Examining persistent effects of extractive institutions in the United States

This paper estimates the medium‐ to long‐run effects of slavery in the United States in a spatial regression discontinuity design. Using the boundary between free and slave states immediately antebellum, we find that legal slavery decreased per capita manufacturing output by as much as 30% in the decades following the Civil War. Perhaps surprisingly, agricultural output and farm values were only briefly depressed in former slave states after the war. Although emancipation ended slavery, political forces kept the institution from being completely disintegrated, and we explore channels through which this was possible. We show that slavery affected the structure of the economy in a given region—specifically through agricultural production decisions—and that these structures persisted long after passage of the 13th Amendment. However, sharecropping played a relatively small role in this region. Our results support mounting evidence in recent literature of the significant and lasting effects of institutions on economic development.