Download this publication

How to cite this publication:

Gretchen Bauer, Akosua K. Darkwah (2024). The Cost of Doing Politics in Ghana: What does violence against politicians look like? Bergen: Chr. Michelsen Institute (CMI Brief 2024:01)

Ghana is characterized by highly partisan politics with the prospects for electoral violence exacerbated by a winner takes all electoral system. Candidates for political office routinely experience some form of political violence often perpetrated by their own and other political parties during primary and general election campaigns.

Key takeaways

  • Nearly all candidates for parliament in Ghana (95 percent) experience some form of political violence, namely, degrading talk or false rumors
  • Physical violence against both men and women is much less common
  • A significant amount of political violence is perpetrated by members and leaders of candidates’ own parties or other parties
  • Political violence is gendered in its form and scope with women more likely to experience sexual violence and more likely to be victimized by members of either their own party or other parties
  • Political violence occurs mostly during the primary and general election campaigns
  • Only about half of incidents of political violence are reported to political parties, with parties taking action on just over half of these reports
  • Women are largely not deterred from standing for political office by political violence despite suffering somewhat more than men from political violence

Context: the Ghana case study

In 2020, Ghana undertook its eighth round of parliamentary and presidential elections since its transition to democracy in 1992. Operating under the candidate-centered single member district (SMD), plurality/majority electoral system, elections in Ghana are at risk for some level of political violence. The risk stems from the focus on individual candidates and the fact that election outcomes under majoritarian electoral systems are ‘winner takes all’ for the winning candidates and winning parties, thus raising the stakes of electoral competition. Plurality/majority electoral systems also tend to result in strong two-party political systems and that has been the case in Ghana since 1992, with the rivalry between the two main political parties – the National Democratic Congress (NDC) and New Patriotic Party (NPP) – intensifying over time, such that politics in Ghana today are highly partisan. Indeed, electoral politics in Ghana today may even be perceived as ‘zero-sum’ contests – with the winner taking all and the loser getting nothing. For the winners, a positive electoral outcome implies not only dominance of the state but access to sources of patronage. And the longer a political party remains out of power, the worse will likely be the potential future fortunes of that party. As a result, scholars have suggested, parties may resort to dispensing patronage goods, but also hate speech, fear and panic, and violence during elections, in an attempt to influence the election outcome. It may be argued that the zero-sum nature of electoral politics in Ghana has been reflected in the harsh language of campaign slogans previewed in the years leading up to the 2024 elections: ‘break the eight’ [pattern of eight years only in office] for the ruling NPP and ‘do or die’ [win – or lose everything] for the opposition NDC.

Another impact of Ghana’s single member district electoral system – and no electoral gender quota – has been one of the lowest representations of women in parliament – in Africa and the world – since the political transition in 1992. A women-in-politics literature on Ghana identifies the SMD electoral system, no gender quota, the financial cost of standing for political office, a politics of insult targeted especially at women, and a concern about the weakness of the National Assembly as some of the factors keeping women from standing for political office in Ghana. This brief also reports the potential impact of political violence on women’s participation in electoral politics in Ghana.

The survey

This brief presents findings from surveys conducted on Ghana’s 2020 parliamentary elections. The population for the survey consisted of all those who stood as candidates for parliament in the 2020 election., including the 550 individuals who stood as candidates for one of the two main political parties, the NDC or the NPP (one from each party in 275 constituencies), as well as a small number of candidates from smaller parties distributed over the constituencies. The survey asked respondents about whether they had experienced a range of different forms of political violence, whether they had anticipated such violence, the extent to which the violence was sexually connoted, the sources and timing of the violence, political party responses to the violence, and the effects of the violence on their political plans and behavior.

Gender, age, education, and political party

In total, we surveyed 192 members of parliament, 167 men and 25 women, and 221 (unsuccessful) candidates, 202 men and 19 women – for a total of about 10 percent women respondents from the two groups combined. Only 14.5 percent of MPs and under 14 percent of candidates overall were women. On average, members of parliament in Ghana are not as old as legislators in some other parts of the world, but they are still typically much older than the majority of Ghanaians. Only 8 percent of MPs and 12.5 percent of candidates are below 40 years of age even though the median age of Ghanaians in 2023 was 20.3 years. Members of parliament in Ghana, like their counterparts around the world, are highly educated and much more so than their fellow Ghanaians. So, 99.5 percent of MPs and 87.5 percent of candidates have a university education as compared to a Ghanaian population in which 13.4 percent are university educated, as per the 2021 census. Since the 2020 election there are only NPP or NDC members of parliament; there were candidates from a few smaller parties, of whom we surveyed a handful. Between the two major parties, there was a 60/40 NDC/NPP split of respondents during both phases of the surveys, and for both men and women.

Findings: what does political violence look like in Ghana?

Scope and form of political violence

Some forms of political violence are pervasive in Ghana. For example, as shown in the table, about 95 percent of respondents experienced degrading talk and false rumors as candidates. By contrast, about half of respondents experienced threats in relation to their political roles while an even smaller percentage experienced physical violence or damage to personal property.

Forms of Violence

Not Elected


Man %

Woman %

Man %

Woman %

Experienced degrading talk or false rumors in relation to political role





Experienced threats in relation to political role





Personally experienced physical violence in relation to political role





Experienced damage to personal or party/government property in relation to political role





Some persons associated with the respondent experienced intimidation in relation to political role





Sources and timing of political violence

The sources of political violence are varied. Nonetheless, our findings show that most of the violence is perpetrated by members of candidates’ own political party or the other parties. Almost 75 percent of our respondents had experienced violence perpetrated by members of their own parties and 85 percent had experienced violence from members of other parties.

There was a clear pattern to the experience of violence in terms of the timing. With reference to physical violence, in particular, the general election campaign was the most violence-prone period followed by the nomination phase (party primaries) and election day. Once an individual was elected, however, they were not likely to experience much physical violence. As an MP then, an individual could focus on the work at hand without having to worry about the physical violence of the primary or general election campaigns and election day that led them to their current roles.

Political party response (to reports of violence)

The vast majority of respondents (97.3 percent) were of the view that parties had ethical guidelines with only 1.2 percent unsure if they did and 1.4 percent of the view that they did not have guidelines. Although almost all the respondents believed that their parties had guidelines regarding behavior, only about half of all incidents were reported to the party organization for redress. In the cases where an incident was reported, the party took action in about 60 percent of the cases. Overall, then, only about 25 percent of cases were being reported and addressed in one way or the other by the political parties. Although there were no significant gender differences, the parties seem to be a bit more responsive to women’s reports of violence.

Gendered differences

There is a gendered dimension to the experience of violence; half of the women (50 percent) as compared to about a quarter of the men (27 percent) had experienced any sexually connoted violence. Another element of the gendered violence was that it extended to family members as well. Significantly more women (80 percent) than men (62 percent) indicated that someone close to them experienced violence.

In addition to there being a gendered element to the experience of violence, there were gender differences in the sources of violence. Party members were often perpetrators of violence, but women were much more likely to find that either members of their own party or members of other parties were the perpetrators of the violence they experienced. Four-fifths of women (83 percent) compared to two-thirds of men (65 percent) had experienced violence from within their own parties. The numbers were slightly higher for perpetrators from outside their own parties, but the gender difference still persisted; 91.5 percent of women as compared to 78 percent of men had experienced violence perpetrated by members of other parties.

Although there were gender differences in the experience of violence and the perpetrators of violence, there were no gender differences in terms of the timing of the violent incidents.

Impact of violence

Although both men and women respondents experienced different forms of violence, it would not deter them from seeking to stand again, they asserted. Overall, very few of our respondents expressed an unwillingness to run again, with the majority stating their willingness to run again. With respect to a willingness to speak on certain controversial issues, a quarter of respondents (both male and female) were undeterred from speaking out despite experiencing some form of political violence.

Conclusion: What can be done?

Overall, survey respondents conveyed a sense of a ‘normalization’ of political violence on the campaign trail, especially the degrading talk and false rumors which are rampant for all aspirants and candidates and more vicious for women than for men. However, despite some high-profile cases of political violence during the 2020 elections, the overall conclusion based on the survey reveals that political violence is not an intractable problem in Ghana.

Several measures that can mitigate the disruptive potential of political violence have been attempted. And yet, for the most part, candidates do not see the efforts of national stakeholders like the Electoral Commission, the National Commission for Civic Education or the National Peace Council as effective. Outside observers have recommended backchannel negotiations for Ghana as well as the timely identification of potential hotspots, early monitoring and observation, and the use of conflict resolution mechanisms. Suggestions from candidates themselves for mitigating violence during elections include better training and less partisanship from security services, candidate pledges to avoid political violence during their respective campaigns, enhanced civic education across the board in the country, calls for peaceful elections from pastors and imams, and football matches for the youth allied to the respective parties (who might otherwise engage in vigilante activities) - with candidates in attendance to reinforce the message about avoiding political violence during elections. Despite the high stakes of Ghana’s winner takes all electoral system, candidates for political office do not want to see the quality of elections – or of Ghana’s democracy – undermined by political violence during election campaigns or on election day.

We would like to thank Mr George Asare from GABSResearch, LLC, Accra, for his tireless work on our behalf, as well as several University of Ghana students who served as enumerators for the surveys on which this brief is based. We would also like to acknowledge Mr Jude Holali Sablah for assistance with tables.


Asante, Richard, and Ransford E. Gyampo. 2021. Explaining Political Party Vigilantism and Violence in Parliamentary By-Elections in Ghana. Journal of African Politics 1: 89-105.

Bauer, Gretchen, and Akosua K. Darkwah. 2020. ‘We Would Rather be Leaders Than Parliamentarians’: Women and Political Office in Ghana. European Journal of Politics and Gender 3, 1: 101-119.

Bekoe, Dorina, and Stephanie Burchard. 2021. Robust Electoral Violence Prevention: An Example from Ghana, African Affairs 120, 481: 543-567.

Fjelde, Hanne, and Kristine Hoglund. 2016. Electoral Institutions and Electoral Violence in Sub-Saharan Africa. BJPolS 46: 297-320.

Ichino, Nahomi, and Noah Nathan. 2012. Primaries on Demand? Intra-party Politics and Nominations in Ghana. BJPolS 42: 769-791.

Kuyini Mohammed, Abdulai. 2023. An Evaluation of the Quality of Democracy of Ghana. Democracy and Security 19, 4: 347-388.