There is no consensus among specialists in agricultural contracts over whether the long‐term inefficiencies that classical economists attributed to sharecropping actually exist. This article maintains that they do exist and are partly caused by the fact that sharecropping is hardly compatible with the tenant being compensated for improvements, viticulture being the main historical exception. In line with recent contributions to the sharecropping literature, the article contends that the widely held belief among scholars of agricultural contracts that sharecropping was very frequent in Europe's vineyards is incorrect. However, it also provides evidence of an issue whose importance has gone largely unnoticed: prior to the twentieth century, many of the European vineyards worked by sharecroppers had been created by the sharecroppers themselves, through contracts which entitled them to compensation. Those contracts abounded while viticulture depended basically on two inputs, land and labour. When viticulture became a heavy consumer of capital, they were rapidly abandoned, but not in Catalonia, with a paradoxical result: the Catalan rabassa morta contract, which for centuries had made it possible to eliminate both the long‐ and short‐term inefficiencies of sharecropping, ended up becoming an obstacle to overcoming the short‐term inefficiencies. The article discusses why that happened.