## Notes from 30 January 2026 [[2026-01-29|← Previous note]] ┃ [[2026-01-31|Next note →]] Coming from a Brazilian federal context, I always find it striking how _country-specific_ “decentralisation” debates can be - especially in the UK, where devolution is asymmetrical. Scotland, Wales, and Northern Ireland have their own devolved legislatures, yet their MPs still sit and vote in the UK Parliament (including on many England-only matters), while England has no separate devolved parliament of its own. That institutional patchwork shapes how people talk about “local government” and “state capacity” in ways that don’t map neatly onto the federal intuitions I’m used to. What I liked in Jack Shaw’s argument (“[Forget state capacity: what about local capacity?](https://jackshaw.substack.com/p/forget-state-capacity-what-about)”, 1 Feb 2026) is precisely that it cuts through constitutional uniqueness and lands on a practical institutional point: you can’t “fix Whitehall” if the delivery state is capacity-constrained locally —and you don’t need a grand redesign to start addressing that. The most elegant proposal is a _tour of duty_: making Fast Streamers spend six months in a strategic/local authority, both to inject scarce generalist capacity and to force central policymakers to learn, firsthand, how services actually operate on the ground. It’s the kind of simple mechanism that creates real cross-level understanding. He also proposes a set of practical bridges: making local government a more attractive career destination (including time-limited cohorts of specialists - behavioral science, AI - embedded in councils), explicitly reassessing Whitehall headcount and roles as devolution matures, and even physically co-locating civil servants with local officials to create everyday collaboration and informal knowledge exchange. He also argues for a stronger “policy–university ecosystem” in places - building infrastructure to embed local academic expertise in local problem-solving, drawing on existing fellowship-style models.