Jo Michell

@jomichell.bsky.social

Professor of economics at UWE Bristol. Chair of Post-Keynesian Economics Society. Progressive Economy Forum council member. Interested in macro, finance, banking, climate change, inequality, demographics. https://people.uwe.ac.uk/Person/JoMichell

Very good from @dianecoyle1859.bsky.social. Entirely correct that a fundamental reset of the institutional framework is needed. We’ve tested the current one to its limit. It doesn’t work.

The system needs a reset, or the mythical fiscal "headroom" will only ever shrink. Public spending needs to become more effective by devolving decision-making powers and accountability for outcomes. Decentralised decisions benefit from detailed information and easier co-ordination than is possible in Whitehall. Local people know who is responsible if services are not up to scratch.
The Treasury and the OBR need to agree a strategic fiscal framework instead of short-termist game-playing. It should include public investment, which takes many years to affect growth and so is never "scored" highly in the OBR'S forecasts; if it does not go ahead, we will never get the growth.
One lesson from Manchester is that it takes time for education, employment and health policies to show up in GDP statistics. But meanwhile, people have better skills, jobs and health; almost 20 years on from the first devolution deal, the city region has the growth. My reading of Burnham's announcements so far is that he has taken this to heart.
Giles Wilkes@gilesyb.bsky.social · yesterday

How to end the cycle of mistrust between the Treasury and everybody else - giftarticle.ft.com/giftarticle/... Must read from @dianecoyle1859.bsky.social The need to micromanage is really hard to uproot. Something for Healey's advisers to think about too

1. Open weights has caught the frontier 2. Qwen imho is serially guilty of benchmaxxing, so we’ll see 3. openai and anthropic both have unreleased models that beat this by a lot 4. those releases are delayed to accommodate government-mandated release gates

Important stuff here from @lmacfarlane.bsky.social. I had a busy week (school holidays); I spotted the “higher tax rate loses money for Scotland” headline, thought it sounded dubious at best, and mentally made a note to come back and check it. Fortunately Laurie has done it for me.

Laurie Macfarlane @lmacfarlane.bsky.social · 4d ago

🚨 NEW: @danneidle.bsky.social claims Scotland’s 48% tax rate has lost public revenue. But is this actually true? My new deep-dive says no – Dan’s analysis doesn’t prove anything. It’s a long read, so buckle up 🧵

I realise the details are unlikely to be forthcoming, but I'd like to to know more about why OpenAI and Anthropic are unable to set up secure test environments without internet connectivity. It doesn't appear a particularly difficult task.

I suppose we should be thankful that the rise in gilt yields is not attributed to Burnham and Healey themselves.

Andy Gibson (@AndyGibsonTV) / InformingBritain.com@informingbritain.com · 2w ago

"IRAN WAR NARROWS HEALEY'S OPTIONS" #DelphineStrauss / @LucyFisher.FT.com / @IanKMSmith.FT.com for @FinancialTimes.com #IranWar pressures are shrinking Chancellor #JohnHealey’s fiscal options and squeezing the government’s buffer. #FTWeekend #TomorrowsPapersToday

In the past we would sometimes forgive even very bad writing if a paper contained a correct proof of an important, interesting result. But we probably should not do that anymore. An important job of the human researcher in possession of a proof machine is to digest the proof and

I find myself strangely underwhelmed by the OpenAI hugging face story. Once we had autonomously running LLMs connected to the internet this kind of thing was a dead cert — it was just a case of if not when.

Finding that I can still summon up a bit of savage indignation for governments blowing smoke on public spending with real world consequences. No, the aid money is not paying for the bus subsidies. But the aid is still being cut.