📈 economics

Persistent Divergence and Club Formation: Long-Run Growth Trajectories in the CEMAC and WAEMU Regions (1960–2024)

This study analyzes long-run growth trajectories in CEMAC and WAEMU regions from 1960 to 2024, revealing overall economic divergence driven by resource booms and conflict, and identifying three distinct convergence clubs that necessitate targeted policies addressing structural barriers, commodity dependence, and institutional heterogeneity.

Original authors: Humphred Watard

Published 2026-06-26

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Humphred Watard

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the economies of Central and West Africa as a group of 15 runners in a very long race that started in 1960 and is still running today (2024). The goal of the race was supposed to be "convergence," meaning the slower runners were expected to speed up and catch up to the faster ones, so that by the end, everyone would be running at roughly the same pace and be close together.

This study, by Humphred Watard, checks the race results to see if that happened. Here is what the paper found, explained simply:

1. The Race Didn't End in a Tight Pack (Sigma Divergence)

If you look at the whole group of 15 runners, they are actually further apart now than when they started.

The Analogy: Imagine a group of hikers starting a trail together. Over 60 years, some hikers found a secret shortcut through a gold mine (oil-rich countries like Gabon and Equatorial Guinea) and zoomed ahead. Others got stuck in mud or had to deal with injuries and storms (conflict-ridden countries like the Central African Republic) and barely moved.
The Result: The "spread" or distance between the richest and poorest runners has actually grown wider, not narrower. The paper calls this Sigma Divergence. The gap between the top and bottom is bigger in 2024 than it was in 1960.

2. The "Catch-Up" Theory is Weak (Beta Convergence)

Economic theory often says that poorer runners should naturally run faster to catch up to the rich ones (like a heavy weight falling faster than a light one).

The Reality: The study found only a very weak hint that this is happening. While some poorer countries did grow fast, it wasn't a consistent rule. Sometimes a poor country grew fast because of a lucky oil discovery, not because it was naturally "catching up." Sometimes a rich country slowed down because of bad luck.
The Metaphor: It's like a classroom where the students who started with the lowest grades sometimes got better grades, but not always, and not because they were naturally catching up to the top students. It was too messy and inconsistent to call it a real trend.

3. The "Clubs" Formed (Club Convergence)

Since the whole group didn't converge, the study used a special tool (the Phillips and Sul algorithm) to see if the runners naturally split into smaller, tighter groups. They did. The 15 countries formed three distinct "clubs":

Club 1: The Steady Climbers (The "Diversified" Group)
- Who: Cameroon, Côte d'Ivoire, Senegal, Mali, Burkina Faso, and Benin.
- The Vibe: These runners are running together. They are catching up to each other. They rely on a mix of farming, trade, and services. Because they are similar, they are getting closer in income.
- The Metaphor: Think of a group of hikers on a well-marked trail. They are all moving at a steady pace, and the ones who started slower are slowly catching up to the leaders.
Club 2: The Rollercoaster Riders (The "Resource" Group)
- Who: Equatorial Guinea, Gabon, Congo Republic, Chad, Togo, and Guinea-Bissau.
- The Vibe: These runners are on a wild ride. When oil prices go up, they fly; when they crash, they fall. They are technically moving toward a similar destination, but the path is so bumpy and volatile that it's hard to tell if they are really catching up.
- The Metaphor: Imagine a group of people on a rollercoaster. They are all on the same ride, but the ups and downs are so extreme that sometimes they are far apart, and sometimes they are close. They are "converging" in the long run, but it's a shaky, slow convergence.
Club 3: The Stuck in the Mud (The "Conflict" Group)
- Who: Niger and the Central African Republic.
- The Vibe: These two runners are stuck in deep mud. They are moving together, but they are both stuck at a very low level. They are "converging" to a low-income trap. They are not catching up to the other clubs; they are falling further behind.
- The Metaphor: Two hikers who are stuck in a swamp. They are walking side-by-side, but they aren't making progress toward the finish line. They are converging on a state of poverty.

4. Why Can't the Clubs Merge?

The study tried to see if Club 1 and Club 2 could merge into one big happy group, or if Club 2 and Club 3 could join forces.

The Result: No. The differences between these groups are too deep. The "Diversified" group and the "Resource" group are running on totally different tracks. The "Resource" group and the "Stuck" group are separated by a canyon. They cannot be combined into a single group that moves together.

The Bottom Line

The paper concludes that the dream of all these African nations naturally catching up to each other under a shared currency (like the CFA franc) hasn't happened as a single group. Instead, the region has split into three separate teams:

A team that is successfully catching up to each other (Club 1).
A team that is moving together but is very unstable (Club 2).
A team that is stuck together in poverty (Club 3).

The paper suggests that because these groups are so different, they need different strategies to move forward, rather than treating all 15 countries as if they are on the same path.

Technical Summary: Persistent Divergence and Club Formation in CEMAC and WAEMU (1960–2024)

Problem Statement and Motivation
Despite the theoretical premise of monetary unions (CEMAC and WAEMU) and neoclassical growth models predicting that shared currencies and harmonized policies would foster convergent growth, empirical observation suggests a persistent and widening economic gap among member states. Resource-rich nations (e.g., Gabon, Equatorial Guinea) have historically outperformed agrarian or conflict-affected economies (e.g., Central African Republic, DRC). The study addresses whether six decades of regional integration have led to economic convergence or if structural heterogeneity has resulted in divergence and the formation of distinct "clubs" of economies with similar growth trajectories.

Methodology and Data
The study employs a quantitative, ex-post facto research design using an unbalanced panel of 15 French-speaking countries in Central and West Africa from 1960 to 2024.

Data Source: Annual GDP per capita in constant 2015 US dollars, sourced from the World Bank's World Development Indicators (WDI).
Analytical Approach: The analysis utilizes three primary techniques:
1. Sigma ( $\sigma$ ) Convergence: Measured by the cross-sectional standard deviation of log GDP per capita to assess the evolution of income dispersion over time.
2. Absolute Beta ( $\beta$ ) Convergence: Tested via regression of average annual growth rates against initial log GDP per capita to determine if poorer economies grew faster than richer ones across the entire panel.
3. Phillips and Sul (2007) Club Convergence: The core methodological contribution involves the application of the log- $t$ test and an iterative clustering algorithm. This approach endogenously identifies convergence clubs without imposing a priori groupings. It decomposes GDP per capita into a common growth component and a time-varying idiosyncratic element. The algorithm sequentially forms core groups, tests for convergence, and recursively clusters remaining countries, followed by merging tests to determine if adjacent clubs can be combined.
Software: Analysis was conducted using the ConvergenceClubs package in R.

Key Results

Overall Divergence ( $\sigma$ -convergence): The study finds strong evidence of $\sigma$ -divergence. The standard deviation of log GDP per capita increased from 0.95 in 1960 to 1.25 in 2024. This divergence is driven by resource booms in oil-exporting nations versus stagnation in conflict-affected states.
Weak Absolute Convergence ( $\beta$ -convergence): The regression for the full panel yields a negative coefficient ( $\beta = -0.307$ ) that is only marginally significant ( $p = 0.064$ ). The low $R^2$ (0.24) indicates that initial income levels explain only a small fraction of growth variation, suggesting that unconditional convergence is weak across the region.
Identification of Three Convergence Clubs: The Phillips and Sul log- $t$ $t$ test rejects full panel convergence ( $\hat{\gamma} = -0.472, p < 0.001$ $\overset{γ}{^} = - 0.472, p < 0.001$ ) and identifies three distinct clubs:
- Club 1 (High-Income/Diversified): Comprising Cameroon, Côte d'Ivoire, Senegal, Mali, Burkina Faso, and Benin. This group exhibits significant within-club convergence ( $\hat{\gamma} = -0.198, p = 0.092$ ) and strong absolute $\beta$ -convergence ( $\beta = -0.68, p = 0.031$ ). These economies share structural characteristics such as diversified agriculture and relatively stable institutions.
- Club 2 (Middle-Income/Resource-Driven): Comprising Equatorial Guinea, Gabon, Congo Republic, Chad, Togo, and Guinea-Bissau. The log- $t$ test accepts convergence ( $\hat{\gamma} = -0.087$ ), though the speed is slow and within-club $\beta$ -convergence is statistically insignificant ( $p > 0.10$ ). This group is characterized by high volatility due to commodity dependence and political instability.
- Club 3 (Low-Income/Stagnation): Comprising Niger and the Central African Republic. These countries converge to a common low-income equilibrium ( $\hat{\gamma} = 0.152$ ), representing a "poverty trap" characterized by chronic insecurity and lack of structural transformation.
Rejection of Merging: Statistical tests confirm that adjacent clubs cannot be merged, validating the existence of three distinct, non-converging growth paths.

Significance and Contributions
The paper claims significance in three specific areas:

Policy Relevance: The findings challenge the assumption that monetary integration alone guarantees convergence. The study argues for targeted, club-specific policies: diversification and human capital investment for Club 1; resource rent management and fiscal rules for Club 2; and conflict resolution and infrastructure development for Club 3.
Theoretical Contribution: By applying the Phillips and Sul (2007) methodology to this specific region, the study provides empirical support for the club convergence hypothesis (Galor, 1996; Durlauf & Johnson, 1995) in Sub-Saharan Africa, demonstrating that economies converge to multiple local steady-states rather than a single global one.
Methodological Value: The study demonstrates the utility of the log- $t$ clustering algorithm in detecting transitional behaviors and time-varying heterogeneity that traditional cross-sectional regressions might miss, particularly in volatile, commodity-dependent economies.

Limitations
The authors acknowledge limitations, including the reliance on unconditional convergence due to data constraints over the 64-year period, the exclusion of conditional factors (e.g., education, institutional quality) in the primary regression, and the focus on GDP per capita rather than consumption metrics. Future research is suggested to incorporate these controls and non-income indicators.

1. The Race Didn't End in a Tight Pack (Sigma Divergence)

2. The "Catch-Up" Theory is Weak (Beta Convergence)

3. The "Clubs" Formed (Club Convergence)

4. Why Can't the Clubs Merge?

The Bottom Line

More like this