Membership, stability and internal institutions in European cartels

The first topic this thesis examines is that of firms which enter and leave cartels without \naffecting the existence of the cartel. The first chapter predicts which firms in markets \nwill choose to join and leave cartels. The findings align with a group of theoretical \nmodels identified in the literature review, indicating that cartel membership is explained \nby�firms' individual preferences for collusion, which are consistent over time. The firms \nmost likely to join and leave cartels are small firms in large cartels. \nThe second chapter questions what effect this behaviour by firms has on the survivability \nof cartels. Theory is ambiguous on this, since entry and exit by firms could signal poor \ndiscipline among cartelists which prevents the cartel from raising prices substantially due \nto undercutting by outsiders or it could signal a structurally stable cartel which marginal firms take advantage of in their membership decisions. Cartels which experienced more \nentry and exit by firms had a lower risk of breakdown in each period than cartels \nwith more static membership, indicating that member firms recognise when cartels are \nstrong and take advantage of this by constantly re-evaluating their membership decisions. \nThe�final chapter discusses a different topic: the types of agreement formed by cartels. \nAll cartels must agree to either �x prices, restrict the output of its members, allocate \nexclusive territories, allocation customers, or rig bids in order to fulfil their objective \nof raising member profits. Many cartels engage in more than one of these practices \nsimultaneously. Structural variables are poor at predicting the presence of agreement \ntypes in the cartels studied, but distinct strategy pro�les where certain agreement types \nsubstitute for each other or complement each other are present. These strategy profiles \nappear to be associated with particular industries and cartels of common geographical \nscope.