top of page

Who Prefers Democratic Regimes in Africa?

Louis Tomavo

Democracy in Africa in recent years has undergone a series of questions and public debates. While the development and progress of democracy is celebrated at the international level, coups d’etat are becoming more and more recurrent in African countries. In case after case, governments resulting from democratic elections are rejected by a fringe of the population through action by the military. Freedom House (2018) notes a significant decline in political rights, civil liberties, and the overall quality of democracy. Afrobarometer data, however, shows that some 70% of Africans still express a preference for democracy. Who are they, and why do they continue to support democratic government while 30% do not?


As Professor Abdoulaye Bathily pointed out to the ECOWAS parliament, democratization as a process reflects extremely contradictory demands. African systems are based on the rejection of the authoritarianism of the “nationalist” regimes that emerged from decolonization; but at the same time, they express a revolt against the economic and social effects of the liberal policies implemented by structural adjustments. Public opinion on democracy in Africa is therefore still embodied by divergences. The responses from the questions asked by Afrobarometer surveyors during Round 8 in the countries covered by the network present a more varied and complex picture.


In this essay, we draw on these surveys to highlight the profile of people who prefer democracy to any other form of governance in Africa. We first present the frequency of the latter in the consolidated database of the Afrobarometer Round 8 in Africa and then proceed to the estimation by the Logit model to measure the explanatory variables. We explore several hypotheses:

H1: Unemployment may be a source of support for non-democratic regimes.


H2: The performance of the president and the economic performance of the country could lead to a rejection of a democratic regime.

H3: Support for democratic rule depends on certain demographic criteria.



To conduct our study that aims to see the profile of people who support democracy or not in African countries, Afrobarometer Round 8 data is used. Afrobarometer Round 8 was collected between 2019 and 2021 with over 48,000 adult respondents in 34 countries. The data produces a result with a marginal error of +/- 3% and a confidence interval of 97%.


Table 1 below describes the dependent variable for which the paper aims to provide a profile of respondents who prefer a democratic government. It turns out that about 7 out of 10 Africans support democracy. However, it also appears that about 3 out of 10 people don’t support democracy. In other words, they approve a non-democratic government under certain circumstances or do not care which regime governs their country.

Question to respondents: Which of these three statements is closest to your opinion?


The objective of our work is to provide a plausible explanation for those who choose a democratic regime even when democracy appears to be one the decline globally. In the following section, this variable will be subdivided into two options. The first is support for democracy, which considers the fact that democracy is preferable to any type of government, and the second considers the last two options for answering the above question. These are referred to in this paper as non-support for democracy.


Questions asked of respondents: Which of these three statements is closest to your opinion?

  • Statement 1: Democracy is preferable to any other form of government.

  • Statement 2: Non-democratic government may be preferable in some circumstances.

  • Statement 3: For someone like me, it doesn’t matter what kind of government we have


In general, the average African supports democracy. However, there is a 53-point gap between the countries with the highest and lowest levels of support. People living in Ethiopia, Senegal, Sierra Leone, Zambia, etc. have a strong preference for democracy; but it is also true that less than half of all citizens prefer democracy in Angola, South Africa, Mozambique.


Overview of the variables that explain the preference for democracy

The purpose of this section is to briefly describe the variables that can explain the preference for democracy in Africa. It is expressed through the socio-demographic characteristics of the people, their satisfaction with the performance of the president and his management of the economy but also through the unemployment situation in our countries.


Support for democracy according to socio-demographic factors

One possible explanation is gender (Table 2). The data shows that nearly 7 out of 10 men (71%) prefer democracy to all other forms of governance, unlike women, whose rate is lower (66%). However, young people, particularly those between the ages of 18 and 35, have a lower level of preference for democracy (67%), which should attract the attention of leaders, since youth is the lever for Africa’s future development. Also, the poorer people are, the more they believe that another type of government other than democratic can improve their living conditions (67%).

Support for democracy/satisfaction with

democracy: the data suggests strongly

that the more the population is

satisfied with the functioning of

democracy (74%), the more they are

inclined to a preference for democracy

in their country. (See Graph 2.)

Supporting democracy and creating jobs:

despite a high rate (67%) of preference for

democracy, the level of unemployment

follows a high rate (69%), and these people

may in the long term want a non-democratic

regime. But, as Graph 3 shows, this analysis

showed no significant differences in support

for democracy.

Support for democracy and President’s

performance: the president’s performance in

office could also be an indicator of support

for democracy, but the data again do not show

any significant differences in support for

democracy based on satisfaction with the

President’s performance.

Study Variables

The dependent variable in this study is support for democracy. We examined a set of variables. based on the literature which might explain preference for democracy in Africa: presidential performance, satisfaction with democracy, job creation, management of the economy, poverty, gender, education, and age. The analysis in (Table 3) shows that there is a dependence between all the explanatory variables and the choice of democracy (Pvalue less than 5%).


Factors explaining the preference for democracy

This section aims to establish a plausible link between attitudes toward the performance of the president and support for a democratic government. In addition to satisfaction with the functioning of democracy, we considered management of the economy, insecurity issues, and so forth as potential factors in the choice of a democratic government or not. To further investigate this study, we used the econometric model with binary qualitative Logit variables (see Appendix 1 for the model.)



The search for explanatory factors for support for democracy in African countries led us to estimate using a binary (Logit) model. To achieve the objective, four regressions were performed. The first was for President’s Performance, while the second considered the effect of Satisfaction with democracy on support for democracy. The third regression explored the impact of Job Creating on support for democracy and then the last one considered Management of the Economy and Freedom of Speech. Finally, demographic variables such as gender, experienced poverty, and education were used as control variables to better assess the explanatory factors of support for democracy. The results show that the estimated model is globally significant and well-specified with all variables considered. The analysis of the results reveals that the performance of presidents in managing the city has a significant effect at 1% on support for democracy. Indeed, the good performance of the president of the Republic leads about 2% of the adult population to support democracy. Similarly, a better living condition of the people ensures greater support for democracy. Controlling for demographic characteristics, the results indicate that the president’s performance leads women to be 3.4% more likely to prefer democracy to any other form of government than men. Finally, when the president’s performance is good across all African countries, people often prefer democracy in general and 0.3% of the fluctuations in the probability of support for democracy are explained by the president’s performance in running the city. Satisfaction with democracy being a fundamental element of support for democracy, the results of our estimations show that the more people are satisfied with the functioning of democracy, the more they are inclined to support a democratic regime (1-0.086 = 0.914, i.e., nearly 92%). The level of poverty and the level of education are elements that explain this phenomenon: with satisfaction with the functioning of democracy, more than 99% of the population supports the democratic regime regardless of their standard of living or their level of education. It is also noticed that the good functioning of democracy makes women 3% more likely to want to support a democratic regime than men. Countries like Senegal, Sierra Leone, Benin, Botswana among others have a very significant threshold of satisfaction with democracy (1%) unlike South Africa, Mozambique, or Tunisia among others which present a less significant threshold (10% approximately) of satisfaction with democracy.


The issue of job creation does not necessarily translate into lower unemployment. Everything depends on the balance between entries into the labor market (especially young people who have finished their studies) and exits (especially people retiring). The unemployment rate will only fall if job creation is sufficient to absorb an increase in the labor force. Our estimation results show that if the leaders of our states provide job opportunities, more than 85 percent of the population will tend to support a democratic regime. This does not change the propensity of female to support democracy relative to male.


As far as the management of the economy is concerned, the more the country’s resources are well allocated to sustainable development actions, the more the population feels better in their daily lives and therefore supports democracy. The better management of the economy is, the most 83% of the population prefers the democratic regime. The support of women remains constant even with a perception of better management of the economy. Since we cannot speak of democracy without freedom of expression, we have made a comparison between these notions, and we must admit that freedom of expression, free choice to join a political organization of one’s choice or the freedom to choose without pressure the person to vote for have a significant (1%) on the support to a democratic regime in our countries.


At the end of these different findings, we can affirm that our hypotheses are validated; in other words, the performance of the president, the economic performance of the country as well as the issue of job creation are prerequisites for supporting democracy in our countries, as well as the demographic variables help explain this phenomenon.



From the results of our study, we note both positive and negative advances on the state of democracy in Africa. In most African countries, support for democracy remains an important element in the eyes of the average population in most countries. Beyond the continental averages, pro-democracy attitudes are widespread in some countries (Angola, South Africa, Mozambique) but dangerously sporadic in others.


The leaders of the African states must also take the well-being of the youth into account because their development is a guarantee for the development of our continent. But a question remains: is democracy compatible with the African context?

Appendix 1: Logit Regression Model

Screen Shot 2023-11-09 at 4.54.57 PM.png
Screen Shot 2023-11-09 at 4.59.33 PM.png
Screen Shot 2023-11-09 at 5.04.57 PM.png
Screen Shot 2023-11-09 at 5.10.27 PM.png
Screen Shot 2023-11-09 at 5.18.31 PM.png
Screen Shot 2023-11-10 at 11.58.44 PM.png
bottom of page