## Notes from 01 February 2026 [[2026-01-31|← Previous note]] ┃ [[2026-02-01|Next note →]] I found this provocative working paper (from this year!), and I think it will interest anyone interested in civil service reform and the politization of administration in subnational settings. [Matthew Stenberg and Jarle Trondal](https://www.sv.uio.no/arena/english/research/publications/arena-working-papers/2026/wp-01-26-final.pdf) investigate how state-level public administrators in the US perceive their independence from political pressure. Using 44 years of survey data from the American State Administrators Project (n ≈ 11,500), the authors challenge the common assumption that state bureaucracies are sensitive to partisan shifts or radical civil service reforms. The core takeaway is a direct challenge to the "Laboratories of Democracy" metaphor. While political rhetoric suggests a bureaucracy under siege by partisan polarization, the longitudinal data from 1964–2008 shows a robust stability in perceived autonomy. It is a reminder that the administrative state functions more like a self-insulating fortress than a flexible political instrument. The findings regarding civil service "deregulation" are particularly striking. Even in states like Georgia and Florida, where radical reforms significantly weakened employment protections, public administrators did not report a corresponding drop in their day-to-day operational autonomy. This suggests that technical expertise and organizational complexity provide a more effective shield against political encroachment than formal legal statutes or "at-will" employment clauses. The primary institutional exceptions occur in states with a "Weak Governor" model, a concept distinct from the centralized executive power typical in countries like Brazil: - **The Plural Executive (Texas & Mississippi):** Unlike the Brazilian model where the Governor holds nearly absolute authority over the cabinet, these states feature a "fragmented executive." Many top officials (e.g., Attorney General, Land Commissioner) are elected independently, meaning the Governor cannot simply command the entire executive branch. - **Legislative Supremacy:** In these jurisdictions, bureaucrats report that state legislatures (rather than the governor) exert the primary influence over budgets and major policy shifts. The Governor is often just one of many competing political principals. - **Institutional Path Dependency:** This fragmentation acts as a natural buffer for career bureaucrats. By operating in a system with "many bosses" and no single clear authority, agencies successfully leverage this lack of top-down coordination to maintain their own professional logic. The paper is surprising because it effectively dismantles the narrative of a "politicized machine" at the subnational level. Despite the partisan realignment of the American South and the intensifying polarization of the late 20th century, the perceived "room for maneuver" for top-level administrators remained a nearly flat line. Bureaucrats seem to successfully leverage internal resources to maintain independence, making them far less vulnerable to democratic shifts than is often assumed. This leaves me with a question: if neither radical reform nor the "will of the people" via elections significantly alters how the machine is managed, what actually does? The stability described here is both a guarantor of professional standards and a significant barrier to any meaningful structural change.