In this project, we develop new theoretical insight into the anatomy of hard-core cartels and combine it with a rich data set on the recent German cement cartel. This private data set comprises about 340.000 market transactions from 36 customers of German cement producers and encompasses most of the period during which the cartel was functioning as well as a period after the collapse of the cartel. We will first investigate the impact of collusion not only on the price level but also on the structure of prices and thus analyse to what extent collusion reduces the use of efficiency-enhancing customer-specific prices. Secondly, we will explore and quantify the productive inefficiencies created by collusion causing an inefficient matching of buyers and sellers. Thirdly, we plan to test a well-known theory of collusive pricing, which suggests a close link between collusion and price reactions triggered by demand changes.