Recent research suggests that co-optation at the elite level may be a key factor when trying to understand the stability of autocratic regimes. This paper seeks to contribute to these debates by arguing that the stability of autocratic regimes is determined by the inclusiveness of their ‘elite bargain’, i.e. the distribution of access to positions of state power (political, military, economic and territorial) between the country’s most salient social groups. I hypothesise that the co-optation of all relevant group leaders will favour regime stability and peace, whereas the exclusion of relevant leaders will lead to instability and violence. This hypothesis is tested by studying two autocratic regimes in Sub-Saharan Africa, namely Zambia under Kenneth Kaunda (1964-1991) and Uganda under Milton Obote and Idi Amin (1962-1986). I find that the relative stability of Kaunda’s autocratic regime was indeed facilitated by systematic efforts for ‘tribal balancing’ between the country’s main linguistic and ethnic groups. At the same time, the failure to effectively co-opt class interests in guise of the powerful trade unions ultimately led to the downfall of the Kaunda regime. In Uganda, the Obote and Amin regimes failed to effectively co-opt and accommodate ethnic, regional and religious interests. This resulted in high levels of regime instability, evident in recurrent coups and civil war. My findings are based on 103 interviews, a comprehensive set of original data on the inter-group distribution of political, military, economic and territorial posts, and in-depth historical analysis.
