Political Regimes in Africa: The Impact of Resource Wealth

Resource Wealth and Political Regimes in Africa

Introduction

Political economists view economic development, poverty, and income inequality as key determinants of political regimes.
This study provides empirical evidence that there's a strong negative correlation between a sizable natural resource sector and democracy in Africa.
Resource abundance impacts democratic transition and consolidation.
Post-Cold War democratic reforms succeeded in resource-poor countries (Benin, Mali, Madagascar).
Resource-rich countries (Nigeria, Gabon) need strong accountability mechanisms for democratization.

Resource Abundance and Political Regimes

Executive discretion in distributing resource rents significantly impacts political regimes in Africa.
Discretion in distributing oil or mineral revenues leads to democratic breakdown or authoritarian endurance.
Rentier states depend on external rents from a few economic actors, typically from natural resource exploitation.
Rents are not generated from production (labor), investment (interest), or risk management (profit).
Rentier states are autonomous due to large natural resource endowments, reducing the need for taxes.
Mahdavy (1970) argues that this explains the lack of pressure for democratic change in the Middle East.
Yates (1996) suggests rentier states have poor governance because officials use resource rents for unpopular or illegal objectives.
Karl (1996) posits that fiscal reliance on petrodollars weakens the state and creates political instability.
Wantchekon (1999) finds resource abundance increases income inequality and dictatorial regimes.
Ross (1999) notes the rentier state literature is based on case studies of wealthy petrostates and lacks cross-sectional or panel testing.
Wantchekon (1999) and Ross (2000) filled this gap; this analysis provides a robustness test in the African context.
Executive discretion over resource rents leads to less political liberalization and greater democratic breakdown.
Empirical analysis shows resource-dependent economies:
- Were more likely to be authoritarian.
- Exhibit higher government spending.
- Are associated with worse governance.
- Were more likely to lead to breakdown in democracy after the third wave of democratic transitions in the 1990s.

Resource Dependence and Democracy

Resource dependence is negatively correlated with changes in the level of democracy, corroborating Bratton's (1998) finding of a decline in democracy in post-third wave democratization.
Bratton's data shows:
- Decline in leadership alternation (37% to 6.6%).
- Increase in opposition boycotts (11% to 73%).
- Increase in winner's vote share (61.4% to 69.1% for presidential elections, 62.7% to 72.0% in parliamentary elections).
Opposition boycotts and election rigging mostly occurred in petrostates or resource-dependent countries (Gabon, Cameroon, Togo, Zambia).
In resource-dependent countries (Algeria, Congo, former Zaire, Sierra Leone), democratization led to civil wars.
From 1965 to 1990, nearly all African low-income countries, including resource-dependent ones, were authoritarian.
The correlation between political regimes and resource dependence became evident after the third wave of democratization.
Natural resource abundance impacts democratic consolidation in Africa.

Theoretical Argument

Sub-Saharan African countries with abundant natural resources tend to have authoritarian governments.
Resource abundance increases competition for control of the state, leading to political violence.
Ruling parties use resource rents to maintain political power.
Resource-poor countries (Mali, Benin) have less competition for state control, favoring elite cooperation and democratic governance.
Voters in systems with high resource rents select parties based on promises to deliver rents.
Incumbents offer resource rents to gain minimum winning coalitions.
In authoritarian systems with no credible opposition (Gabon), incumbents offer resource rents to a select group for political support.
Higher resource rents translate to higher support for incumbent politicians, hindering democratic transition.
Nigeria's government became increasingly centralized and oppressive as oil revenues increased from 1% of GDP in 1960 and 30% in 1964 to over 90% after 1979 (Bienen, 1983).
Only 4 of Nigeria’s past 35 years of political history have been under civilian rule.
More than half (55%) of oil rents accrue directly to the federal government, which is responsible for distributing an additional 35% of these profits to states (Khan Ahmad, 1994).
Regional and ethnic competitions for oil revenues have contributed to Nigeria’s political system of institutionalized patronage.

Case Studies

Congo (Brazzaville): Oil industry gains importance with 1973 oil shock, Marien Ngouabi undertakes economic reforms financed by oil revenues.
After Ngouabi's assassination, Joachim Yhombi-Opango and later Denis Sassou-Nguesso take power.
Sassou-Nguesso consolidates rule with the second OPEC oil shock of 1979 and 1980 and a massive 5-year plan in 1982, increasing public expenditures to over 40% of the 1981 GDP (Tommasi, 1999).
Democratic reforms in 1991 lead to Patrice Lissouba's election, but a civil war breaks out, leading to Sassou-Nguesso's return to power.
Guinea: Over 50,000 civil servants consume over half the budget.
Botswana: Civil service employment explodes from 1964 to 1984, fueled by diamond revenues (Niemann, 1993).
Zambia: United National Independence Party uses copper revenues to generate employment favorable to the regime (Bratton, 1994).
Cameroon: Oil wealth used as information advantage and resource to buy support.
President Ahmadu Ahidjo (1960-1982) keeps oil revenues in secret bank accounts, repatriating them to finance state activities (Van de Walle, 1994).

Resource Dependence and Authoritarianism

Even in democratic systems, natural resource dependence can lead to authoritarian government by making democratic consolidation difficult.
Weak state capacity allows incumbents discretion in allocating resource rents for political support.
Opposition parties resort to nonconstitutional means (riots, coups) to combat incumbency advantage, causing political unrest.
Incumbent politicians ban or force opposition parties to merge with the ruling party.
The key mechanism is an incumbent’s discretion over the distribution of natural resource rents.
Weak political institutions allow incumbents to distribute resource rents for political gain.
Odedokun (1990) finds Nigerian states use federal allocation changes during election years in favor of consumption expenditures and against capital expenditures.
Picard (1987) links the dominance of the Botswana Democratic Party to resource wealth.
Van de Walle (1994) argues Cameroon’s “patrimonial orientation” and management of oil wealth allowed the authoritarian regime to endure.
High natural resource rents and executive discretion may lead incumbents to buy off the opposition (Nigeria, Niger).

Summary of Resource Abundance and Authoritarianism

Resource abundance leads to authoritarianism because:
- It allows an already dominant party to consolidate power, hindering democratic transition and consolidation.
- It generates incumbency advantage and political instability, inciting incumbents to adopt repressive policies toward the opposition.
- It generates extra-constitutional conflict (civil war), resulting in dictatorship.

Data and Methodology

Panel data from Robert Bates' (1999) Africa Research Program at Harvard University:
- Political data, economic data, violence data, controls for 46 sub-Saharan countries (1960-1995).
Dependent variable: Political regime type from Polity III data set (Jaggers and Gurr, 1998).
- Democracy score: Polity III democracy score minus Polity III autocracy score, rescaled by adding 10.
- Ordinal measure of regime type: 0 to 20.
Control variables from Bates’s (1999) data set:
- Log of GDP per capita.
- GDP growth.
- Number of coups.
- Number of government crises.
- Number of demonstrations.
- Number of riots.
- Number of strikes.
Measure of natural resource dependence:
- Constructed from World Bank (1999) data on fuel, mineral, and metal exports as a percentage of merchandise exports.
- Ranges from 1 to 4 (low to high resource dependence) for 46 countries.
- Classification:
  - Less than 25%: 1.
  - 25% to 50%: 2.
  - 50% to 75%: 3.
  - Greater than 75%: 4.
Cross-sectional data set for 40 African countries.
Six measures of governance from the World Bank’s Worldwide Governance Research Indicators Dataset.
- Voice and accountability.
- Political stability and lack of violence.
- Governance effectiveness.
- Regulatory framework.
- Rule of law.
- Control of corruption.
Cross-sectional data set for 46 countries to analyze the effects of natural resource dependence since democratization waves.
- Dependent variable: Democracy (1998) from Polity 98 data set (Jaggers and Gurr, 1998).
- Independent variables from Bratton and van de Walle’s (1997) Political Regimes and Regime Transitions in Africa 1910-1994:
  - Number of elections.
  - Percentage of seats held by the largest party in 1989.
  - Protest frequency.
  - Military role.
- Control for natural resource dependence: Natural resources (1990).
- Gross national product (GNP) per capita.
- Measure of past GNP growth rates.

Empirical Results

Time-series cross-sectional regressions.
Random effects (generalized least squares [GLS]) and fixed effects (ordinary least squares [OLS]) regressions used.
Hauseman test failed to reject the null hypothesis of no systematic difference between coefficients.
Models estimated using random-effects OLS regression with panel-corrected standard errors.
Model 1: Level of democracy regressed on log of GDP per capita, GDP growth, resource dependence, and decade dummies.
- High GDP per capita and economic growth have positive effects on democratic institutions.
- Higher natural resource dependence leads to lower democracy scores.
- Resource-dependent countries (Nigeria, Gabon, Zaire, Angola) are predicted to have democracy scores 1.59 lower than the least dependent countries.
- The average level of democracy was only 5.63 out of 20.
Model 2: Added former colony dummy variables.
- Colonial heritage has an effect on a country’s political regime.
- Former British colonies have the highest democracy scores; former Portuguese colonies have the lowest.
Empirical tests with different variables on protest and violence (coups, government crises, demonstrations, riots, strikes) demonstrate that even still the effect of natural resource dependence on political regimes is unchanged.

Natural Resources and Government Consumption

Countries with large natural resource endowments use a larger percentage of resource rents to maintain political power and spend more.
The difference between a resource-rich (4) and a resource-poor (1) country amounts to over 3% of the GDP.

Natural Resources and Governance

Natural resource rents are associated with worse governance, even when controlling for political regimes.
Negative relationship between measures of voice and accountability, the rule of law, and control of corruption.
Mixed results on government effectiveness, with natural resources having a negative impact but estimates are weakly significant.
Natural resources have a negative coefficient for political stability and the regulatory framework, but these estimates are not statistically significant.
Countries with large endowments of natural resources are associated with worse government performance.

Natural Resources and Democratic Breakdown

Post-third wave democratizations in Africa provide a test of the causal mechanism linking natural resource rents and democratic breakdown.
Transitions from authoritarian rule are more likely to collapse back into authoritarianism in natural resource–dependent countries.
Cross-sectional OLS and ordered probit analysis on democracy changes between 1994 and 1998.
Regressions show natural resources contributed to a decline in democracy after 1994.
Political variables (protests, electoral history) contributed to the initial wave of democratization (Bratton and van de Walle (1997).
Dependence on natural resources has a dramatic effect on the likelihood of democracy enduring.

Concluding Remarks

African rentier economies tend to generate authoritarian governments or undermine democratic governance.
Lack of transparency and executive discretion in revenue allocation affects electoral outcomes.
Natural resource dependence has a serious negative impact on democratic transition and consolidation.
From 1970 to 1995, countries with higher natural resource dependence tended to be more authoritarian.
Higher natural resources were associated with higher government consumption and worse government performance.
After the initial wave of democratization, countries with higher natural resource dependence experienced a backslide toward authoritarian rule.
The correlation between authoritarianism and resource abundance is essentially a post–third wave phenomenon.