Which electoral systems succeed at providing proportionality and concentration? An investigation of promising (mixed) designs and risky tools

Electoral systems are typically faced with the problem of being asked to provide both proportional representation and party system concentration leading to accountable government. Which electoral system designs are able to successfully deliver on both these challenges and thus optimize the representativeness-accountability trade-off? This paper investigates the performance of different general electoral system designs as well as their specific technical details (such as legal threshold, tier-linkages, and compensation mechanisms) based on a dataset of 590 elections in 56 countries. The key results are that both proportional representation systems with moderate district magnitudes and mixed-member proportional systems are able to optimize performance. Going to the level of details confirms these results and deepens our understanding further: while different technical changes are able to improve the chances of reaching the best of both worlds, some of these (e.g. raising the legal threshold) also increase the risk of ending up with the worst.