Democratic backsliding is a gradual erosion and hollowing out of democratic rules, institutions, rights, and accountability. Instead of a sudden collapse, it often happens when elected leaders manipulate legal systems, undermine electoral agencies, constrain opposition parties, repress civil society and restrict civil liberties. Therefore, it is not just about whether there are elections but about whether the broader democratic system supports their importance.
The International Foundation for Electoral Systems describes democratic backsliding as occurring in three stages: democratic erosion, democratic breakdown, and autocratic deepening. Erosion is more difficult to identify because it can resemble routine legal and political reform even as it eats away at accountability.
Breakdowns are more visible and occur more quickly, involving actions such as shutting down media outlets, cancelling the registration of opposition parties, removing officials, or suspending constitutional rule. Autocratic deepening is when authoritarian or hybrid regimes get more repressive, further shutting the door on democracy by imprisoning journalists, repressing opposition figures, and using courts to repress dissent.
These processes can take different forms depending on the type of regime. In more established democracies, backsliding may take the form of corruption, attacks on independent institutions, or a decline in civil liberties from a strong democratic baseline.
Hybrid regimes may have democratic institutions in name, such as constitutions, elections, courts and parliaments, but these institutions operate with little independence or accountability. In more authoritarian contexts, the problem might be less the loss of democracy than the strengthening of autocracy.
Norwegian research institute CMI notes that democratic decline in Africa is typically gradual and takes place within regimes, as democracies become less liberal and autocracies become less competitive and more repressive.
This raises a conceptual question: can the term “democratic backsliding” capture the political conditions and diverse democratic trajectories of African countries, or does it risk overlooking cases where democracy was never truly consolidated in the first place? The idea of democratic backsliding rests on the underlying assumption that democracy was previously well established, though this may not always be the case.
In several African contexts, the patterns we observe are more characterised by regional instability, stagnation, or fluctuation, which are frequently explained by specific local circumstances rather than by a consistent decline in democracy across the continent.
Consequently, when democracy is primarily seen as formal or procedural rather than genuinely meaningful, events such as the weakening of institutions, the diminishing of civil freedoms, low civic engagement, and coups may signal both democratic collapse and the limitations of democratic consolidation.
In the African context, democracy is much more nuanced than dominant political theories can describe, which brings into question whether the recent decline in democratic institutions and systems can be described as part of the wider democratic backsliding many parts of the world are facing.
To examine this phenomenon and its applicability in the African context, we set out to explore the current state of democratic institutions across the continent and, if they are declining, where they are declining.
We first divided the most prevalent aspects of democracy into four main categories: the extent of civil rights and liberties, the quality of electoral institutions, the degree of participation in democratic processes, and the quality of liberal institutions, including the rule of law.
These dimensions are important because the weakening or collapse of systems required to sustain democracy can reduce the quality of democratic rule, with consequences for rights, stability and development.
Based on these sets of democratic characteristics, we analysed data from several sources that measure them across the African continent.
Table 1 provides an overview of the data sources and variables used in our analysis for each of the five features of democracy.
Table 1: Data sources used in analysis:
| Democratic Characteristic | Data Source (Year) | Variable used (Code from source) | Variable Definition | Scale | Range |
| Quality of liberal institutions | V-Dem (2026) | Liberal Component Index (v2x_liberal) | Measures the extent to which liberal democracy is achieved through a constitution, rule of law, and independent judiciary. | 0-1 scale, with 1 being a perfect system of accountability through elections. | 2015-2025 |
| Quality of electoral institutions | V-Dem (2026) | Electoral Component Index (v2x_EDcomp_thick) | It measures the extent to which elections ensure accountability and action between leaders and citizens. It considers the level of suffrage, NGO freedoms, the quality of elections and the extent to which elections decide the executive leader. | 0-1 scale, with 1 being perfect citizen involvement in democratic systems | 2015-2025 |
| Extent of participation within democratic processes | V-Dem (2026) | It measures how free people are in terms of their civil rights. It considers freedom of expression, associational and organisational rights, the rule of law and individual rights. | Participatory Component Index (v2x_particip) | 0-1 scale with 1 being perfect citizen involvement in democratic systems | 2015-2025 |
| Extent of civil rights and liberties | Freedom House’s Freedom in the World Data (2026) | Civil Liberties (CL) | It measures how free people are in terms of their civil rights. It considers freedom of expression, associational and organizational rights, rule of law and individual rights. | 0-60 scale | 2015-2025 |
Our analysis focuses on the 10-year period between 2015 and 2025. This is because even though most scholars agree that the democratic backsliding phenomenon started to emerge in the late 2000s. It was only in 2017 that this phenomenon became more prevalent in Africa. By looking at data from 2015 to 2025, we can better understand whether these increases are due to democratic backsliding or just part of the democratic experience.
Figure 1 presents the first democratic element that we analysed, the quality of liberal institutions from V-Dem. The graphic shows the 2025 scores relative to the 2015 scores, with the absolute change for each country shown. Southern and western Africa tend to have the strongest liberal institutions, with means of 0.7 and 0.5, respectively. However, in the case of the former, the data suggest that the quality of liberal institutions is stagnant, with some of the region’s major democracies, such as South Africa and Namibia, not improving their institutions over the 10-year period.
As such, while a mean of 0.7 suggests stronger democracies, it is not a good sign to see no increase over a longer period. Western Africa, with an average of 0.5, shows quite strong signs of institutional collapse across democratic institutions. Countries like Burkina Faso, Benin, Niger and Guinea-Bissau have seen significant decreases in the quality of liberal institutions. Middle Africa is quite interesting because it has significant declines, including in the Central African Republic.
However, the region has the lowest institutional quality score, indicating that liberal institutions have not fully consolidated there. Overall, these findings show that different regions exhibit different trends, with some, such as western Africa, showing widespread decreases, while others, such as eastern Africa, are more varied, with some declines and some increases, but nothing as significant.
However, considering the quality of liberal institutions on its own does not provide an indication of whether these regressions are systemic. To do that, we also considered the quality of electoral institutions and the extent of participation as depicted in Figures 2 and 3.
Figure 1: Liberal institutions index ccore across African countries (2015-2025).
Firstly, in terms of electoral institutions shown in Figure 2, western Africa again stands out as a key region of concern, with only Gambia making significant improvements. Countries that experienced a coup in western Africa are seen to have experienced significant decreases in electoral quality. This makes sense, as the transfer of power shifted from elections to military coups, which would then lead to a significant drop in those institutions.
Tunisia also stands out again as an outlier in North Africa, with the 2021 coup having significant consequences on the quality of electoral institutions. Southern Africa has the highest mean for the quality of electoral institutions. However, there are no significant increases or decreases; Botswana and Namibia saw some decreases, though they’re quite small, dropping by less than 0.1 points over 10 years. Eastern Africa is also quite consistent, with most falling quite close to the mean and some outliers. However, even then, the outliers are not much less or more than the mean.
Figure 2: Electoral quality Iindex score across African countries (2015-2025).
Secondly, regarding the extent of participation in the democracy variable shown in Figure 3, southern Africa stands out as one of the most concerning regions, with declines across the region except in Lesotho. This is concerning because southern Africa has some of the strongest democracies, and seeing such declines raises questions around how the growing lack of participation will affect the other characteristics. Interestingly, western Africa has the highest mean in 2025, but it is quite widespread, with some countries showing smaller increases or decreases and others showing greater ones.
One thing to note is that, unlike the other characteristics, having a coup has not seen a drastic decrease in the participation score. This might suggest that participation in these regions is quite strong regardless of other political factors. Middle and northern Africa have similar means and spreads, with countries like Libya and Cameroon showing greater decreases, while countries like Tunisia and Angola show smaller changes over the 10-year period.
Figure 3: Participatory democracy index score across African countries (2015-2025).
Lastly, the civil liberties depicted in Figure 4 are based on data from Freedom House, which rates countries on a scale of 0-60 for the extent and quality of their civil liberties. Southern Africa is quite consistent with most countries seeing small declines, except for Lesotho. However, it is important to note that across southern Africa, countries are declining in civil liberties, albeit at an incremental rate.
Western Africa follows similar patterns to those of the liberal and electoral institutions variables: countries that had a military coup in the past 10 years tended to decrease, and in some instances, such as Burkina Faso, the decrease was quite significant. Eastern Africa is also diverse, with its points spread out over a wide area. Tanzania stands out as having the largest decrease in the region, dropping by 11 points over the 10-year period.
Figure 4: Civil liberties index score across African countries (2015-2025).
Each of these variables offers a compelling overview of how democracy is faring across the continent, but what do they mean for democratic backsliding? At face value, one could argue that the various characteristics give convincing evidence for democratic backsliding across Africa. However, when we look at these figures with more nuance and context, we begin to question whether democratic backsliding is the most appropriate explanation for what is happening. Africa is arguably more nuanced and complex when it comes to democracy. Many countries are considered democracies on paper, but are not democratic. This question is whether one can apply the term ‘democratic backsliding’ to countries that are not democratic.
Figure 5 provides an overview of regime type across Africa from 1999 to 2025. The regimes are based on V-Dem categories and split into liberal democracy, electoral democracy, electoral autocracy, and closed autocracy. Most African countries fall into the electoral autocracy category, meaning they have many of the institutions and practices of a democratic state, but they are not substantive and exist only on paper. Between 2009 and 2017, Africa experienced a period of democratisation with an increase in electoral and liberal democracies.
Figure 5: Regime transitions in Africa (1999-2025).
However, by 2017, this number began to decline, accompanied by an increase in electoral autocracies. Given the number of autocracies in Africa, there is an argument to be made that describing the growing number of autocracies as democratic backsliding misses much of the nuance behind the numbers. For example, in western Africa, the data above shows that countries that experienced coups generally saw quite significant decreases. However, this does not mean there was democratic backsliding; rather, it indicates full regime change followed by autocratic consolidation, as regime change usually halts these institutions, which would explain the massive decreases the data shows. Such nuances are missed when using the term ‘democratic backsliding’ to explain the decline.
Furthermore, even when considering countries that did not experience regime change, the data cannot be fully explained by democratic backsliding. For example, in Zimbabwe, which saw decreases across the variables, one could argue that there is democratic backsliding. However, since the 1990s, Zimbabwe has been considered an electoral autocracy, suggesting that the decreases observed are not necessarily democratic backsliding but rather the consolidation of an autocratic regime. By not making this distinction, it raises the question of where the line lies between a country’s political trajectory being considered democratic backsliding or autocratic consolidation.
This is not to say that ‘democratic backsliding’ should not be used in the African context, but there needs to be more discourse on how these terms are used to shape perceptions of the continent and whether they help explain these social and political phenomena.
Stuart Morrison is a Data Analyst with the Governance Insights and Analytics team at GGA. His research and expertise mainly focuses on the nexus between local governance, urbanisation and elections. He has a MA in E-science (Data Science) at the University of Witwatersrand, Johannesburg, with a background in Political Science, International Relations and Development studies. His multi-disciplinary approach incorporating data science and quantitative methods allows him to provide a nuanced and data-driven approach towards his research and policy work.
Owami is a junior data analyst intern in the Governance Insights and Analytics programme at Good Governance Africa. She is completing a Master’s degree in e-Science at the University of the Witwatersrand, building on a foundation in International Relations and Political Science. Her research interests focus on political economy with a particular focus on development and elections. She is especially interested in the intersection of governance, economic development, and institutional quality.
