Unpacking public corruption in the U.S.: An actor-action-target analysis

Prior corruption research in the United States has primarily focused on the causes and consequences of corruption through macro-level analysis of aggregated data. While they provide valuable insights into systemic and institutional drivers, micro-level descriptive research is essential for uncovering the patterns and vulnerabilities of public corruption, which are often obscured by aggregate data in macro-level analyses. This study introduces the Actor-Action- Target (AAT) framework to systematically examine who engages in corruption (actor), what corrupt actions they undertake (action), and what targets are exploited for illicit gain (target). Hence, it provides a landscape of public corruption in the United States through a micro-level analysis of real public corruption cases. The dataset is obtained by applying generative AI to compile over 2,000 public corruption cases disclosed by the FBI from 2014 to 2025. The results reveal systematic patterns linking actors' institutional authority to the forms of corruption committed and the target pursued. By highlighting these relationships, the AAT framework offers both conceptual understanding of public corruption and practical relevance, providing a structured basis for targeted anti-corruption strategies.