 1980). But this double competition can reinforce inter-regional inequalities and increase the risk of conflict.

Tax competition between jurisdictions can reinforce interregional inequality as high-income individuals flock to low-tax jurisdictions and displace low-income individuals, which in turn could exacerbate social disparities (Zimmermann and Henke 2001) and lead to political breakdown. This is all the more likely when this allocation does not benefit all jurisdictions equally (Alesina et al. 2000). Thus, it may happen in a country that jurisdictions that benefit less from public allocations harbor resentment towards jurisdictions that benefit more, and especially towards the central government responsible for this state of affairs. This is more likely to occur in developing countries where administrative divisions are more communal and where mobilities are generally migrations from villages to a few cities, since most of these countries have only one or two cities that house almost all of their affairs (Letelier 2005). Seen in this light, mobility in developing countries does not necessarily follow the principle of inter-jurisdictional competition that is supposed to ensure government efficiency in the delivery of public services. In addition, resources transferred to local governments may be diverted (Tanzi 2008) or be misused because of a lack of competence (Prud'homme 1995) and undermine the legitimacy of the state. More realistically, fiscal decentralization can serve personal ends through the maximization of political support (Hettich & Winer 1984 1988).

In an African context where decentralization is characterized by insufficient resources for local governments and inadequate financial arrangements with the central state (Yatta 2009), it can be a source of conflict. Considering that African local finances depend on intergovernmental transfers, there is a case for fiscal decentralization to serve political purposes. Indeed, policy makers at the central level may use fiscal transfers to maximize their chances of re-election (Grossman & West 1994). As such, there may be either a partisan effect in the allocation of federal transfers to subnational governments, which consists of giving local governments, which are supporters of the incumbent, more budgetary responsibilities than non-partisan governments (Case 2001; Cox 1986), or, on the contrary, a tactical effect, which consists of granting more budgetary responsibilities to opposition local governments in order to rally them to their cause (Caldeira 2011). In either case, there may be potential for conflict. Therefore, decentralization processes must take into account the risks associated with independence claims. All else being equal, governments concerned about the risk of secession are more likely to decentralize policy and encourage regional protectionism, but less likely to decentralize fiscal authority (Sorens 2016). As Sorens (2016) himself points out, the fiscal autonomy of subnational units is a potential source of funding for a future insurgency. This view is shared by Bardhan (2002) and Tanzi (2000). Moreover, like small states, jurisdictions can gain autonomy in the context of trade liberalization and economic integration (Alesina et al. 1995; Alesina et al. 2000; Alesina & Wacziarg 1998)

Based on the current literature, it is safe to say that there is no consensus regarding the relationship between decentralization and conflict. By giving local areas more taxing, spending and political power decentralization may indeed help avoid conflict or ease the tension between different groups in a country. At the same time, there is the risk that local autonomy over re-sources is used to fuel more conflict especially when there is no monitoring of the use of those resources by the central government, which can happen if there are no legal-institutional mechanisms in place for accountability (Mainali and al. 2022).

2.2. Synthesis of some empirical works

Empirical work analyzing the role of decentralization in conflict resolution is extensive, but work specifically studying the effect of fiscal decentralization on internal conflict is limited (Ezcurra 2015). Of the existing studies, while some have shown the inability of fiscal decentralization to resolve internal conflicts (Eaton 2005; Sanchez and Palau 2006; Green, 2008), others indicate that conflicts in a country can be avoided or resolved through fiscal decentralization (Ezcurra 2015; Farzanegan et al. 2018). But this effect depends on other factors (Green 2008; Siegle and O’mahony 2009; Sugiyanto and al. 2018).

Sánchez & Palau (2006) showed using the spatial Tobit method that between 1974 and 2004, the Colombian decentralization process transformed internal conflict into a dispute between local governments by intensifying violence in order to appropriate a share of public goods and resources. Green (2008) reached similar conclusions by showing that in Uganda fiscal decentralization reduces the risk of conflict at the national level but increases it at the local level because of the concentration of power in the hands of local elected officials and the emergence of new jurisdictions that have accentuated identity-based withdrawal. Before them, Eaton (2005) have shown that decentralization reforms may have come to Colombia too late such that decentralization reforms in the 1980s may have generated resources and a degree of political autonomy at the local level that contributed to more conflict. He notes however that decentralization would have helped if it were adopted before the conflict started in the 1960s. Based on the current literature, it is safe to say that there is no consensus regarding the relationship between decentralization and conflict. In line with this observation, some authors such as Mansoob et al (2009), Ezcurra (2015) and Farzanegan et al (2018) have reached interests results.

Using a new dataset, Mansoob et al (2009) examined the relationship between ordinary/daily violence and fiscal decentralisation in 98 districts on the Indonesian island of Java. Poisson regression results show that despite the uniform implementation of fiscal decentralisation, sub-national entities have diverse experiences of decentralisation. But a common consequence is an increase in the size of local government. Fiscal decentralisation and an increase in the size of local government can alleviate pent-up frustrations with a centralised state, as local government spending is perceived as meeting the needs of communities with which people identify more closely. The authors also find that the greater the proportion of resources generated locally, the lower the number of violent incidents; but this ability to generate more local income is found mainly in the wealthier districts. Consequently, wealthier districts are likely to have a lower incidence of violence. Ezcurra (2015) analysed the relationship between fiscal decentralization and internal armed conflict for a panel of 77 countries between 1972 and 2000. Applying probit and logit, the author shows that devolution of fiscal power to subnational governments reduces the incidence of civil conflict. Farzanegan et al. (2018) study the influence of fiscal and political decentralization on the relationship between natural resource rents and the risk of internal conflict for a panel of 90 countries between 1984 and 2004. Applying the fixed effects method, they show that fiscal and political decentralization reduces the propensity of resource rents to increase conflict risk. However, these optimistic findings are mitigated by these of Green (2008), Siegle and O'Mahony (2009), and Sugiyanto et al. (2018).

Siegle and O'Mahony (2009) showed in a cross-sectional study of 51 countries for the year 1995 that countries with a high level of expenditure decentralization were less prone to internal conflict than these with high levels of local or regional autonomy. This is virtually the same result Green had found a year earlier in Uganda. He argues that fiscal decentralization may have reduced conflict at the central level of government, but it transferred it to the local level. More recently, Sugiyanto et al. (2018) demonstrated that fiscal decentralization is more effective in reducing routine or permanent conflict in Indonesian provinces, but only if there are good institutions. Thus, by applying the fixed-effect negative binomial regression, they show that the phenomenon of elite capture present at all levels of government reduces the ability of fiscal decentralization to mitigate conflict in Indonesia. In addition, Nickson and Cutting (2016) provide a critical analysis of the situation in Sierra Leone; an example of a country where decentralisation is credited with helping to ease internal conflicts. They conclude that external pressure can indeed ‘lock-in’ reforms but that a more nuanced, iterative and locally contextualised approach based on sound political economy analysis is needed in order to foster and sustain reform gains. Additional for those mixed results, Mainali and al. (2022) has shown show that a decrease in conflict in response to the intergovernmental transfer reform, but the exact response varies by conflict region. Oil reserves in some municipalities seem to have played a role in the conflict outcomes as well.

Our methodological strategy proceeds in four steps. First, we present the empirical model. Then, after defining the variables, we identify the data sources. Finally, we present the estimation technique.

3.1. Specification of the empirical model

The empirical model below is based on Sorens (2016) and Sugiyanto et al. (2018) as follows:

$$\varvec{I}\varvec{n}\varvec{t}\varvec{e}\varvec{r}\varvec{n}{\varvec{a}\varvec{l}\_\varvec{C}\varvec{o}\varvec{n}\varvec{f}\varvec{i}\varvec{c}\varvec{t}}_{\varvec{i}\varvec{t}}=\varvec{\alpha }+\varvec{\gamma }{\varvec{F}\varvec{D}}_{\varvec{i}\varvec{t}}+\varvec{\delta }{\varvec{X}}_{\varvec{i}\varvec{t}}+{\varvec{\mu }}_{\varvec{i}\varvec{t}}$$

Where ${FD}_{it}$ is the fiscal decentralization of country i in period t ; ${X}_{it}$ is the matrix of control variables accounting for external conflict, socioeconomic conditions, religious tensions, ethnic tensions, corruption; $\gamma \text{a}\text{n}\text{d} \delta$ are the parameters to be estimated, and ${\mu }_{ti}$ the error term, where $i=\text{1,2},\dots , 24$, representing the sample size and$t=2002,\dots , 2019$, the time series.

3.2. Description of variables

3.2.1. Dependent variable: Internal Conflict

Internal conflict measures political violence in the country and its actual or potential impact on governance. The highest rating is given to countries where there is no armed or civil opposition to the government and where the government does not engage in arbitrary violence, direct or indirect, against its own population. The lowest rating is assigned to a country with an ongoing civil war. The risk score assigned is the sum of three subcomponents (civil war/threat of coup, terrorism/political violence, civil unrest), each with a maximum score of 4 points and a minimum score of 0. A score of 4 points equals very low risk and a score of 0 points equals high risk.

3.2.2. Independent variable: fiscal decentralization

Fiscal decentralization is the set of intergovernmental fiscal relations. The measurement of fiscal or budgetary decentralization is debated in the literature because it has several aspects and dimensions (Qiao et al. 2019). No single index is able to capture all the different aspects of it (Martinez-Vazquez & McNab 2003; Stegarescu 2006). The Government Finance Statistics (GFS) published by the IMF provide at least seven indicators of fiscal decentralization. In this study, we exploring tree different aspects of revenue decentralization index. These indices take to account total revenue, tax and transfer indices. The revenue index measures the share of total local government revenue in total general government revenue (central + local). The tax index measures the share of taxes collected on total local revenue and the transfer indicator is measure as the proportion of local revenue due to central government transfer/dotation. In addition, expenditure index is takes in to the account on this study. The four indicators mentioned above where measured follow Bodman (2011).

3.2.3 Explanatory variables: (Xit)

- External conflict

External conflict assesses the risk to an incumbent government of foreign action, ranging from nonviolent external pressures (diplomatic pressure, denial of aid, trade restrictions, territorial disputes, sanctions, etc.) to violent external pressures (cross-border conflict or all-out war) (Howell 2011). External conflict can affect foreign companies in many ways, ranging from restrictions on operations to trade and investment sanctions to denial of access to information. Like internal conflict, its score is the sum of three subcomponents (war, cross-border conflict, and foreign pressure) with the same scores.

- Socio-economic conditions

This is an assessment of the socio-economic pressures at work in society that could constrain government action or fuel social discontent (Howell, 2011). The risk score assigned is the sum of three subcomponents (unemployment, consumer confidence, and poverty), each with a maximum score of 4 points (very low risk) and a minimum score of 0 points (very high risk). Countries with poor socio-economic conditions are usually the hotbeds of conflict.

- Religious tensions

Religious tensions can result from the domination of society and/or governance by a single religious group that seeks to replace civil law with religious law and exclude other religions from society (Howell 2011). The risks associated with these situations range from inexperience to the imposition of inappropriate policies, to civil dissent, to civil war. This risk is scored between 0 (very high tension) and 6 (very low tension).

- Ethnic tensions

This component is an assessment of the degree (from 0 to 6) of tension attributable to racial, nationality, or language divisions within a country (Howell 2011). Countries with high racial and national tensions are rated lower because opposing groups are intolerant and unwilling to get along. Countries with minimal tensions receive higher ratings, and vice versa. These tensions hinder social cohesion and exacerbate conflict in a state.

- Corruption

This is an assessment of corruption within the political system. This corruption poses a threat to foreign investment for several reasons: it distorts the economic and financial environment; it reduces the efficiency of government and business by allowing individuals to gain access to positions of power through patronage rather than through their skills; and, last but not least, it introduces inherent instability into the political process (Howell 2011). A positive effect is therefore expected.

3.3. Data and sources

The data used in this study come from a variety of sources. They cover 21 countries in Sub-Saharan Africa between 2002 and 2019 (Burkina-Faso, Cameroon, Congo, Democratic Republic of Congo, Ivory Coast, Ethiopia, Ghana, Kenya, Madagascar, Malawi, Mali, Namibia, Niger, Nigeria, Senegal, Somalia, South Africa, Tanzania, Togo, Uganda and Zambia). Data on revenues, expenditures, and intergovernmental transfers used to calculate the fiscal decentralization indices are provided by the IMF's Government Finance Statistics (GFS 2020). The International Country Risk Guide (2020) provided the data on internal conflict and most of the control variables. The definitions of the variables and data sources are shown in Table 1 below.

Table 1

Data descriptions
Variables	Description	Sources of data
IntConf	The risk of conflict is between 0 (very high) and 12 (very low)	ICRG (2020)
EthTens	It takes the score from 0 (very high) to 6 (very low)	ICRG (2020)
SoEcCon	The risk of deteriorating socio-economic conditions 0 (very high) to 12 (very low)	ICRG (2020)
ExtConf	The risk of conflict is between 0 (very high) and 12 (very low)	ICRG (2020)
Corr	Corruption is rated from 0 (very corrupt) to 6 (no corruption)	ICRG (2020)
RelTens	It takes the score from 0 (very high) to 6 (very low)	ICRG (2020)
TranFD	Share of local government revenue due to central government transfer (between 0 and 1)	Author's calculation from GFS (2020)
RevFD	Share of local government revenue in total state revenue between 0 (absence) and 1 (full decentralization)	Author's calculation from GFS (2020)
TaxFD	Share of local government tax in total general government tax. it takes between 0 (absence of fiscal autonomy) and 1 (full fiscal autonomy)	Author's calculation from GFS (2020)
ExpFD	Share of local government expenditure in total general government expenditure between 0 (absence) and 1 (full decentralization)	Author's calculation from GFS (2020)

Source : authors

Table 2

Descriptive statistics
Variable	Obs	Mean	Std. Dev.	Min	Max
RevFD	361	0,1025063	0,1094633	0,0013995	0,5780661
TaxFD	198	0,0397847	0,0678317	0	0,3679018
TranFD	216	0,6138712	0,2806765	0,011	1
ExpFD	216	0,0899478	0,0963749	0,02589	0,4813925
EthTens	378	4,135472	2,568846	0	12
IntConf	378	8,023038	1,541611	3,5	11
ExtConf	378	8,308532	2,969875	1	12
RelTens	378	3,8222	1,316766	0,5	6
SoCond	378	3,064145	1,329456	0	6,666667
Corr	378	1,96131	0,6676022	0	4

Source: authors

The choice of time period, 2002 to 2019, is dictated by the availability of data obtained simultaneously for the countries in the sample. The descriptive statistics summarized in Table 2 indicate low standard deviations of our variables of interest for our data to be dispersed. The correlation matrix (Table 3) shows that the different correlations between the variables used are generally not high enough to create serious multicollinearity problems.

Table 3

Correlation matrix
	RevFD	TFD	TranFD	EFD	EthTens	IntConf	ExtConf	RelTens	SoCond	Corr
RevFD	1.0000
TFD	0.0012	1.0000
TranFD	0.1859	-0.2440	1.0000
EFD	0.3316	-0.0818	0.2233	1.0000
EthTens	-0.2855	0.1108	-0.2307	-0.0232	1.0000
IntConf	-0.5191	-0.0807	0.1444	0.0185	0.6376	1.0000
ExtConf	-0.6388	0.1156	-0.1526	0.0346	0.5782	0.8345	1.0000
RelTens	-0.1289	-0.2387	-0.2964	0.0484	0.4938	0.6865	0.4495	1.0000
SoCond	-0.2474	0.1022	-0.1214	0.2696	0.2886	0.6182	0.6730	0.4724	1.0000
Corr	-0.2073	0.0852	-0.0620	0.0632	0.4388	0.6408	0.5986	0.3937	0.5234	1.0000

Source: authors

3.4. Econometric strategy

The effect of fiscal decentralization on the internal conflict was firstly analyse through Eq. (1) by three methods including pooled ordinary least squares (POLS), fixed effects (FE) and random effects (RE). First, POLS was performed with the assumption of exogenous regressors. Then, FE and RE are employed to remove the effects of country unobserved characteristics. Next, the Breusch and Pagan Lagrangian multiplier test was performed to choose between POLS or RE. The result indicates that the panel models are better than POLS. Here, Hausman test was used to choose between FE and RE. FE are seemed to be the appropriate one. Finally, the cross-sectional dependence test by Pesaran (2021) indicated that our panels models are correlated. We then adopt an estimation strategy to overcome this cross-sectional dependence. Some studies (Hoechle 2007; Reed and Ye, 2011; Chudik and Pesaran, 2015) demonstrate that the estimation methods that overlook the cross-sectional issues lead to biased and inconsistent estimators. Furthermore, the presence of cross-sectional and temporal dependence in the panel data means the slackening of some econometrics traditional stochastic assumptions (Pesaran 2021). The alternative estimation techniques must be considered to get suitable results. The econometrics and economic literature highlight several estimation approaches to tackle the cross-sectional dependence in the static panel models. The spatial econometrics method falls within these approaches (see Corrado and Fingleton 2012; Moundigbaye and al. 2018 for the comments).

Besides the spatial econometrics’ method, the standard econometrics offers some estimation methods that enable to master the cross-sectional dependence. These methods include the Feasible Generalized Least Squares (FGLS) by Parks (1967), Panel Corrected Standard Error (PCSE) by Beck and Katz (1995), Common Correlated Effects Estimator (CCEE) by Pesaran (2006) and estimation technique by Hoechle (2007). The latter is based on the non-parametric estimation technique by Driscoll and Kraay (1998). Although both latters’ estimation techniques surmount the cross-sectional dependence issue, they usually result in the misleading specification by considerably reducing the number of parameters to be estimated (Moundigbaye & al. 2018; Moundigbaye and al. 2020). The FGLS by Parks (1967) and PCSE by Beck and Katz (1995) achieve the same estimation objectives. The two estimation approaches consider at once the cross-sectional dependence, heteroscedasticity and errors autocorrelation (Reed and Ye 2011; Moundigbaye and al. 2020). However, to choose between both estimation methods depends on the ratio between the time dimension (T) and individual dimension (N) (Reed and Ye 2011; Moundigbaye and al. 2018). When this ratio is less than one (1) (meaning T < N), the FGLS estimation strategy provides lower reliable estimation outcomes or is even impracticable due to the inability to compute the errors variance-covariance matrix inverse (Beck and Katz 1995; Reed and Ye 2011). However, when this ratio is greater than one (1) (meaning N < T), the FGLS estimation strategy is preferred to PCSE as it produces more efficient estimators and remains better in performance and precision than the PCSE (Chen and al. 2010; Reed and Webb 2010; Reed and Ye 2011). Given that our panel configuration is in accordance with N < T from transfer, expenditure and tax and decentralization, we perform our estimations using the FGLS estimation technique. And from revenue decentralization which is assume to have T > N, PCSE is used. For robustness checking, 2SLS by Lewbel (2018) has been used. The results of different estimations are presented in the section below.

We first present results for baseline model and then we present the robustness.

4.1. Baseline results

As baseline results, this subsection presents the RE, FE, FGLS and PCSE. FE and RE results are reported in Table 4 while FGLS and PCSE results are in Table 5. As it can be seen, fixed and random effects estimations provide statistically significant results at the 1% level for the resources, taxes and transfers variables. While resources and taxes have a negative effect on the risk of internal conflict in sub-Saharan Africa, transfers have a positive effect. In concrete terms, decentralization of resources and decentralization of taxes increase the risk of internal conflict in these sub-Saharan African countries, while financial transfers from central to local government reduce it. Furthermore, in the eight models, the control variables are not all significant, nor is the decentralized spending variable. This is certainly due to the cross-sectional dependence and heteroscedasticity problems identified through the Wooldridge (2010) and Pesaran (2021) tests. The search for solutions to these problems has led to the use of the FGLS and PCSE techniques, which also make it possible to correct the problems of cross-sectional dependence and multi-collinearity (Hoechle 2007). The results of these estimates are presented in Table 6 below. It can be seen that all our variables of interest are significant at the 1% level. Unlike the results in the previous table, where the decentralization of expenditure had a negative but insignificant correlation, the results in this table indicate a positive and significant correlation. In addition, most of the control variables are significant at the 5% level.

Concerning the positive effects of expenditure decentralization (column 2) and intergovernmental transfer (column 4), it can be analysed as the fact that more decentralized expenditure and transfer are led to the minimum within countries conflict. In other words, the countries that decentralize the most in terms of spending and resources transfers generally have a very low risk of internal conflict. This is because an increase in local government spending is likely to be accompanied by an increase in basic public services, the inadequacy, scarcity and sometimes even absence of which is most often at the root of tensions between populations and governments. Education and unemployment are some of these services which, when they become bogged down, lead to poverty and idleness, two major problems that characterize the populations of countries in conflict. Faced with this situation, it would seem that an increase in intergovernmental transfers is likely to increase appeasement or reduce the risk of conflict within a country. These results are in line with theoretical findings on the merits of decentralization and corroborate the work of Ezcurra (2015), providing indeed an empirical support for the social cohesion hypothesis resulting from the allocative efficiency induced by fiscal decentralization.

The negative effect of revenue (column 1) and tax (column 3) decentralization means that internal conflict risk increases with more revenue and tax decentralization. This can be explained by the fact that in many African countries, resources and tax decentralization are not used for the initial purpose. If we accept that in most of these states’ resources are often misappropriated by the leaders, decentralizing the collection of tax revenues and increasing the resources of local governments can have the same effects. As a result, local authorities can use resources for their own personal ends. This point of view is reflected in the dangers/pitfalls of decentralization, often highlighted by Prud'homme (1995) and Tanzi (1996; 2000), among others. Both authors point out that fiscal decentralization increases the risk of secession in a country. These results contradict those of Farzanegan and al. (2018), who show that increasing the resources of local governments and, above all, increasing their autonomy helps to reduce the risk of conflict.

Table 4

effect of fiscal decentralization on internal conflict through FE and RE
Variables	Internal Conflict
	FE				RE
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
RevFD	-1.873**				-1.775**
	(0.893)				(0.759)
ExpFD		-0.715				-1.058
		(1.099)				(0.828)
TaxFD			-2.243**				-2.880***
			(0.909)				(0.858)
TranFD				0.844***				0.592***
				(0.237)				(0.215)
EthTens	0.0441	0.421	0.433	0.629*	0.302***	0.373***	0.406***	0.340***
	(0.119)	(0.281)	(0.280)	(0.340)	(0.0605)	(0.132)	(0.128)	(0.113)
ExtConf	0.434***	0.414***	0.439***	0.305***	0.354***	0.446***	0.460***	0.444***
	(0.0768)	(0.0685)	(0.0703)	(0.0818)	(0.0530)	(0.0557)	(0.0596)	(0.0591)
RelTens	0.0834	-0.0553	-0.270**	-0.272*	0.285***	0.257***	0.177**	0.366***
	(0.146)	(0.120)	(0.129)	(0.154)	(0.0818)	(0.0783)	(0.0824)	(0.0728)
SoCond	0.221**	-0.00536	0.102	0.216*	0.185**	0.0622	0.104	0.0981
	(0.0904)	(0.0958)	(0.108)	(0.118)	(0.0721)	(0.0716)	(0.0739)	(0.0695)
Corr	0.0677	0.254***	0.247***	0.115	0.0632	0.301***	0.295***	0.202*
	(0.0917)	(0.0968)	(0.0938)	(0.111)	(0.0904)	(0.0946)	(0.0939)	(0.112)
Constant	3.305***	2.600**	2.764**	2.733**	2.224***	1.070**	1.004**	0.421
	(0.988)	(1.108)	(1.135)	(1.382)	(0.634)	(0.494)	(0.492)	(0.469)
Obser.	361	216	198	216	361	216	198	216
R²	0.156	0.197	0.259	0.166	0.8557	0.9480	0.9476	0.9526
Nbr Ind.	21	12	11	12	21	12	11	12
Fisher	10.28	8.09	10.55	6.65	166.27	225.33	260.73	338.56

Notes: *; **; *** significance at the 10%, 5%, and 1% levels, respectively, and ( ) standard deviations.

Source: Authors

Concerning explanatory variables, the results obtain are those was expected. External conflicts, ethnic tensions and religious tensions are all significant at the 1% level in the four models (columns 1 to 4), while corruption is significant for the expenditure model and decentralised taxes (columns 2 and 3). Socio-economic conditions are significant for resources, expenditure and transfers (columns 1, 2 and 4), but not for taxes. The positive and significant effect of external conflict, ethnics tensions, religious tensions and corruption means that they contribute to increase the risk of internal conflict within a country. The existence of ethnic and religious tensions in a country is evidence of a lack of social cohesion. A country in which there are religious and ethnic tensions is exposed to civil war. The case of tensions between Tutsi and Ottu in Rwandan is a perfect illustration. If we add to this the corruption that erodes the population's confidence in the government, its aggravation can lead to demands for the removal of the governing authorities. Joined this with the EthConf and RelConf effects, these results are consistent with the results of Siegle and O'Mahony (2009) who found that religious and ethnic conflicts increase the risk of internal conflict. Despite the fact that most of sub-Saharan African countries are characterized by poor socioeconomic conditions, our results are showing that improving socio-economic conditions reduces the risk of conflict. Sanchaz and Palau (2006) have shown that poor socio-economic conditions accentuated the risks of war in Colombia.

Table 5

effect of fiscal decentralization on internal conflict through PCSE and FGLS
Variables	Internal Conflict
	PCSE	FGLS
	(1)	(2)	(3)	(4)
RevFD	-2.298***
	(0.121)
ExpFD		1.382***
		(0.208)
TaxFD			-3.352***
			(0.339)
TranFD				0.346***
				(0.0654)
EthTens	0.246***	0.199***	0.198***	0.148***
	(0.0450)	(0.0324)	(0.0331)	(0.0350)
ExtConf	0.339***	0.268***	0.351***	0.163***
	(0.0196)	(0.0225)	(0.0255)	(0.0274)
RelTens	0.260***	0.122***	0.188***	0.354***
	(0.0351)	(0.0200)	(0.0281)	(0.0249)
SoEcoCond	-0.126**	-0.0884***	-0.0258	-0.0505*
	(0.0532)	(0.0229)	(0.0269)	(0.0269)
Corr	0.0683	0.111***	0.0859**	0.0359
	(0.0581)	(0.0362)	(0.0364)	(0.0352)
Constant	2.675***	4.243***	3.562***	4.456***
	(0.635)	(0.223)	(0.266)	(0.251)
Observations	361	216	198	216
Number of ind (N)	21	12	11	12
Wald	4934.97	636.11	582.88	1018.48
R²	0.918	-	-	-
Methods criteria	T > N	T < N
Notes: ; ; * significance at the 10%, 5%, and 1% levels, respectively, and ( ) standard deviations.*

Source: Authors

The proximity of a state to hotbeds of tension and conflict is another determinant of insecurity within that state. In general, it is almost inevitable to suffer the pitfalls of a confrontation in the neighbouring country. The Boko Haram crisis that affects Nigeria, Chad and Cameroon due to their territorial proximity is a perfect illustration. External conflicts constitute precedents that influence the birth of other conflicts in other countries. And as Sorenz (2015) said, even if most countries use transfers as a fiscal appeasement mechanism, this can still lead to social tensions when local governments find themselves asphyxiated because of delays in transfers that are often insufficient to meet the needs of the population.

4.2. Robustness cheeks

This subsection is concerned by results robustness check. For this purpose, 2SLS by Lewbel (2018) is used to analyse the effect of fiscal decentralization on internal conflict in SSA. The results are presented in table 6 below and as it can be seen, they are globally consistent with those previously obtained. Indeed, revenue and tax decentralization are negatively and significantly associate to internal conflict. Elsewhere, expenditure decentralization and intergovernmental transfer both have a positive and significant effect on internal conflict. Precisely, the rise of 1% of RevFD (model 1) is associated with the growing in the 1.42% of internal conflict risk. In the same way, the increase of TaxFD (Model 3) by 1% reduced at 2.72% of appeasement situation. In contrast, rising ExpFD (Model 2) by 1% helps to reduce the risk of internal conflict by 0.98%. And in the last model where fiscal decentralization is captured by transfers from central to local government (TranFD), the rise up of it at 1% lead to reduce the risk conflict within the sub-Saharan African countries at 0.83%.

In addition, the explanatory variables are all significant excepted socioeconomics condition. Corruption is a major constraint on government effectiveness. This inefficiency, coupled with ethnic and religious tensions, undermines social cohesion and can ultimately encourage and generate verbal clashes between socioeconomic, ethnic, and religious groups. These clashes can lead to civil wars if they are between populations. But they can also lead to secessionist claims, as is the case in Cameroon, when part of the population feels that the poor socio-economic conditions to which they are subjected are the result of the unequal management between French and anglophone peoples. Moreover, the resurgence of coups d'état observed in West and Central Africa testifies to a generalization of the government's inefficiency in guaranteeing quality socioeconomic conditions for its population.

Table 7

Robustness check trough Lewbel 2SLS instrumental variable
Variables	Internal Conflict
	Lewbel 2SLS
	(Model1)	(Model 2)	(Model 3)	(Model 4)
RevFD	-1.427***
	(0.520)
ExpFD		0.989**
		(0.479)
TaxFD			-2.720***
			(0.744)
Transfer				0.831***
				(0.309)
ExtConf	0.392***	0.531***	0.532***	0.565***
	(0.0343)	(0.0386)	(0.0420)	(0.0528)
EthTens	0.384***	0.186***	0.207***	0.113
	(0.0371)	(0.0684)	(0.0702)	(0.0892)
RelTens	0.391***	0.444***	0.400***	0.525***
	(0.0410)	(0.0390)	(0.0448)	(0.0603)
SoEcoCond	-0.0393	-0.00859	-0.00681	-0.0304
	(0.0483)	(0.0448)	(0.0440)	(0.0571)
Corr	0.163*	0.444***	0.456***	0.313***
	(0.0882)	(0.0823)	(0.0838)	(0.108)
Constant	1.399***	0.180	0.242	-0.482
	(0.365)	(0.242)	(0.259)	(0.439)
Observations	340	204	187	196
R-squared	0.698	0.871	0.876	0.801
F Statistics	128.78	221.66	213.37	126.21
Notes: ; ; * significance at the 10%, 5%, and 1% levels, respectively, and ( ) standard deviations.*

Source: Authors

The aim of this paper was to analyse the effect of different aspects of fiscal decentralization on the risk of internal conflict in sub-Saharan African countries. Using four indicators of fiscal decentralization, this study mobilized data for a sample of 21 countries for resource decentralization, 12 countries for expenditure and transfers, and 11 countries for taxes over the period from 2002 to 2019. The use of several econometric estimation techniques led to two fundamental results: (1) the decentralization of resources and the empowerment of local authorities through greater taxing power increase the risk of internal conflict in sub-Saharan Africa; (2) increased spending at the local level and the financial transfers from central to local governments contribute to a state of appeasement. In addition to these results, it has been noted that ethnic and religious tensions, corruption and external conflicts are catalysts for conflict within a country. But an improvement in socio-economic conditions is likely to reduce the related risk.

Three main economic policy implications can be drawn from the results to further reduce the risk of internal conflict and secession. Firstly, financial transfers to local authorities should be increased, while respecting the principle of subsidiarity. Similarly, the level of their expenditure should be increased. Recognizing that decentralization initiatives can pose risks, those strategies should be accompanied by a comprehensive conflict risk analysis. Secondly, improve the institutional framework, in particular by promoting better governance and controlling corruption, so as to increase the confidence of the beneficiary populations in the governing authority and their interest in the important issues of city management. Finally, improve socio-economic conditions through effective and efficient investment. An interesting extension of this study would be to assess, for these same countries, the impact of the existence of natural resources on the emergence and management of conflicts. This could also be done in a context of decentralization.

Disclosure statement

The authors have no relevant financial or non-financial interests to disclose.

Data availability statement

The data of this paper are available and will be sent on request.

AKUF (2020) Zwei neue Kriege in diesem Jahr, Universität Hamburg.
Alesina A, Perotti R, & Spolaore E (1995) The Political Economy of Fiscal Discipline Togetheror separately ? Issues on the costs and benefits of political and fiscal unions. Journal of Political Economy, 39, 751–758.
Alesina A, Spolaore E, Wacziarg R (2000) Economic Integration and Political Disintegration. American Economic Review, 90(5), 1276–1296. https://doi.org/10.1257/AER.90.5.1276
Alesina A, Wacziarg R (1998) Openness, country size and government. Journal of Public Economics, 69(3), 305–321. https://doi.org/10.1016/S0047-2727(98)00010-3
Bardhan P (2002) Decentralization of governance and development. Journal of Economic Perspectives, 16(4), 185–205. https://doi.org/10.1257/089533002320951037
Beck N, Katz J N (1995) What to Do (and not to Do) with Time-Series Cross-Section Data.” American Political Science Review 89 (3): 634–647.
Bird R, Vaillancourt F (1998) Décentralisation financière et pays en développement : concepts, mesure et évaluation. L’Actualité Économique, 74(3), 343–362. https://doi.org/10.7202/602266ar
Bodman P (2011) Fiscal decentralization and economic growth in the OECD. Applied Economics, 43(23), 3021–3035. https://doi.org/10.1080/00036840903427208
Brennan G, Buchanan, J. M. (1980) The Power to Tax: Analytical Foundations of a Fiscal Constitution (Vol. 48, Issue 192) Press syndicate of Cambridge University. https://doi.org/10.2307/2553704
Caldeira E (2011) Essais sur la decentrajisation dans les pays en développement [Auvergne - Clermont-Ferrand I]. In HAL archives-ouvertes. https://tel.archives-ouvertes.fr/tel-00643382
Cheng H, Cen L, Wang Y, Li H (2020) “Business Cycle Co-Movements and Transmission Channels: Evidence from China. Journal of the Asia Pacific Economy 25 (2): 289–306.
Chudik A, Pesaran M H (2015) “Common Correlated Effects Estimation of Heterogeneous Dynamic Panel Data Models with Weakly Exogenous Regressors.” Journal of Econometrics 188 (2): 393–420.
Clark T S, Linzer D A (2015) Should I Use Fixed or Random Effects? Political Science Research and Methods, vol. 3(2), pp.399-408.
Corrado L, Fingleton B (2012) Where Is the Economics in Spatial Econometrics?” Journal of Regional Science 52 (2): 210–239.
Driscoll J C, Kraay A C (1998) Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data.” Review of Economics and Statistics 80 (4): 549–560.
Eaton K (2006) The downside of decentralization. Armed clientelism in Colombia. Security Studies, 15(4):533–62.
Ezcurra R (2015) Fiscal decentralization and internal conflict: an empirical investigation. Environment and Planning C: Government and Policy, 33(3), 580–600. https://doi.org/10.1068/c13141r
Farzanegan M R, Lessmann C, Markwardt G (2018) Natural resource rents and internal conflicts: Can decentralization lift the curse? Economic Systems, 42(2), 186–205. https://doi.org/10.1016/j.ecosys.2017.05.009
Frey B S, Luechinger S (2004) Decentralization as a disincentive for terror. European Journal of Political Economy, 20(2), 509–515.
Green E D (2008) Decentralisation and conflict in Uganda. Conflict, Security & Development, 8(4), 427–450. https://doi.org/10.1080/14678800802539317
Grossman P J, West E G (1994) Federalism and the Growth of Government Revisited. Public Choice, 79(1/2), 19–32.
Hettich W, Winer S (1984) A Positive Modele of Tax Structure. Journal of Public Economics, 24, 67–87.
Hettich W, Winer S L (1988) Economic and Political Foundations of Tax Structure. American Economic Review, 78(4), 701–712.
Hoechle D (2007) Robust Standard Errors for Panel Regressions with Cross-Sectional Dependence.” The Stata Journal 7 (3): 281–312.
Howell L D (2011) International country risk guide methodology. 17. https://www.prsgroup.com/
Kee J E (2004) Fiscal decentralization: theory as reform. Financial Management Theory in the Public Sector, September, 253–286.
Keil S, Anderson P (2018) Decentralization as a tool for conflict resolution. in: Detterbeck, K. and Hepburn, E. (ed.) Handbook of Territorial Politics, Cheltenham and Northampton Edward Elgar. pp. 89-106.
Lewbel A (2018) Identification and Estimation Using Heteroscedasticity Without Instruments: The Binary Endogenous Regressor Case. Economics Letters, 165, 10-12.
Mainali R, Tosun M S, Yang J (2022) Fiscal decentralization, intergovernmental transfer reform and conflict in Colombian municipalities. Socio-Economic Planning Sciences, 83, 101108.
Mansoob S, Zulfan Tadjoeddin M, Chowdhury A (2009) Is fiscal decentralization conflict abating? Routine violence and district level government in Java, Indonesia. Oxford Development Studies, 37(4), 397-421.
Martinez-Vazquez J, McNab R M (2003) Fiscal decentralization and economic growth. World Development, 31(9), 1597–1616.
Moundigbaye M, Messemer C, Parks R W, Reed W R (2020) Bootstrap Methods for Inference in the Parks Model.” Economics 14 (1): 1–18.
Moundigbaye M, Rea W S, Reed W R (2018) Which Panel Data Estimator Should I Use? : A Corrigendum and Extension.” Economics 12 (1): 1–31.
Musgrave R A, Musgrave P B (1989) Public Finance in Theory and Practice. (McGraw-Hill (ed.); McGraw-Hil) McGraw-Hill Book Company.
Nickson A, Cutting J (2016) The role of decentralisation in post-conflict reconstruction in Sierra Leone. Third World Thematics: A TWQ Journal, 1(6), 799-816.
Oates W E (1972) Fiscal Federalism. In H. B. Jovanovich (Ed.), Harbrace series in business and economics. Harcourt Brace Jovanovich.
Parks R W (1967) “Efficient Estimation of a System of Regression Equations When Disturbances Are Both Serially and Contemporaneously Correlated.” Journal of the American Statistical Association 62 (318): 500–509.
Pesaran M H (2006) “Estimation and Inference in Large Heterogeneous Panels with a Multifactor Error Structure.” Econometrica 74 (4): 967–1012.
Pesaran M H (2021) General Diagnostic Tests for Cross-Sectional Dependence in Panels.” Empirical Economics 60: 13–50.
Qiao M, Ding S, Liu Y (2019) Fiscal decentralization and government size: The role of democracy. European Journal of Political Economy, 59(November 2018), 316–330.
Reed W R, Webb R (2010) The PCSE Estimator Is Good – Just not as Good as You Think.” Journal of Time Series Econometrics 2 (1)
Reed W R, Ye H (2011) Which Panel Data Estimator Should I Use?” Applied Economics 43 (8): 985–1000.
Sánchez F, Palau M (2006) Conflict, Decentralisation and Local Governance in Colombia, 1974-2004. Documentos CEDE, 20.
Siegle J, O’mahony P (2009) Asessing the Merits of Decentralisation as a Conflit Mitigation Strategy.
Sorens J P (2016) Secession Risk and Fiscal Federalism. Publius, 46(1), 25–50.
Stegarescu D (2006) Public Sector Decentralisation: Measurement and International Trends. Fiscal Studies, 26(3), 301–333.
Sugiyanto E, Digdowiseiso K, Zulmasyhur, Setiawan H (2018) Fiscal decentralization and routine conflict in Indonesia. Journal of Applied Economic Sciences, 13(4), 961–969.
Tanzi V (2001) Pitfalls on the road to fiscal decentralization.
Tanzi V (2008) The future of fiscal federalism. European Journal of Political Economy, 24(3), 705–712.
Tiebout C M (1956) A Pure Theory of Local Expenditures. Journal of Political Economy, 64(5), 416–424.
Treisman D (2006) Explaining fiscal decentralisation: Geography, colonial history, economic development and political institutions. In Commonwealth and Comparative Politics (Vol. 44, Issue 3)
Wooldridge J M (2010) Econometric analysis of cross section and panel data. The MIT Press.
World Bank (1997) World Development Report: The State in a Changing World. New York: Oxford University Press.
Yatta P F (2009) Fiscal Decentralization in Africa: Issues and Perspectives, Ed. Karthala, Paris.
Zimmermann H (2001) Finanzwissenschaft, Eine Einführung in die Lehre der öffentlichen Finanzwirtschaft, 7. Auflage.

Often there is a community that feels disadvantaged and in addition there is the factor of organizing the struggle on behalf of that community, explains Wolfgang Schreiber, one of the authors of the report.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Is fiscal decentralization in help for internal conflict risks mitigation? Sub-Saharan African countries focus

Status:

Version 1

Abstract

Figures

Introduction

1. Fiscal decentralization and internal conflict: some stylized facts

1.1. Fiscal decentralization in Sub-Saharan Africa remains constant

1.2. Internal conflicts in sub-Saharan Africa have increased over the time

2. Literature review

2.1. The theoretical foundations for the effect of fiscal decentralization on internal conflicts

2.2. Synthesis of some empirical works

3. Methodological strategy

3.1. Specification of the empirical model

3.2. Description of variables

3.2.1. Dependent variable: Internal Conflict

3.2.2. Independent variable: fiscal decentralization

- External conflict

- Socio-economic conditions

- Religious tensions

- Ethnic tensions

- Corruption

3.3. Data and sources

Source : authors

Source: authors

Source: authors

3.4. Econometric strategy

4. Results and discussion

4.1. Baseline results

Source: Authors

Source: Authors

4.2. Robustness cheeks

Source: Authors

Conclusion and policy implications

Declarations

References

Footnotes

Additional Declarations

Status:

Version 1