iiinova — where your digital self keeps evolving

Jo Michell

@jomichell.bsky.social

9.9K followers 1.1K following 7.4K posts

Professor of economics at UWE Bristol. Chair of Post-Keynesian Economics Society. Progressive Economy Forum council member. Official OBR messenger shooter 🚨. people.uwe.ac.uk/Person/JoMichell

Posts Replies Media

Jo Michell @jomichell.bsky.social · 39m

Arun Advani talking sense here.

Arun Advani, a tax specialist at Warwick university, said: "The problem with a quiet budget that doesn't fully fund the spending envelope for the rest of the parliament is that it leaves uncertainty over future tax policy hanging over business going into next year."
"This perpetuates the ongoing drag on investment as the government struggles to get on top of its fiscal problems," he added.

Jo Michell @jomichell.bsky.social · 40m

The predicted available margin on the (different) margin between an uncertain forecast and an arbitrary target five years hence is not a “fund” and is itself subject to substantial contingencies. Other than that, this seems fine.

Government figures see the fiscal buffer as a "contingency fund" against unexpected global shocks such as the Iran war, as Labour MPs warn against big tax rises now that risk choking off growth and alienating voters.

Jo Michell @jomichell.bsky.social · 01/10/2026

That's a lot of IT capex.

Jo Michell @jomichell.bsky.social · 01/10/2026

After years of experimenting and tinkering, I think I've finally found a way to simulate the Solow model in a way that could work for undergraduate teaching.

Jo Michell @jomichell.bsky.social · 01/10/2026

Just came across this from Victoria Chick on MMT (in a 2020 interview by Lavoie). Succinct and to the point, as ever.

What about modern money theory (MMT or neo-Chartalism). Have you had a look at their
works?
Yes, sure. I think one of the things they did, and I wish post-Keynesians around here had
said it, was to point out that government spends before the taxes come in. That is exactly
parallel to investment preceding saving. It is not even rocket science, but no one had really
said it. But it is another thing to say that, therefore, there is no limit on government
expenditure. This proposition may be true in the little framework that they are using,
but if you broaden it out to the economy as a whole, of course there are constraints on
government spending, like the balance of payments. It is a question of how big your fra-
mework is, whether you see those constraints or not.

Jo Michell @jomichell.bsky.social · 01/10/2026

Jo Michell @jomichell.bsky.social · 30/09/2026

full disclosure

LLMs were used in the analysis and writing of this article. For this project, Schwartz worked as a
contractor for Anthropic; the results and views are not endorsed by Anthropic. The views expressed
herein are those of the authors and do not necessarily reflect the views of the National Bureau of
Economic Research.
At least one co-author has disclosed additional relationships of potential relevance for this research.
Further information is available online at http://www.nber.org/papers/w35782

Jo Michell @jomichell.bsky.social · 30/09/2026

game over, lads www.nber.org/system/files...

An LLM Workflow That Reproduces, Improves, and Extends
Published Economics Research ∗
Matthew D. Schwartz1
, Isaiah Andrews2
, and Jesse M. Shapiro3
1Harvard University
2MIT and NBER
3Harvard University and NBER
Wednesday 23rd September, 2026
Abstract
We introduce an open-source workflow that enables an LLM to reproduce, improve, and extend an economics article using the article’s published replication package. First, the workflow attempts to reproduce
the original calculations, checks for discrepancies with published findings, and performs automated sensitivity analysis. Across 4,452 published replication packages for five economics journals, the workflow
flags discrepancies in 3,460 articles or their appendices. Second, the workflow improves the original calculations by using a different implementation or algorithm. In 496 articles, the workflow is able to reduce
a calculation’s computation time, at similar or greater accuracy, by more than a factor of 10. Third, the
workflow extends the original analysis. In 923 articles, the workflow develops an extension that does not
appear in the original article and that is aligned with the original article’s goals and assumptions.

Jo Michell @jomichell.bsky.social · 30/09/2026

Revised GDP figures: not much change.

• ons.gov.uk
• UK gross domestic product (GDP) is estimated to have grown by 0.5% in Quarter 2 (Apr to June) 2026 (revised up by 0.1 percentage points from the previous estimate), following an unrevised increase of 0.6% in Quarter 1
(Jan to Mar) 2026.
• GDP for 2025 as a whole is estimated to
have increased by 1.2%, revised down by 0.1 percentage points from the previous estimate.

Jo Michell @jomichell.bsky.social · 29/09/2026

The triple lock is dead, long live the triple lock?

Government says Labour adjusting, not scrapping, triple lock
Government officials have insisted that Andy Burnham is retaining a triple lock on the state pension, but concede that he is planning to adjust its terms from April 2030 if Labour wins the next election.

At present the triple lock means that the state pension is uprated annually by whichever is highest out of three metrics: CPI inflation, 2.5 per cent or average earnings growth that year.

Under the prime minister’s plans, earnings growth would be tracked over time instead of annually — while the other two metrics would remain unchanged.

Helen Miller, director of the Institute for Fiscal Studies think-tank, said on X: “I think the ‘tweak’ he mentioned can only imply that old triple lock is gone. New one is similar but without ratchet.”

Jo Michell @jomichell.bsky.social · 29/09/2026

The most sensible MPC member, by a margin. www.ft.com/content/4aa0...

However, in an environment shaped by large external supply shocks and weak domestic investment, the monetary policy response required to restore price stability leaves the economy more vulnerable to the next shock. Without broader resilience-building, depressed demand and restrictive financial conditions undercut the expansion of supply capacity just when it is needed most.

Jo Michell @jomichell.bsky.social · 27/09/2026

Jo Michell @jomichell.bsky.social · 27/09/2026

Jo Michell @jomichell.bsky.social · 26/09/2026

I don’t think we usually learn much from these hedge fund vox pops but “the yield curve is flattening, actually” seems to contain some information.

"The market is revising the Fed rate path to be higher for longer," said Seb Barker, chief market strategist at hedge fund firm Marshall Wace. "For all the drama around the long end, the [yield] curve has flattened year to date. What has changed is the pricing for the Fed."

Jo Michell @jomichell.bsky.social · 24/09/2026

Jo Michell @jomichell.bsky.social · 24/09/2026

Therefore, while the measure of the slowdown in output per worker is largely unaffected, today’s estimates reduce the size of the productivity slowdown by around half. However, overall, the “productivity puzzle” remains, albeit more muted than before, as these data still point to a fundamental shift in the UK economy following the financial crisis.

Jo Michell @jomichell.bsky.social · 24/09/2026

That's not really how the ONS themselves are framing it. blog.ons.gov.uk/2026/09/17/m...

Our new component-based method improves our estimate of both the number of workers contributing to UK output and the number of hours they work. Whilst it does not materially change our account of the slowdown in output per worker, there is a material change to why this occurred. We now think more of the slowdown can be explained by continued falls in average hours worked, leaving less to be explained by falls in output per hour.

Under the new approach, it remains clearly visible in the data that, following the 2008 financial crisis, the UK experienced a slowdown in productivity growth per hour compared with the historic long-term trend.

Jo Michell @jomichell.bsky.social · 24/09/2026

As Dan pointed out yesterday we are now treating the headroom against the target as a new target.

One person involved in the government's discussions has suggested that headroom of £15bn would be sufficient, while another suggested closer to £2obn, and a third said no figure was yet being targeted.
Another senior government figure said there was "no magic figure" to demonstrate credibility to the market, arguing that Britain's plan for rapid deficit reduction was potentially more important to borrowers.

Jo Michell @jomichell.bsky.social · 23/09/2026

I might appoint you official bluesky bio editor :)

Jo Michell @jomichell.bsky.social · 23/09/2026

Ha yes. Done. Deleted the pointless earnest fluff. Might need to add ‘siren’ somewhere too.

Jo Michell @jomichell.bsky.social · 23/09/2026

Seriously though, what do we think would happen if the OBR said “The Fiscal Rules are Breached” and the Chancellor said “you know what squire, I think I’ll take the pen anyway!”

28. That point was amplified by the Resolution Foundation, which wrote:
More than anything, those who lay the blame for our fiscal woes at the door of the OBR are fundamentally misattributing the problem. The source of sharp trade-offs and struggles to meet self-imposed rules are not a result of the role of the OBR. This is analogous to blaming referees for the poor state of football.
29. Similarly Ruth Curtice told us the OBR "does not veto government policy. This is a game of football where, if the Chancellor disagrees with the referee, she can take the penalty anyway."

Jo Michell @jomichell.bsky.social · 23/09/2026

And the chaser. (Obvs. I knew we’d submitted evidence back in the mists of time, just didn’t expect to be reading it again over my cornflakes this morning.)

Prof Jo Michell, an economist at the University of the West of England, and Dr Robert Calvert Jump, an economist at the University of Greenwich, told the committee that the OBR's forecasts about the level of spare financial capacity, often referred to as headroom, at the end of a five-year forecast had come to dictate government spending policies.
"The OBR has evolved into a body that determines the bounds of permissible fiscal policy, despite lacking the democratic mandate to fulfil such a role [...] It does not merely produce forecasts; it determines whether the government passes or fails its fiscal rules. As a result, its 'headroom' figures have become the binding constraint on policy," they said.

Jo Michell @jomichell.bsky.social · 22/09/2026

Yeah, maybe I'll skip the news break and get on with my admin.

Guardian headline: OpenAI's George Osborne says datacentre nimbys holding back Britain

Jo Michell @jomichell.bsky.social · 21/09/2026

We explore the interactions with demand regimes, Gibrat's Law, the granularity hypothesis and more.

Figures showing the division of 2D parameter spaces into different growth regimes.

A page from the appendix exploring Gibrat's law and heavy-tailed firm size distributions.

Jo Michell @jomichell.bsky.social · 21/09/2026

We also calibrate the ABM to a selection of moments from the US Compustat data with perhaps a surprising degree of success.

A table of simulated moments compared with empirical moments:

Table A2. Empirical and Simulated Moments.

Description Empirical moment Simulated moment
Mean firm-level profit share 0.1864 0.2067
Median firm-level profit share 0.1497 0.1513
Variance of firm-level profit share 0.0217 0.0211
Skewness of firm-level profit share 1.647 2.576
Normalised HHI of market shares 0.1456 0.1431
Skewness of market share 4.003 4.053
Mean firm-level capital growth rate 0.0914 0.1037
Mean aggregate capacity utilisation 0.7702 0.7049

Jo Michell @jomichell.bsky.social · 21/09/2026

As noted earlier, the analytical version allows us to to derive conditions under which various steady states occur and a bunch of other stuff. Although we can't also do this for the ABM, the results seem to carry over pretty well. We think this gets us quite a long way on the 'black box' problem.

A page of algebraic analysis of the conditions which produce a steady state with monopolistic concentration.

Another page of algebraic analysis of the conditions which produce a steady state with monopolistic concentration.

Jo Michell @jomichell.bsky.social · 21/09/2026

The emergence of monopoly power leads to a falling wage share and - under certain parameterisations, stagnating growth as a result of constrained wage-consumption-goods demand. The function distribution of income is the outcome of both market share and markup pricing.

Simulation results showing a rising aggregate markup and falling aggregate rates of growth and utilisation.

Jo Michell @jomichell.bsky.social · 21/09/2026

For example, this simulation shows a duopoly emerging: two firms become dominant.

A simulation run of the agent-based model showing two firms dominating in terms of market share, capital stock (firm size), and profit shares.

Jo Michell @jomichell.bsky.social · 21/09/2026

Not sure there are many Kalecki heads on this website but I'm going to give this a try anyway. If this isn't your cup of tea, you might want to hit mute now...

A journal cover page for "An analytical heterogeneous agent macro model of concentration, markups, and falling labour shares" by Karsten Kohler, Jo Michell and Ayoze Alfageme, published in the Journal of Economic Dynamics and Control.

The abstract text is: This paper presents an analytical heterogeneous agent model (HAM) to explore micro-level concentration dynamics and their effects on macro-level distribution and growth. It contributes to recent work on macroeconomic agent-based models (ABMs) by comparing a small-scale ABM with bilateral firm-customer matching to a simplified HAM without direct agent interactions. The HAM replicates key ABM results while allowing for analytical solutions. In the model, larger firms benefit from lower unit costs due to economies of scale, gain larger market shares, and set higher markups. More profitable firms temporarily grow faster, leading to endogenous changes in the firm size distribution. The resulting right-skewed distributions create a divergence between the average profit share across firms and the aggregate profit share, which is driven by the largest firms. The decline in the labour share reduces aggregate consumption and slows growth. Despite its simplicity, the model captures key empirical patterns related to superstar firms and declining labour shares.

Jo Michell @jomichell.bsky.social · 18/09/2026

This guy sounds sensible. I’ve long been a big fan of Raspberry Pi, and this hasn’t done anything to change my mind. giftarticle.ft.com/giftarticle/... Raspberry Pi founder Eben Upton: ‘I’m an omni-geek’

He appears deeply sceptical of the Silicon Valley evangelists, intent on creating a machine God. "It can be a little bit hard to tell whether they're talking this way because that's how you raise money or whether they're talking that way because they really believe it. But I think there are a lot of them for whom it's both."

Jo Michell @jomichell.bsky.social · 16/09/2026

Is this really the best the Tories can manage in response to the CPI release?

Andrew Griffith, shadow chancellor, took aim at the government for August’s increase, saying: “Inflation is now at 3.1 per cent, meaning that every family is paying the price for Labour’s choices.”

Jo Michell @jomichell.bsky.social · 15/09/2026

Charlie Bean is in the growing list of experts who acknowledge the need for coordination of monetary and fiscal policies.

Charlie Bean, a former deputy governor of the Bank of England, said: “I do not think it is politically sustainable for the MPC to be able to take such decisions without the involvement of the [Treasury] or else somehow reduce the magnitude of spillovers [to the Treasury].”

Jo Michell @jomichell.bsky.social · 15/09/2026

The policy conclusion in the paper where that line of argument started is easy to find. progressiveeconomyforum.com/wp-content/u...

A rational policy response at this point in time would be a cautious, ‘wait and see’ approach.
Spending should not be cut, while taxes on higher earnings and income from wealth could
reasonably be increased to cover any further increases in borrowing costs if they occur. For
an interactive tool to explore these options, see https://arunadvani.com/taxreform.html.

100

Jo Michell @jomichell.bsky.social · 15/09/2026

I've argued for higher borrowing for *investment*, alongside higher taxation to cover current spending and interest payments. This is nearly two years old, for example: phenomenalworld.org/analysis/bey...

Starmer’s Labour Party took power committed to an impossible policy offer. Pre-election pledges ruled out increased taxes on “working people”—widely interpreted to mean not only income tax, but also national insurance and VAT—as well as increases in corporation tax. This placed around 75 percent of the tax base out of reach. Labour retained fiscal rules inherited from the previous administration, including a commitment to ensure falling debt-to-GDP on a rolling five-year basis. Despite apparently ruling out increases in either taxes or borrowing, Labour were somehow adamant that there would be no return to austerity in the form of spending cuts.

A range of fiscal demand-constraint policies are available, including the use of taxation and household saving incentives. How should these be structured? The progressive answer is that additional revenue should be generated by increasing the contributions of the most well off. However, the structure of tax in the UK is already highly progressive: the top 10 percent by income already provide over half of tax revenues. Thus, while the share of tax in national income is at a historically high level of around 36 percent of GDP, the effective personal tax rate for the average earner is at its lowest since 1975. Further tax increases on average earners will be increasingly hard to avoid if the intention is to raise growth and avoid inflation while sustaining tight labor markets.

Jo Michell @jomichell.bsky.social · 15/09/2026

Literally the next paragraph and yet no connection is drawn. Can anyone remind me — what was the UK government’s economic response to the 07-08 crisis, who was responsible for designing it, and what is the broad consensus on the results, including the effects on growth?

But the economy still bears the scars of the global financial crisis and the pandemic.
GDP per capita would be 30 per cent higher today if the pre-2008 trend had continued,

Jo Michell @jomichell.bsky.social · 15/09/2026

Give me strength.

"I think everyone shares Andy Burnham's desire not to be in hock to the bond market," says Rupert Harrison, senior adviser at asset manager Pimco and former chief of staff to Conservative chancellor George Osborne.
"But in this environment it's the bond market that's defining his Budget choices."

Jo Michell @jomichell.bsky.social · 15/09/2026

This is ludicrous. The basic shape of the UK’s fiscal problem has been visible from space for years. giftarticle.ft.com/giftarticle/...

Jo Michell @jomichell.bsky.social · 12/09/2026

Also: the ‘saving rate’ in these models has little to do with actual saving. It’s an investment rate. Saving — not consuming — does not generate new machines. Building new machines does. The conflation is misleading.

If, in addition, the saving rate responds to the soaring return on machines (green lines), growth reaches 15% already by 2034 and 20% by 2045.

Jo Michell @jomichell.bsky.social · 12/09/2026

In a Solow model, alpha, the exponent on the capital stock, is also the capital share of national income. If alpha goes to one, the labour share goes to zero.

$The second building block is the standard Solow model: a constant fraction of output is saved and invested in machines. In reduced form, the model boils down to two familiar equations: where Z = Z(A,α) and A is the level of technology which grows at a constant rate. The key property is that the capital share of this Cobb-Douglas production function equals the share of automatable tasks α. That is, automation raises the exponent on capital.$

Jo Michell @jomichell.bsky.social · 12/09/2026

I think they could have made a bit more of this — some of the other objections and caveats are perhaps less important than this one, if we are discussing how feasible the scenario is.

Jo Michell @jomichell.bsky.social · 11/09/2026

Including this spectacular archive clip from the FT leader column.

In that piece I linked to a then recent Editorial (not just an op-ed) in the Financial Times laying out how society ought to change when it was all over, which spoke of how things looked and felt, even to those raised on ‘sound’ economics, while they were stuck at home:

Radical reforms — reversing the prevailing policy direction of the last four decades — will need to be put on the table. Governments will have to accept a more active role in the economy. They must see public services as investments rather than liabilities, and look for ways to make labour markets less insecure. Redistribution will again be on the agenda; the privileges of the elderly and wealthy in question. Policies until recently considered eccentric, such as basic income and wealth taxes, will have to be in the mix.

Jo Michell @jomichell.bsky.social · 11/09/2026

Not sure about OAI but anthropic has opt out options for use of chat transcripts for training. Whether they actually adhere to it, who knows.

Jo Michell @jomichell.bsky.social · 10/09/2026

The Makerfield byelection really seems to be dragging on.

Jo Michell @jomichell.bsky.social · 09/09/2026

Martin Wolf endorsing Warsh’s quasi monetarist views.

Another area in which I believe conventional economics has been proved misleading is, as Warsh indicated in his speech, in deciding that the quantity of money is irrelevant. The post-pandemic rise in the price level, after the huge monetary expansion of 2000, is, as I have previously argued, an indication that this view is quite mistaken.

Jo Michell @jomichell.bsky.social · 09/09/2026

The reserves must be running pretty low by this point.

Jo Michell @jomichell.bsky.social · 08/09/2026

giftarticle.ft.com/giftarticle/...

We are invited to giggle at Musk and his goons' cluelessness on how state institutions function, but the laughter abruptly stops as the human cost of halting USAID payments is counted: the estimated 14mn people who, we are told, will die due to USAID cuts, 4.5mn of them children under five.

Jo Michell @jomichell.bsky.social · 07/09/2026

Still got my X61. Still my favourite.

Jo Michell @jomichell.bsky.social · 04/09/2026

Higher productivity requires higher interest rates. This is why interest rates were so much higher in 2011 than in 1874.

Faster, more productive economies can absorb - nay, require - a higher cost of capital to keep savings and investment in balance. Thus it is my view, shared by

Jo Michell @jomichell.bsky.social · 04/09/2026

Impressive number of conceptual errors for a single paragraph, even considering the fact that the subject — the natural rate of interest — is itself deeply confused and incoherent.

Higher real rates are the main reason yields are rising. These reflect an economy's so-called natural rate (or what bores like to call r*). This is the real cost of money that neither stimulates nor restrains growth — rising when the underlying growth potential expands and vice versa.

Jo Michell @jomichell.bsky.social · 04/09/2026

I’d been wondering about this. I realise that Cambridge academics don’t have to do any actual work, but still — holding down a full time academic post plus head of policy for Reform plus all the hanging around with weirdos in the US. Seems quite a lot.

The FT has been told about more
widespread concern from students that Orr is struggling to combine his academic work with his Reform role and overseas lecturing commitments. He is also a faculty member at Jordan Peterson's online Academy and a fellow at the private US liberal arts institution Ralston College, which is funded by GB News and Spectator owner Paul Marshall. One of his most significant faculty roles at Cambridge is as chair of undergraduate exams. In June this year, he arrived late and appeared not on top of his brief to the online exam board meeting for a final marking review, according to people with knowledge of the meeting. That week