Reform delays emerge frequently in politics but can be solved using compensation payments. A higher degree of fragmentation among the addressees may- according to the theory- reduce these costs. The number of groups and the inherent uncertainty normally influence agents' behaviour. When this prediction holds and behaviour is in fact influenced by the number of groups, fragmentation will not outperform a less fragmented society concerning compensation costs. An experiment is conducted to evaluate the effects of fragmentation on agents' decisions and shows that the theoretic result as such cannot be applied to the behaviour of agents.