## Notes from 07 April 2025 [[2025-04-06|← Previous note]] ┃ [[2025-04-08|Next note →]] Revisited this [[Samuel Hammond]] (a consistent banger!) [tweet](https://twitter.com/hamandcheese/status/1628944996830898176) on differing views of "state capacity". He contrasts two perspectives: <table> <thead> <tr> <th>Feature</th> <th>'Left' View</th> <th>'Right' View</th> </tr> </thead> <tbody> <tr> <td><strong>Focus</strong></td> <td>Inputs (personnel, rules)</td> <td>Structure, leadership, outcomes</td> </tr> <tr> <td><strong>Means</strong></td> <td>More staff, higher pay, relaxed procedures</td> <td>Lean organizations, hierarchy, high agency</td> </tr> <tr> <td><strong>Emphasis</strong></td> <td>Bureaucratic resources</td> <td>Executive decisiveness</td> </tr> </tbody> </table> Hammond notes these aren't strictly mutually exclusive (cites Singapore as combining elements) but represent distinct approaches. He also points out that periods of high US state capacity (FDR-Eisenhower era: Hoover Dam, Manhattan Project, Apollo, ARPA) relied heavily on hierarchical structures led by high-agency individuals. The "[[DARPA model]]" was once just _the_ model. The tension here is between improving the state by bolstering its existing operational inputs versus reshaping its structure to empower decisive leadership. One focuses on _resources_, the other on _agency_ and _structure_. This resonates with the [[Centre for British Progress]] [paper's diagnosis](https://britishprogress.org/articles/rediscovering-british-progress) of institutional 'impotence' and its call to 'rediscover agency' within the state, exemplified by models like ARIA/DARPA. Their analysis highlights "executive decisiveness" via empowered Programme Directors/Managers who have end-to-end ownership and minimal bureaucratic drag, mirroring Hammond's 'high agency leader'. However, the paper crucially notes that establishing ARIA in the UK required creating an "air gap" – significant exemptions from standard rules (Treasury/Budget, procurement, FOIA, Civil Service codes). This wasn't systemic reform but a [[Regulatory Carve-out|workaround]], suggesting the broader state apparatus is ill-suited to this high-agency model. Perhaps underlining the perceived need for exceptional talent to justify such exceptional operating conditions, ARIA recently detailed their sophisticated Programme Director selection process [in their newsletter](https://ariaresearch.substack.com/p/how-we-found-our-new-programme-directors). Both the rigor of the selection method itself, and their transparency in communicating it, seem to underscore a departure from standard procedures, arguably aiming to build confidence in leaders granted unusual autonomy. My takeaway: The emphasis on high-agency leadership seems critical. It suggests effective government action may hinge less on optimizing existing processes or simply adding personnel, and more on selecting capable leaders and then granting them significant operational freedom (budget, team selection, decision authority) within a, let's call it, "mission-driven" structure. The need for ARIA's insulation highlights the challenge: _embedding_ agency requires more than just creating pockets of excellence, it requires confronting the default state structure itself.