Computational Institutional Analysis of Bureaucracies

Abstract

Bureaucracies are crucial to policy performance and qualify political regimes. Theories portray them as inertial or responsive, entrepreneurial or risk-averse, siloed or collaborative, vulnerable to capture or stalwart. Such diversity is ascribed to differences in the configurations of rules that shape their structure, coordination, and agency. However, the mechanisms connecting specific rule configurations to models of bureaucratic agency and delivered goods and values can seldom rely on systematic and credible evidence as prescriptions may need. This Panel invites Papers that discuss how computational tools can help gauge, render, and evaluate these mechanisms and the conditions for scaling them up.

Title	Details
Detecting Democratic Backsliding in Assessment Reports Using Computational Social Science Tools	View Paper Details
Who is in charge? Explaining the European Union’s changing implementation design through a machine-learning application to the syntactic structure of EU laws	View Paper Details
From democracy to autocracy through bureaucracy: Behavioural and functional implications of backsliding theories tested ‘in silico’	View Paper Details
Neighbors with Benefits: How Politicians' Local Ties Generate Positive Externalities When Bureaucratic Oversight is Limited	View Paper Details
Understanding the information quality of public comments on bureaucratic policymaking in the European Union: a text-as-data approach	View Paper Details