When and why do countries redistribute land to the landless? What political purposes does land reform serve, and what place does it have in today's world? A long-standing literature dating back to Aristotle and echoed in important recent works holds that redistribution should be both higher and more targeted at the poor under democracy. Yet comprehensive historical data to test this claim has been lacking. This book shows that land redistribution - the most consequential form of redistribution in the developing world - occurs more often under dictatorship than democracy. It offers a novel theory of land reform and develops a typology of land reform policies. Albertus leverages original data spanning the world and dating back to 1900 to extensively test the theory using statistical analysis and case studies of key countries such as Egypt, Peru, Venezuela, and Zimbabwe. These findings call for rethinking much of the common wisdom about redistribution and regimes.