Forced recruitment has been found to retain members longer than voluntary recruitment. This raises a puzzle, both as coercion is costly, but also for allegiance. In most settings, tactics are dynamic over the course of conflict. Yet, prior literature has assumed these as static. Allegiance is further highlighted in prior research as key to success in forced rebel recruitment, but has rarely been measured as an outcome. This thesis seeks to contribute to these gaps in a twofold way, by theorising that a combination of physical and psychological coercive tactics in forced rebel recruitment affect the level of allegiance across rebel groups. The examined rebel groups are Lord’s Resistance Army in Uganda (LRA) and Mozambique’s National Resistance (RENAMO) in Mozambique. This is conducted by a qualitative, small-N study with a comparative case study method. The theory and hypothesis found some support in the empirical findings even though some discrepancies were identified. In general, a high and intense combination of physical and psychological coercive tactics were found to cause high levels of allegiance in LRA, whereas a lower and less intense combination were found to cause moderate levels of allegiance in RENAMO. Both relationships were reinforced by causal mechanisms of social identification.