Stephen Jarvis

How we’re making data centers more flexible to benefit power grids An update on how we’re developing flexible data center capabilities to support power grids and build a sustainable, affordable energy system.

Very exciting to see Google step up on load flexibility of data centers. Hope we'll see other hyperscalers do the same. Critical for developing AI while maintaining reliable grids and reasonable electricity prices. (my blog: energyathaas.wordpress.com/2025/04/14/c...) 🔌💡

blog.google/inside-googl...

Congrats!

Motherfucking wind farms…

Agreed. That many UK universities are going through financial difficulties at the moment (largely due to avoidable gov policy failings) is also terrible timing.

Tax. The. Gambling. Industry. More.

The high-tech "open the window" strategy strikes again :) Agreed this tends to do the job and is often better than trying to fiddle with the room controls.

Builds up fast doesn’t it! My least favorite is when the classroom does not have windows that are easily opened

Definitely think Covid made this all a lot more salient, although some just seem to have erased this from their memories.

This! I have a strong preference for classrooms on campus that are newer and have one of these CO2 detectors on the wall. Being stuck doing back-to-back teaching in an old room with minimal ventilation in winter is the worst.

Food prices surge amid extreme weather

#climateflation graphic of the day, via @financialtimes.com

Hmm, could well be. I'm also all for students exploring different social science / interdisciplinary approaches, and I'm sure I'm not alone in having thoughts on where economics teaching, or the wider profession, could do better. But this kind of "it's all corrupt/broken" stuff isn't super helpful.

All disciplines have gatekeepers and can be slow to change - perhaps econ more than most. But this notion that students just learn about rational agents and efficient markets before they head out to destroy the planet is pretty reductive, and doesn't align with the reality of econ teaching/research.

The dominance of neoclassical economics in our university curricula has created a world where we are told there is no alternative — only technical adjustments to a system that is fundamentally fair, rational and efficient. But this is fiction. Economics today resembles Catholic theology in medieval Europe: a rigid doctrine guarded by a modern priesthood who claim to possess the sole truth. Dissenters are shunned. Non-economists are told to “think like an economist” or not think at all. This is not education. It’s indoctrination.

Also this trope of an "orthodox neoclassical priesthood" suppressing the plucky, heterodox "dissenters" and "activists" is really quite silly.

Anyway, with that I'll go dig out my cassock and get back to prepping my teaching materials - these students won't indoctrinate themselves.

Rethinking Economics, a group of activists that emerged from the post-crash student rebellion, recently took stock of the state of economics taught at UK universities in its curriculum health check. The findings are sobering. “The climate crisis and socio-ecological issues are broadly absent from economic curricula,” it found. Seventy-five per cent of universities “do not teach any ecological economics”. The report shows how the mainstream’s best-ranked universities often perform the worst in preparing students for the real world. We are training future economists to fail again.

The broad approach of "mainstream" environmental economics (market failures, externalities etc) may not be to the author's liking. But these core economic theories are still valuable tools for understanding real-world phenomena, and the quantitative methods students learn have lots of practical uses

Economics teaching has become the Aeroflot of ideas The discipline is failing students by ignoring the biggest social, political and ecological challenges facing the world today

Personally I find this kind of "review" quite unhelpful and misleading. Economics teaching of environmental issues (climate, biodiversity, pollution etc) has expanded a lot in recent decades. There's huge interest from students and on the research side the environment field keeps growing rapidly. 📈📉

Sometimes quietly wish my research occasionally involved superconducting magnets producing 10x luminosity. Science is cool.

Zonal/nodal reforms were always a long-term technical debate about efficiency gains. Agreed that opposition from those that would lose relative to the status quo wasn’t some kind of nefarious con - they have private interests to protect. Alas fear of disappointing me wasn’t enough to shift the dial.

Thanks! Yes locational pricing was never going to be the panacea for high bills. I’d say government will have to throw some real money at that problem to shift costs off bills, enhance low-income discounts etc., preferably in a way that reduces (rather than worsens) existing distortions.

Quietly bookmarking these to share ahead of my upcoming class on energy markets. Keep 'em coming :)

It's a great point. In other nodal markets residential/commercial consumers actually tend to face a regional price that averages across a bunch of nodes (to avoid this exact concern). It's big industrial consumers and generators who can respond / manage the risk that face a specific nodal price.

They were really dragging it out by the final episode weren't they. Probably should have wrapped it up after the first or second season.

Thanks! Glad you found it helpful.

Honestly the thread could probably have been even longer but I think people will generally tolerate a brief trip into the weeds, not getting completely lost in them :)

Thanks Paul, and your earlier post did prompt me to give that UKERC piece a read too.

Makes sense. Easier to model long-term costs if you've got a smooth set of annual projected price paths and you're not thinking about some kind of stochastic gas price shock from a war.

Thanks for sharing. I'm with you on the point about long-term impacts and on modelling. Not sure I've seen an issue so heavily studied in ways that support either conclusion! Partly why I would fall back on some economic first principles. But agree the timing/implementation risks are hard to ignore.

Oh very cool! Weird to be in the "quote someones work back to them" position. Curious what you made of the whole debate on this topic.

Thanks! And yes, this did snowball a bit :)

Chart shows outturn balancing costs from 2018-2023. Constraint costs have roughly doubled, although receded slightly from peak during 2022 energy crisis.

Forecast balancing costs out to 2040 with costs rising through to around £3-5bn/yr by 2030, before significant new network upgrades come online.

Yes good point, and this comes through clearly with the soaring "cost of other services".

But even absent the gas price spike constraint costs have been rising and are forecast to keep growing to 2030 until various network upgrades are completed (see for example these rather ugly charts from NESO).

Thanks that's kind of you. The debate on this did get a bit spicy for sure! For me it feels like a missed opportunity, but I'm sure there is a reasonable case that big market shakeups in the middle of a huge renewable/grid buildout isn't ideal timing. I hope the issue is revisited eventually though.

It does in media/public debate but rarely in active policy discussion (esp. generation, although the newly created GB Energy will take some public equity stakes in upcoming projects). I'd say the UK rank ordering of "taking things back into public ownership" is probably: rail > water > mail > energy