Subscribe now

Introducing: K chat

I kept hearing it. In podcasts, X posts, media interviews, you name it. The K-shaped economy. By mid-2025 I couldn’t open an economics newsletter without someone drawing two diverging lines on a chart and calling it a revelation. The basic claim was clean enough: top earners were pulling further ahead while ordinary incomes stagnated. One economist described it as “the jaws of a crocodile,” which is the kind of metaphor that sounds better than it explains. Top-third spending rose 4% year-on-year while the bottom-third grew less than 1%. Two lines, shaped like a K. Case closed, supposedly.

But is it true?

Where the K came from

The term traces back to 2020, when an anonymous Twitter account called Ivan the K made a playful reference to a “K recovering” in the early weeks of COVID. Peter Atwater, a behavioural economist at William & Mary, picked it up and ran with it. Atwater had noticed a gap opening between white-collar workers who shifted to home offices and blue-collar workers who couldn’t. The white-collar crowd felt fine. The hospital workers, supermarket clerks, and warehouse staff felt worse by the week. Atwater used the letter K to describe an economy splitting in real time: one arm rising, one falling. It was a good metaphor, because it was visual. You could draw it on a napkin and people would nod.

The term went viral almost immediately. By late 2020, Al Jazeera was explaining it as “Wall Street soars, Main Street struggles.” Within a year it was in every economics podcast, every op-ed section, every investor note. Fortune called it one of the biggest business buzzwords of 2025. It joined the alphabet soup of recession shapes (V, U, W, L) but added something new: not just speed of recovery, but divergence. For the first time, the metaphor said the economy wasn’t just slow or fast. It was splitting apart.

The post-COVID K shaped recovery

The K became the default explanation for the post-pandemic economy. The story went like this: white-collar workers shifted to home offices and kept earning. Lower-income workers got furloughed, put on benefits, or lost their jobs entirely. When the economy reopened, the white-collar labour market tightened fast. Employers competed for knowledge workers, a salary premium emerged, and the gap widened. Top-third consumer spending rose while the bottom-third barely moved. The IFS noted that poorer households faced higher effective inflation because they spent more of their budgets on food and energy, so even where incomes held steady, purchasing power didn’t. Draw it on a chart and it looks like a K. For a moment in 2024 and early 2025, this seemed to explain everything

Then the Minneapolis Federal Reserve published a review, and things got messier. Their researchers looked across years of household data and found “little support for a consistent K-shaped pattern in recent years.” Before 2025 hit, lower-income households often outperformed higher-income ones in terms of income growth. The narrative started to look shaky.

Noah Smith made a similar point on Noahpinion, arguing that income inequality had actually started declining post-COVID, and that some of the wealth gap was “imaginary” because it was driven by stock market valuations rather than anything tangible. Krugman went the other way, arguing the K was real but politically contingent: Biden-era wage gains had briefly narrowed it, and the Trump-era economy was pulling it apart again.

The income K, it turned out, depended entirely on who you asked and what they measured. It didn’t show two clean groups pulling apart. It showed something messier: three or four tiers, shifting at different speeds depending on the metric. Not a clean split. A fracture.

But what if the K is somewhere else?

I think we’ve been missing the point.

The real, durable K-shaped split isn’t income-based at all. It’s about assets, housing, generations, and who owns what. And the story in Britain and Australia is both clearer and sharper than the American one.

Let’s look at the UK first. The Gini coefficient, that standard measure of income inequality, barely moved over the past decade. By that metric, nothing much changed. Yet wealth inequality tells a completely different story. The gap between the wealthiest and poorest British households grew roughly 50% over eight years. The IFS has argued that wealth is now a more important driver of economic divides in the UK than income. The generational split is even sharper. ONS data shows individual wealth peaking in the 60-to-64 age group at nine times the level of the 30-to-34 group, with younger generations less likely to own homes than any cohort before them. One generation rode housing inflation like a winning trade; the next one arrived after the party ended, with deposits to save for homes that cost twice what they did a decade earlier.

Australia shows the same pattern, amplified. ABS data puts it plainly: the bottom quintile of income earners holds just 0.4% of total wealth, while the top quintile holds 63%. Property prices have climbed roughly 2,300% since the early 1970s, while wages rose about 1,400%. The Grattan Institute reports that the typical Australian home has gone from four times median income in the early 2000s to over eight times today. The Treasury’s Intergenerational Report projects $5.4 trillion in wealth transfers over coming decades, with inheritances having more than doubled since 2002.

The Australia Institute now frames the housing crisis not as a standalone policy issue but as the primary driver of inequality in the country. A young Australian earning a decent salary might see real income growth while wealth moves further away. Not because they’re failing, but because the asset base itself has become an entirely different ballgame. The K-shape here is really a property-price shape wearing a K’s costume.

Consider what this means for policy. If the K-shaped recovery was really about consumption smoothing between rich and poor, then you’d want to think about incomes, transfers, and purchasing power. Fiscal policy. Minimum wages. Universal basic income debates. The usual suspects.

But if the K is really about housing and inherited wealth, and the evidence from the UK and Australia suggests it is, then the policy conversation shifts entirely. You’re talking about land supply, property tax reform, intergenerational capital gains taxes, and whether a 50-year asset price bull run should be allowed to continue its march against wage growth. You’re talking about whether future generations should expect to inherit housing or student debt. Those conversations are messier. They threaten people who own (or control) existing assets.

The right letter, the wrong axis

The K-shaped economy was a useful metaphor when it was new. It pointed at something real. But everyone got the right letter for almost entirely the wrong reasons. The K that matters isn’t about who earns and who doesn’t. It’s about who owns and who rents. Capital vs labour. The media saw the letter and ran with it. Economics journals filled with papers. Think tanks produced reports. And all of them were measuring the wrong axis.

We don’t need a new letter of the alphabet. We need more houses. And we probably need to stop inventing clever names for problems we already know how to describe.

Subscribe now

Tangentially Economics publishes articles multiple times a month. Subscribe for more on the economics everyone’s talking about, and the ones they probably should be.

Subscribe now

Another book on land economics

Mike Bird’s new book The Land Trap hit the policy wonk charts (and even the FT long list) in the lead up to Christmas 2025. It explains land is an unique asset class that inevitably skews lives and livelihoods in favour of owners.

The opening premise proposes the immobile and finite nature of land creates a economic environment where property owners capture wealth generated by others.

This idea has been around for a while, but Bird goes further. Once you’re in the land trap, Bird says that policymakers must focus on managing the consequences of this trap, rather than expecting a flight path back to affordability.

Bird describes this by thinking about cities:

The thousands of workers that flow into bustling cities each morning on buses, cars and train are what give the properties in those places their eye-watering rents an exorbitant prices. The same people pay steep rents to live nearby ,often scrimping and saving for decades to acquire some modest share of land wealth for themselves. The windfall from rising land values is reaped by those who has the good fortune to buy it in a particular place, at a particular time, before prices began to rise. It is, for the most part, a game of luck.

Sound a little like Georgism? Henry George, alongside the concept of a single tax gets a good showing in the book. Bird explains that over time land acts as a ‘trap’ that absorbs the financial gains of a developing society.

How we got here

Bird proposed a simple model of three factors for why a land trap develops:

These three crucial factors - the fixed amount of land available, its immobility, and the fact that is does not naturally decay - are what mark it out from every other asset in the world. They explain most of why land assets are still such a store of enormous wealth today, even as its earliest power - are so much less relevant in modern economies and political systems. And they are also why land can be so profoundly dangerous to our collective prosperity today.

This model may well be fairly explanatory… but it only plays out as a sort of arms race when density (read: height) becomes a factor.

In chapter 5, Bird comments on the effects of urbanisation:

Cities across the Anglosphere, From Sydney, to London, to New York, and Vancouver, hosted the political movements that brought land to the centre of political debate. But at the end of the nineteenth century, those booming cities were oddities on the global stage. Most of the world was neither industrial nor urban, and in most of it, the very old models of land ownership persisted. When the guns finally fell silent at the end of the Second World War, less than a third of the world’s population lived in cities at all. As late as 2006, the population of the world as a whole was still more rural than it was urban. Until very recently, the lives of most people across the world would have been relatively recognisable to their ancestors from a thousand years earlier.

Cities may well be the wellsprings of civilisation, but the benefits of agglomeration aren’t equally distributed. Bird summarises the history:

It was the sweat of workers and the innovation of entrepreneurs that made the land increase in value, but they received a much more modest share of the wealth than their idle landlords. Only land, since it was both fixed and immobile, could explain the disparities between such new and tremendous wealth and the crushing poverty with which George was personally acquainted. When any piece of technology or innovation or effort boosted production, landlords simply raised the rents. They needed to do nothing more than collect the income, which was ballooning in the face of the growing urban population. When new infrastructure was built - the roads, bridges, railways and utilities that were being blanketed over urban America - land prices went up, and landlords vacuumed up the benefits, at the expense of the people who built, used and funded those same developments.

This is the outcome of Birds thesis. Land owners are rewarded outsized returns for the activities conducted on their land, increasing the value of the land itself. A self fulfilling prophecy emerges - never to be stopped.

What is to be done?

Bird’s conclusion is as stark as it is scary. Land traps are real, and seemingly irreversible in the short term1. Once an economy becomes addicted to high real estate values and mass homeownership, reform becomes politically and economically perilous.

Bird summarises:

Few places have managed to avoid the pitfalls of the land trap, and none have managed to escape it entirely once the wheels have been set in motion. For nations both rich and poor alike, understanding that the trap exists at all - that land wealth is not like any other wealth, and that it can have pernicious and long-lasting outcomes - is the first step.

So what is to be done? Well firstly, we need to make a judgement call on the claim that buying (relatively cheap) land and not doing much with it encourages more speculative buying. If this is true (and there’s good evidence for it in increasingly dense environments), then the natural follow on outcome is that there is a logical investment strategy to buy land, drive up prices and force out actually useful, value-creating forms of investment (e.g. residential housing or commercial property).

From here, there’s a handful of (at least partial) solutions:

Solution 1: Introduce a land value tax

Instituting a land value tax counters speculation and land banking because it makes holding onto land expensive. In theory, investors stop buying land unless they are able to use the land productively enough to generate more value than it costs to own. The tax rate can be adjusted to equate land yields to returns on other asset classes.

From here, whether it’s developing and selling property, owning rental properties, farming, or through commercial assets - the maths is quite simple. Does the yield outweigh the annual tax?

While some countries (Australia and the UK included) have dipped their toe in this policy arena - namely through vacant property taxes2 - broader land value taxes are a political non-starter.

Solution 2: Radical state led housing

An alternative is to reshape the psyche of just about every western country that home ownership actually isn’t the cornerstone of both the family (and community). Instead, the state is the natural provider of accomodation (at least in dense urban areas).

Good luck with that as a campaign platform.

Bird spends a lot of the book discussing city states like Singapore or Hong Kong3 . These countries may be at an advantage here given their wide sweeping subsidised housing programmes, but it’s their exception that proves the more common rule.

There is a fundamental need for far more state provided housing for the disadvantaged of course, but this is more a of a social welfare policy adjustment, rather than a rewiring of the national psyche round property rights.

Solution 3: Build?

The Land Trap book falls silent on supply side solutions, presumably because you can’t make anymore land. Fair. While this is obviously true, in most dense urban environments you can build up, rather than out.

Building more apartments - especially larger 3 or 4 bedroom apartments - in dense urban areas presents a simple solution. Zoning is often the barrier here (Grattan has been banging this drum loudly) rather than the demand side of the equation, namely stigma against apartment living.

The trouble with this solution is the implementation. Any proposed policy changes to zoning immediately give windfall gains to the incumbent land owners, feeding back into the original land trap. Sure a developer will be happy to sell you a flash 3-bed apartment, albeit at an exorbitant price given the skyrocketing land value.

If there’s one takeaway here, it’s that the land trap is a one-way door. We’re already through the looking glass. Reversing it will be difficult, if not impossible.

The question remains: how do we manage?

Subscribe now

Subscribe now

Hearts in the right places

Burnham and Rotheram’s Head North details the personal motivation for better representation in the North of England. It’s a compelling story, particularly in a time where the Labour party is desperately searching for an economic narrative and scrambling to maintain a healthy dose of economic talent.

Head North explains why the Hillsborough disaster is symptomatic of Westminster1 as a whole. The disconnect between London (and more broadly the South-East of England) is engineered out of centuries of mistrust between the have and have-nots.

The book is written in an unusual format. A chronological memoir where each co-author takes turns to explain the key events of their lives (many of which are the same) from their own point of view. It’s a little clunky - yet if the moral of the story is one of greater unity, it’s certainly on brand.

It’s the third political memoir / manifesto on the future of Westminster I’ve read this year. Yet it brings a welcome, homely, fresh angle to the woes facing the UK compared to both Great Britain? and Failed State.

What is to be done?

Burnham and Rotheram recap the recent history of funding for Manchester and Liverpool (spoilers: it’s going down). Thatcher unsurprisingly gets a good mention for stripping capital planning powers out of councils back yonder, which has never recovered.

The lack of funding flowing north has a much deeper history than Burnham and Rotheram explain. The UK doesn’t rely on a union or federation of independent states (like Australia, Canada, or even the US) that advocate stringently for their regions. Running the country out of Westminster is a feature - not a bug.

Isn’t this the exact role of elected MPs you say? Pragmatically speaking, not anymore. Well certainly not in a two-party system where Whips and seniority control the way the overwhelming majority of MPs vote (especially in the Labour party).

Half of Burnham and Rotheram’s solution is warranted. Capital budgets determined at a local level are more likely to prioritise the ground needs of residents and businesses. There is less information loss between problem and solution, and the temptation for pulling political strings is lower if decisions makers are also a member of the local community.

Yet some realism here is warranted too. Mega-projects (inter-city rail, major new highways, port upgrades) are never going to be locally funded. Not only does council not have the capital, they don’t have the procurement or management talent to run the design/tender/build.

The answer, as with most things, lies somewhere in the middle.

What isn’t to be done?

The other half of Burnham and Rotherham’s solution — scrapping Green Book processes2 for evaluating state significant investments - is far less convincing. Burnham talks of his own experience in HM Treasury:

But I asked the civil servants: why could not a single project in the North of England be identified? That was when the scales fell from my eyes.

‘No project in the North passed the Green Book, Minister,’ said the senior official.

‘How is that possible when the railways of the North are falling apart?’ I countered.

‘Because the Green Book appraises projects on what they will generate for the UK economy within a twenty-five-year period and most projects in the North don’t return enough to meet the threshold.’

If anything had brought me into Parliament, it was a mission to close the North–South divide. As I patiently listened to this explanation, sitting in the splendour of the chief secretary’s office, with its magnificent view onto Horseguards and St James’s Park, I finally realised why it was mission impossible. The North–South divide was no accident. It was the product of permanent UK policy and would continue getting wider for as long as the Green Book remained in place.

While the Treasury was never my natural habitat, I regard my brief time there as critical in my political journey. By seeing how things worked from the inside, I finally understood why England has always felt like two countries. Only the EU funded infrastructure on the basis of social deprivation. The UK government’s economic test was a recipe to give more to the places that already had most and widen regional divides. This is what people call ‘Treasury orthodoxy’. It is a doctrine based on cold numbers which takes a negative view of our own country. In essence it says: those old industrial communities can’t become anything again so why invest in them? Let’s just invest in the places that will provide a quicker economic and financial return. Up until this point, I had assumed that the bias against the North was cultural and emotional – a product of the fact that people from the South tended to dominate positions of power across society and made decisions to favour it. Whilst it is partly that, I genuinely hadn’t realised that it is supplemented by a bias that is hard-wired into the processes which govern the country. Under this system, the North could cry out for investment for years and would never get it and we have seen that play out ever since.

It’s incredible that this angle is being championed from Burnham who once upon a time was chief secretary3 to the Treasury. A move away from ‘Treasury orthodoxy’, including scrapping the Green Book entirely, creates a common enemy to rally against for those who feel like they are missing out. The trouble is what comes next:

The time has come to tear up the Green Book and the Barnett Formula4 and replace them with a modern funding formula which works for the English regions and the home nations.

What is this magical alternative formula? The authors fall silent.

If spending of state funds aren’t to be evaluated through cost-benefit analysis a the Green Book recommends, and ‘social’ considerations are given bigger priority - the Westminster establishment is going to win at that game every day of the week. Westminster politicians can and will rely on the argument that (expensive) projects in London benefit far more people that ‘low utilisation’ projects in the North.

Rather than keeping the North unfairly deprived, the Green Book stops politicians from funnelling cash to pet projects in London.

Instead, stick to the real (boring) solutions

Credit where credit is due - it is true that HM Treasury needs serious reform. The current fiscal rules (discussed extensively in both Great Britain? and Failed State) are absolutely bonkers.

Treating opex and capex as substitutable when balancing the budget books causes politicians to cancel long term infrastructure projects in order to meet the Treasury rules. The people that lose out on this the most live in the lowest density areas - namely the North. The current Labour government has all but committed to stick with them (as to not enrage the fiscally conservative in the party), much to the dismay of most modern economic thinkers.

The saddest part of all this is it’s blatantly obvious that councils are excellent at service delivery (especially for the very young and very old) compared to their Westminster cousins.

More funding flowing out of Westminster to the north for social housing or localised transport would give England’s former industrial powerhouses (Manchester, Liverpool, Birmingham) a real chance of reigniting their engines.

Evolution, rather than revolution, gives the North a real chance.

Subscribe now

I thought I already knew Bayes, I was wrong

Tom Chivers’ Everything is Predictable goes beyond popular science. It’s really a crash course in the history and development of Bayes’ theorem.

Dry you say? Quite the opposite. I devoured this book over the past summer in Australia then gave away my copy to a friend. When I moved to London I was referencing it so often I had to go out and buy another.

Who is Bayes again?

Bayes’ theorem is named after 18th-century British statistician and theologian, Thomas Bayes. His remarkable (yet simple) theorem describes how to update the probability of a hypothesis based on new evidence.

Why is this important? Well, in traditional frequentist statistics (that you get taught in economics, psychology, medicine, and engineering), probability is interpreted as the long-run frequency of events in repeated trials (i.e. sampling).

A p-value from a sampling distribution tells us:

What is the chance of seeing this result, given some hypothesis?

Bayesian statistics, on the other hand, allows us to incorporate prior knowledge or beliefs into our analysis and update those beliefs as we gather more data. The central question of Bayes’ Theorem therefore is:

How likely is the hypothesis to be true, given the data I’ve seen?

This isn’t wordplay or semantics - it’s a completely different way of seeing the world.

We can write Bayes’ theorem using probability notation.

Where:

P(A|B) is the posterior probability (updated belief about A after observing B)

P(B|A) is the likelihood (probability of observing B, given that A is true)

P(A) is the prior probability (initial belief about A, before seeing B)

P(B) is the evidence (total probability of observing B, across all possible values of A)

Flat priors aren’t to be taken for granted

Chivers does a marvellous job in stepping through the history of Bayes’ theorem, and where it departs from the frequentist canon.

We see in the denominator P(A) above, Bayes’ theorem relies on our initial belief about A before seeing B. This is implicit in the frequentists approach too. It’s just that they - without thinking about it - put equal possibility on each possible belief (we’ll start calling them priors from here).

This creates a novel issue, known as Boole’s objection. Sometimes, equally weighting ‘A before seeing B’ involves making big assumptions.

Chivers explains:

Imagine an urn filled with balls, either black or white. If you have a flat prior on the total number of black balls in the urn, then any given mix of black and white balls is equally likely. (If there are only four balls in the urn, you know have three possibilities - two black, one black and zero black - and they’re all equally likely).

But if you assume that each ball is equally likely to be black or white - a flat prior on the probability of drawing a white or black each time - then your prior probability favours (very strongly if there are lots of balls) a roughly fifty-fifty mix in the urn as a whole.

What’s in a prior anyway?

The importance of priors becomes obvious when thinking about disease diagnosis.

Say you go to a doctor for a routine cancer screening. The disease (say, a rare cancer) occurs in 1 out of every 1,000 people.

The test is very accurate (only gives a false positive 2% of the time), but not perfect:

• If you have cancer, the test is positive 100% of the time.

• If you don’t have cancer, the test still gives a false positive 2% of the time.

You take the test, and your result is positive. What is the probability you actually have cancer? Most of us bush mathematicians default to ‘well of course it’s 98%’.

The truth is far more interesting when we plug in the numbers into Bayes formula:

Even though the test is 98% accurate, the rarity of the disease means that most positives are false alarms.

In a group of 1,000 people:

• 1 person has cancer → and they test positive.

• 999 people don’t have cancer → and the test produces 20 false positives (i.e. 2% of 999).

• In total the test shows 21 positives, but only only 1 out of 21 people who receive positive test results (4.8%) actually have cancer.

From first principles

There’s one thought experiment that Bayes covered in his seminal work The Doctrine of Changes that really cracks this whole priors thing open. Chivers explains:

To make his point, Bayes used a metaphor of a table, upon which balls are rolled…The table is hidden from your view, and a white ball is rolled on it in such a way that its final position is entirely random: ‘There shall be the same probability that it rests upon any one equal part of the plane as another.’

When the white ball comes to a rest, it is removed, and a line is drawn across the table where it had been. You are not told where it is. Then a number of red balls are also rolled onto the table. All you are told is how many of the balls lie to the left of the line, and how many to the right. You have to estimate where the line is.

Imagine that five balls are thrown, and you’re told that two of them landed to the left of the line, and three of them landed to the right.

Where do you think the line ought to be? Bayes said that the most likely place is three-sevenths of the way up the table from the left. Intuitively you might think that it should be two-fifths. After all, you’ve just rolled five balls, and two of them ended up on one side, and three on the other.

But Bayes said that you must take into account the prior probability – your best guess of what the situation was, before you got any information. But do you have a best guess? You don’t know anything, do you? The line could be anywhere. But that in itself is a form of prior information: it is equally likely (from your subjective point of view) that the line is right up against the left-hand cushion, or right up against the right-hand cushion, or anywhere in between.

You could draw a graph of the distribution of probability – how likely the line is to be in a given place on the table, before you had rolled another ball. If you have absolutely zero idea where the line is, then the probability of the next ball landing to the left of it is 0.5 – 50 per cent. After all, the line could be far to the right, so the ball would definitely land left; it could be far to the left, so the ball would definitely land right; it could be in the middle, so it would be fifty-fifty; or it could be anywhere else, with corresponding probabilities. The average position is exactly in the middle.

Essentially, Bayes’ big insight was that you must add any new information you get to the information you already have. In this case, you don’t have very much information. But it is something. What that means is that instead of just saying, ‘The most likely position of the line is two-fifths of the way along the table,’ you have to take account of your prior.

So Bayes said that the equation for working out the probability here is not ‘the number of red balls on the left divided by the total number of red balls’ – 2/5 – but the number of red balls on the left of the line PLUS ONE, divided by the total number of red balls PLUS TWO. It is, says Spiegelhalter, ‘equivalent to having already thrown two “imaginary” red balls, and one having landed each side of the dashed line’. That might seem odd, but it makes sense when you think of what it would look like if all the balls had landed on one side or the other. If all five had landed left, and we didn’t include those extra imaginary balls, then we’d say the probability of the next ball landing left was 5/5, or 1, or complete certainty.

But that’s silly – obviously you don’t know for sure that the next ball isn’t going to be to the right. With Bayes’ extra balls, your estimate would be 6/7. And no matter how many balls land on one side, you never end up with absolute certainty – if a million balls land to the left, then your estimate of the next ball’s probability of landing right would be 1/1,000,002. Each piece of new information pushes you closer to certainty, but you never quite get there.

What if I don’t know the prior? Yes, that’s an issue

Okay. The balls rolled on a table thought experiment makes enough sense. Priors are essential.

But, if we know very little about the world, our priors are going to be fuzzy (or at the very least, subjective). This matters.

Chivers drives this home by expanding on the ‘balls in an urn’ example:

In Bayes’ imaginary not-in-fact-a-billiard table, he assumed that it was equally likely that the white ball could be anywhere on the table. That’s called a uniform prior. That’s defensible – you can imagine that if you throw the ball hard enough it’ll be essentially random where it lands. But what about situations where you’re completely ignorant and don’t have good reasons to assume any prior?

A more technical objection was that of the mathematician and logician George Boole, who pointed out that there are different kinds of ignorance. A simplified example taken from Clayton: say that you have an urn with two balls in it. You know the balls are either black or white. Do you assume that two black balls, one black ball and zero black balls are all equally likely outcomes? Or do you assume that each ball is equally likely to be black or white?

This really matters. In the first example, your prior probabilities are ⅓ for each outcome. In the second, you have a binomial distribution: there’s only one way to get two black balls or zero black balls, but two ways to get one of each. So your prior probabilities are ¼ for two blacks, ½ for one of each, ¼ for two whites.

Your two different kinds of ignorance are completely at odds with each other. If you imagine your urn contains not four but 10,000 balls, under the first kind of ignorance, your urn is equally likely to contain one black and 9,999 whites as it is 5,000 of each. But under the second kind of ignorance, that would be like saying you’re just as likely to see 9,999 heads out of 10,000 coin-flips as you are 5,000, which of course is not the case. Under that second kind of ignorance, you know you’re far more likely to see a roughly 50–50 split than a 90–10 or 100–0 split in a large urn with hundreds or thousands of balls, even though you’re supposed to be ignorant.

So which prior do we assume? Do we think the colour of the balls is independent or correlated? You may say that you assume perfect ignorance, but there are different kinds of ‘ignorance’, and you have to pick one.

But the underlying problem of Bayesian priors is a philosophical one: they’re subjective. As we said earlier, they’re a statement not about the world, but about our own knowledge and ignorance. And that’s… uncomfortable.

Think priors are subjective? P-values aren’t much better

A lot of the criticism of Bayesian methods over the years has been the subjective nature of assigning priors. The frequentists have elegance on their side, even if there is an argument that they are measuring entirely the wrong thing (i.e. what is the chance of seeing this result, given some hypothesis?).

Chivers bluntly describes that p-values are wholly arbitrary and subjective too:

If a study finds that a p-value of 0.05 or less, that means that you’d only see those results (or more extreme ones) by chance one time in twenty at most, doesn’t it? So surely, if every study is using that yardstick, you wouldn’t expect to see many false positives?

That is the idea, sure. But it’s not as straightforward as that. The easiest way to get a p<0.05 result - that is, something that you’d only see by coincidence one time in twenty - is to do twenty experiments, and then publish the one that comes up.

Who’s methods are subjective now?

What now?

This is one of those books that left me with more questions than answers. Where do I find good base rate data? Can I still apply Bayesian methods if there’s asymmetric information on priors? Aren’t most LLMs (that rely on predicting the next token) just using a Bayesian approach?

While Chivers’ book didn’t cover everything you could want to know on Bayes, priors, and the limitations of the frequentist approach, there’s a few glaringly obvious takeaways:

1. Priors (e.g. base rates) are crucial, and if you don’t think about them actively you are assuming they’re implicitly flat (i.e. 50/50) anyway

2. P-values are bullshit and arbitrary

3. It’s difficult to make Bayesian methods the default wiring of my brain

Chivers’ book brain wormed me good, and I’m better for it.

Subscribe now

Subscribe now

A lot has been written about Abundance since it was released in March 2025.

I found myself re-reading the book in August once it had well and truly entered the policy wonk and media mainstream.

Back in 2021, Klein and Thompson wrote (seperate) op-eds in the New York Times and Atlantic which set up the early Abundance themes. The book expands on these ideas, being pragmatic without being preachy.

Many reviews over the past 6-months have called Abundance ‘a manifesto for the left’ or ‘a rally cry to embrace supply side economics’. Noah Smith went as far to say ‘the basic thesis of this book is that liberalism — or progressivism, or the left, etc. — has forgotten how to build the things that people want’.

It seems Klein and Thompson have correctly identified that Democrats have over indexed on the ‘managing’, while almost entirely forgetting about the ‘making’. Not only is this bad retail politics, it’s a catastrophic approach to government.

I’ll explain.

Scarcity is a choice we make by failing to reform

One of the most pressing themes of the book is that the environmental reform of the 1970s’ and 1980’s was necessary for the time. But, that time has gone.

Modern day reform (and repeals of what was needed 40 years ago) is just as necessary:

One generation’s solutions can become the next generations problems. After World War II, an explosion of housing and infrastructure enriched the country. But without regulations for clear air and water, the era’s builders despoiled the environment. In response, the US passed a slew of environmental regulations. But these well-meaning laws to protect nature in the twentieth century now block the clear energy projects needed in the twenty-first. Laws meant to ensure that government considers the consequences of its actions have made it too difficult for government to act consequentially. Institutional renewal is a labor that every generation faces anew.

Retaining the reforms of 50-years ago dosen’t help us make progress today. Government (and I’d add the voting public) needs to update their priors.

Managers, not makers

I was asked recently why the north of England has such lower productivity (40 per cent in many areas) compared to London. Infrastructure was my first answer. Our ability to plan, finance, build, and maintain infrastructure underpins productivity growth. The slower we get at ‘making’ (including due to ticking many regulatory boxes), the more vicious the productivity trap becomes.

Klein and Thompson explain:

Now imagine dozing off for another thirty-year nap between 1990 and 2020. You would wonder at the dazzling ingenuity that we funneled into our smartphones and computers. But the physical world would feel much the same. This is reflected in the productivity statistics, which record a slowing of change as the twentieth century wore on. This is not just a problem for our economy. It is a crisis for our politics. The nostalgia that permeates so much of today’s right and no small part of today’s left is no accident. we have lost the faith in the future that once powered our optimist. We fight instead over what we have, or what we had.

Homes haven’t always been ‘assets’

The ‘financialisation’ of housing (or lack of a better term) seems age-old… but it’s really only been with us for 50-odd years. No where is this trend clearer than in Australia (and especially in capital cities). But the US has suffered a similar fate:

Prior to 1970, housing wasn’t a prime asset. You bought a home to live in it.

What impact has this had? Quite a bit:

In the ‘70’s, rising inflation and slowing home building turned the homes people did own into the center of their wealth. But how do you protect the value of that asset? You can insure a home against fire, but you can’t insure it against rising crime rates or local schools slipping in quality or a public housing complex being built down the block.

To manage those risks, you need to control what happens around your home. You do that through zoning and organising. You do it through restricting how many homes and what kinds of homes can be built near you…

It’s hard to look back nowadays and realise the current status quo wasn’t always the case. Klein and Thompson sum up the current state well:

Those who already lived in a place were its stewards, its guardians, its voice. Those who wanted to move to that place were recast as a consumptive horde.

NIMBYism isn’t just prejudice. It’s a rational financial strategy if the family home is both your store, and your engine, of financial security.

Think bigger on NIMBYism and democracy itself

Klein and Thompson make a strong argument that we’ve overcorrected on ‘challenge culture’ in building the physical world we need:

Changing the processes that make building and investing so hard now requires confrontations wth whether the systems liberals have built really reflect the ends they’ve sought. Much that was designed to foster grassroots participation has been captured by incumbents and special interests. It can be difficult, in a raucous town meeting, to look around and remember who is not there: the mother working two jobs, the young family who couldn’t afford the apartment they so badly wanted to move into.

It’s a long overdue reminder that the folks that have the ability (and to be frank, time) to participate in ‘challenge culture’ are usually those with means. This skews the infrastructure that gets built or maintained away from public transport, public hospitals, and public schools.

The 70’s green movement started as a legislative quirk

It was news to me that the original environmental movement in California started as a quirk out of the case Friends of Mammoth v. Board of Supervisors of Mono County.

A HOA sued to stop condos being built near a popular ski resort. The clever part of their argument was that while not a state project:

…yes, actually, it was, because any development that required public permits to be built was inherently a public project.

Friends of Mammoth lost the case in the lower courts but appealed up to the state Supreme Court, which ruled in their favor in a 6-to-1 decision.

The impact of this can’t be understated. All the environmental checks and balances designed to stop state projects trashing the environment now applied to any construction project in which the state regulates private activity (hint: that’s pretty much everything).

Klein and Thompson summarise:

The problem we faced in the 1970s was that we were building too much and too heedlessly. The problem we face in the 2020s is that we are building tool little and we are too often paralysed by process.

In an overly legalistic culture like the US, that egg is going to be hard to unscramble.

Track what gets built rather than the cheque size

If there were one word to describe Abundance, that word would be ‘build’.

Klein and Thompson rightly point out that governments have stopped describing what they intend to build. Instead, ‘capex programmes’ or ‘budget measures’ are the modern day nomenclature:

This is how the scale of such bills is normally described in Washington: by a price tag. The more money, the bigger the bill. That is an incomplete measure, at best.

If bigger cheques (think IRA) mean more stuff gets built, fine. Too often, the two aren’t that neatly related. The left well and truly has learnt that lesson. And now:

The climate crisis demands something different. It demand a liberalism that builds.

The Californian high-speed rail system (familiar to anyone who’s every googled ‘progress studies’) is the perfect example of where capex allocated (and in many cases spent) does not equal infrastructure delivered. For the Brits among us, the fiasco that has become HS2 sings a very similar tune.

What are the costs in a cost blow-out?

Cost blow outs on major infrastructure projects make headlines. But the anatomy of a cost blowout is woefully understood. It’s not bad forecasting on the cost of steel or more rainy days than a builder had hoped for.

Klein and Thompson explain that it’s time and process that empty the till:

What has taken so long on high-speed rail is not hammering nails or pouring concrete. It’s negotiating. Negotiating with courts, with funders, with business owners, with homeowners, with farm owners. Those negotiations cost time, which costs money. Those negotiations lead to changes in the route or the build or the design, which costs money. Those negotiations lead to public disappointment and frustration, which leads to loss of money that might otherwise have been approved if the project were speeding towards completion.

So, are negotiations efficient in that they save re-work down the line? Not necessarily, because each year a project is delayed is more compound costs (financing, staff salaries for lawyers and project managers, media and communications). There is a balancing act here of course. But governments and the public need to get more comfortable that some issues should indeed be left to fix later.

Trust, but verify

Why does has the US struggled so much to build? Part of it is culture… but this administrative burden is also by design. Klein and Thompson explain:

Americans have always mistrusted the government. They’ve particularly mistrusted centralized power. But they also need a government able to wield power. They want the good a government can do.

Given these constraints, a quirky ‘solution’ evolved:

Adversarial legalism was a way of reconciling the government we wanted with the suspicions we harbored. America is unusually legalistic. It always has been.

How has this changed the psyche of those responsible for infrastructure?

When you make legal training he default training for a political career, you make legal thinking the default thinking in politics. And legal thinking centreres around statutory language and commitment to process, not results and outcomes.

And so what’s the ‘end state’ we’re left with?

There is nothing wrong with lawyers. There might be something wrong with a country or a political system that needs so many of them and that makes them so central to its operations. That might be a system so consumed trying to balance its manifold interests that it can no longer perceive what is in the public’s interest.

Bring back a business model for basic research

Klein and Thompson also explain that the emphasis on process rather than outcome has infiltrated the means of creating knowledge itself. The war on ‘basic research’, that is research without a defined practical end, has been widespread across Western universities and labs for the best part of 30-years.

The numbers back this up - especially when you look at NIH grants:

“We find that evaluators uniformly and systematically give lower scores to proposals with increasing novelty”.

This trend in science funding quickly produces a race to the bottom:

Bias against novelty, risks, and edgy thinking is a tragedy, because the most important breakthroughs in scientific history are often wild surprises that emerge from bizarre obsessions.

Klein and Thompson put in a more memorable fashion:

Too many projects get funding because they are probable.

What is the point of science after all?

An idea whose time has come

Abundance resonated with so many because it felt apt for the time:

We are in a rare period in American history, when the decline of one political order makes space for another.

For too long, the left has paraded credentials and followed well formed processes, all while building far too little. The era of renewal is due:

Throughout the 2010s, a slow economic recovery fueled public resentment of inequality, and an affordability crisis gathered steam. In 2020, the pandemic obliterated many Americans’ trust in government, or what was left of it. And between 2021 and 2024, inflation brought national attention to our interlocking crises of scarcity, supply, and unaffordability. For years, the boundaries of American politics has felt fixed, even settled. But now they are falling.

Subscribe now

Subscribe now

A recent cult classic that I’d never read

Super-forecasting has rapidly become a cult classic in the rationalist/economics/tech canon. I found myself noticing references to it everywhere, usually somewhere in the same sentence as ‘updating priors’.

In the past year, I’ve also become fascinated with prediction markets. These have created an economic framework for super-forecasters to both show off, and get paid for their skills. More broadly, the fact that some people are much better at predicting the future than the rest of us seems non-obvious… and worthy of intense study.

Super-forecasting was both an enjoyable and useful read. My primary reflection is that what makes super-forecasters great at predicting the future is simple, yet rare.

Let me explain.

It started with the intelligence agencies, not finance bros

I’d always assumed that forecasting geopolitical events had stemmed from the finance sector. Placing bets on how markets would react to unfolding events or elections is almost the definition of a futures market. But the history is more interesting - it’s the intelligence community, particularly in the US, that historically have been heavily invested in better forecasting.

Tetlock partnered with IARPA in 2011 to run the Good Judgement Project. This aimed to identify cutting-edge methods to forecast geopolitical events. The result found that super-forecasters outperformed intelligence analysts with access to classified data. This is a pretty stunning conclusion. True ‘alpha’ in the intelligence community is not who has the best information - but who uses that information most wisely.

Look for the background rate, then add your own flavour

There’s a simple reason that super-forecasters repeatedly beat the rest of us. Tetlock and Gardner show it’s not raw intellect or fancy degrees - it’s a process.

Super-forecasters address highly complex (yet short term) questions by first looking for background rates of occurrence, regardless of whether it’s a topical issue of the day.

This is best done when you Fermi-ize a question into manageable parts:

Fermi knew people could do much better and the key was to break down the question with more questions like “What would have to be true for this to happen?”. Here, we can break the question down by asking, “What information would allow me to answer the question?”

There’s a useful analogy in the book. What’s the chances a specific family in the US (who happen to have an Italian last name among other traits) own a pet?

If we Fermi-ize this analysis, the ‘outside view’ is the probably that any family in the US owns a pet. Forget the personal characteristics for now… start with the base rate.

Easier said than done, of course:

It’s natural to be drawn to the inside view. It’s usually concrete and filled with engaging detail we an use to craft a story about what’s going on. The outside view is typically abstract, bare, and doesn’t lend itself so readily to storytelling.

This approach has its limits of course. Taking the long time average for a data series spanning multiple decades doesn’t work well where there’s been significant, ‘structural’ change to the event probability (e.g. extreme weather events).

Yet, discarding historical data must be done for very good reason, rather than just recency bias. If you’re thinking that this is essentially the Bayesian method - you’d be correct. Super-forecasters happen to be exceptional at updating their priors with just the right level of precision.

The word ‘likely’ is almost meaningless

As an economist I probably use the word likely as a crutch more than most. Turns out I am not alone. Yet in forecasting, it is dangerous to use such a vague term to make recommendations.

Does ‘likely’ mean 51% or 99%, or anything in-between? There’s a very big difference between taking a course of action with a 51% chance of success compared to something almost certain.

Sherman Kent, the so called ‘father of intelligence analysis’ tried to implement a new system for communications at the CIA. While it was never adopted, it remains a great rule of thumb in day to day conversation.

I may well adopt this myself.

Super-forecasting is skill, not luck

I was skeptical. If you have a group of people flip coins for long enough, one of them is going to flip 10 heads in a row. Is super-forecasting really just putting people in the right hand tail of a distribution on a pedestal? Apparently not.

Tetlock and Gardner explain:

The correlation between how well individuals do from one year to the next is about 0.65… So we should still expect considerable regression o the mean. And we observe just that. Each year, roughly 30% of the individual super-forecasters fall from the ranks of the top 2% next year. But that also implies a good deal of consistency over time: 70% of super-forecasters remain super-forecasters.

Is there some luck in being a super forecaster (e.g. being in the right tail of the distribution)? Of course… but 65% of forecasters aren’t just lucky… there’s a whole chunk of skill in there too.

The best forecasters update and change their mind often

Changing your mind often, especially in light of new data, seems obvious in theory. Yet in practice, almost no one does it.

Tetlock and Gardner explain the concept of ‘active open-mindedness’ goes beyond encountering new perspectives - it’s an ability to rewrite beliefs as needed:

For super-forecasters, beliefs are hypotheses to be tested, not treasures to be guarded. It would be facile to reduce super-forecasting to a bumper-sticker slogan, but if I had to, that would be it.

What’s more, the data from forecasting competitions proves this:

Super-forecasters update much more frequently, on average, than regular forecasters. That obviously matters. An updated forecast is likely to be a better-informed forecast and therefore a more accurate forecast.

But hang on… how much of this is that super-forecasters are just news junkies watching every scrap of new information that becomes available? That’s been tested too:

…super-forcasters’ initial forecasts were at least 50% more accurate than those of regular forecasters. Even if the tournament has asked for only one forecast, and did not permit updating, super-forecasters would have won decisively.

Tetlock and Gardner are quick to point out that a forecaster can both under and overreact to bad news. Overreaction is a fundamental misunderstanding of priors (e.g. just taking the last data point rather than adding that data point into a time series).

However, under reaction seems to be more deep rooted:

…case of what psychologists call “belief perseverance”. People can be astonishingly intransigent - and capable of rationalising like crazy to avoid acknowledging new information that upsets their settled beliefs… More commonly, when we are confronted by facts impossible to ignore, we budge, grudgingly, but the degree of change is likely to be less than it should be.

This makes sense. Splitting the difference or ‘meeting in the middle’ feels like progress (and it generally is in the right direction - but it doesn’t make you correct).

Is this under reaction a function of ego? The authors propose that super-forecasters:

…aren’t deeply committed to their judgements, which makes it easier to admit when a forecast is offtrack and adjust

Is this useful? I imagine it’s incredibly hard to disassociate from the outcomes if you’re a member of the intelligence community.

So what’s the answer? Change your mind as frequently as the data changes:

A forecaster who doesn’t adjust her views in light of new information won’t capture the value of that information, while a forecaster who is so impressed buy the new information that he bases his forecast entirely on it will lose the value of the old information that underpinned his prior forecast. But the forecaster who carefully balances old and new captures the value in both - and puts it into her new forecast. The best way to do that is by updating often but bit by bit.

Don’t bother trying to forecast more than a few years out

Tetlock and Gardner don’t talk about chaos theory by name in the book - but they hint at the same conclusions.

There’s a whole section that recounts the basics of Nassim Taleb’s The Black Swan regarding fat tailed distributions. It’s important that forecasters recognise the size of the tails in the distribution. Some tails are small (human height) and some are enormous (human wealth). The non-zero probabilities on the outer extremes are worthy of consideration.

Yet, even if we assume there are fat tails… that doesn’t help long term (e.g. 10-year) forecasts for sensitive variables like geopolitics:

Taleb, Kahneman, and I agree there is no evidence that geopolitical or economic forecasters can predict anything ten years out beyond the excruciatingly obvious - “there will be conflicts” - and the odd lucky hits that are inevitable whenever lots of forecasters make lots of forecasts. These limits of predictability are the predictable results of the butterfly dynamics of nonlinear systems.

The sweet-spot for super-forecasters seems to be predictions for events less than 18-months away.

Unresolved questions and open loops

First, I there’s a sequel needed for this book to understand how super-forecasters are interacting with the financial markets. They should all be filthy rich is one assumption… however I suspect it’s a mixed bag.

Why aren’t the folks who are brilliant at forecasting geopolitics either making a fortune themselves on polymarket, or consulting for traditional quant houses? Maybe they already are… there’s certainly a book in that story.

Second, AI models are extremely good at weighting variables and updating priors. I don’t think this means AI assisted forecasters will be able to forecast farther into the future… more so that the fidelity of a updating a prediction (e.g. is it 60 per cent or 62 per cent) may get more accurate.

I also imagine an AI model can learn your biases quickly and help course correct your under/over reactions if given enough data.

If I had to make a bet, I think we’re going to see more super-forecasters in the next 5-years then we have since measuring Brier scores began.

Subscribe now

Subscribe now

I too am now a migrant

I’ve lived overseas before, but always with a rough return date in mind. Now, moving to London at age 29, for the first time in my life I truly identify with the word ‘migrant’.

Most Australians, wherever they are in the world, have a natural interest in - if not strong opinion of - migration. Australia prides itself on being one of most multicultural nations. Over half of Australians have a parent born overseas. Malcom Turnbull goes further proclaiming Australia as ‘the most successful multicultural society in the world’.

Hein de Haas’ How Migration Really Works cracked open my world view to the longer term ‘migrant story’.

Here in the UK, that story that plays out in the media is laced more-so with the word ‘threat’ than ‘opportunity’. Some of this is cultural. Australia is far more egalitarian both economically and socially. Yet part of that must be a fundamental misunderstanding of the history, nature, and forward-look for migration.

The faltering UK economy since 2008 has everyone searching for answers. Some are seeking to assign blame too. Alas correlation between slow economic growth and widely publicised negative migrant stories lead pundits to a pretty simple conclusion.

And yet, I still felt this story as I understood it was incomplete. This book was a solid crash course in why.

De Hass structures it in terms of ‘myths’ that are firstly explained, then corrected. I found myself flicking back and forward in the book to align different arguments or counterarguments that seem to conflict. What I now know is that like many things, the truth resists simplicity. I guess that’s why it’s never been easier to have a (strongly held) opinion on migration.

I’ve recommended this book to more people than I can count (niche, I know).

Here’s my best attempt in summarising just what I learnt.

Immigration is not new or particularly novel

This insight seems obvious in hindsight, but I didn’t know it before reading the book:

Current levels of international migration are neither exceptionally high nor increasing. In fact, over the past decades, global migration levels have remained remarkable stable.

De Hass even crunched the data from the United National Population Division for 1960-2017:

So, if we express the number of internal migrants as a share of the world population, we see that relatively levels of migration have remained stable at around 3 per cent.

I would have thought migration rates would have been more correlated with major geopolitical events, but as I discovered - it’s more a function of regional economics.

It takes money (and aspiration) to migrate

Most migrants come from middle income countries. They have some access to capital and education/qualifications, and most importantly have tangible aspiration - largely driven by peers or networks that have also migrated.

The push and pull model of migration (popularised by Lee in 1966, and taught to me in university in 2014) seems not to stack up to the modern evidence.

In the most extreme cases this model may be true (e.g. fleeing on a leaking boat to escape persecution). But for the majority the ‘pull’ factors are far more significant to overcome the significant obstacles of migration.

De Haas explains:

…emigration is highest from countries with moderate and rapidly falling levels of population growth. This has less to do with demographic factors than with the fact that these are typically middle-income countries where economic development and increasing education have led to a general expansion of people’s aspirations and capabilities to migrate. In fact, countries with the highest population growth, such as those in sub-Saharan Africa, actually tend to have the lowest levels of long-distance emigration.

There is an execption to this trend. Labour shortages in middle-income countries (due to an ageing population) may start to attract migrants from nearby low-income countries. De Haas goes further saying:

China seems bound to become a major migration magnet

This trend seems important and not talked about enough.

Most migrants do have jobs to come to

While there’s a myriad of reasons to pack up and move, de Hass shows a very close link between business cycles and immigration levels, concluding:

Destination-country labour demand is the main driver of international migration.

There’s a clear narrative for this too:

Although illegal migration attracts the most attention, it’s important to emphasize that the vast majority of migrants move legally, and that this legal migration is predominately driven by destination-country labour demand. It’s a myth that most migrants just how up at the border without having any idea what to do. This may be true for some young adventurous - there are always exceptions - but most people will only move when they have a concrete job offer, because they were recruited, or because family or friends living abroad told them about vacancies and other opportunities.

Again, this point seems obvious in hindsight. It also aligns with my own experience1 and most around me.

The natural follow on question to the above is ‘well what jobs are migrants taking?’.

De Hass explains:

Contrary to what politicians tell us, most migrants do have jobs to come to. And, contrary to another popular belief, the biggest demand is not for higher-skilled workers by rather lower and mid-skilled migrants workers. Only a minority of immigrants fulfil the ‘expat’2 stereotype of engineers, doctors, and managers. While immigrants are still over represented in manual jobs in fields such as agriculture, factory work, domestic work, food processing, cleaning and landscaping, an increasing number of migrant workers to mid-skilled jobs as plumbers, carpenters, builders, chefs, nurses and other caregivers.

I imagine this trend is highly country dependant (e.g. Australia’s skilled worker programme, the UK’s recent crackdown on visa sponsorships, and the US’ green card lottery).

While I accept the premise, seeing the data on a country by country level seems essential.

Is an immigrant going to come and take my job?

The passages above re. immigration being a response to labour demand also explains why the ‘take my job’ narrative is silly. De Haas writes:

It is true that lower-income earners have barely benefited from economic growth over the past forty years, and they have seen job security and labour standards eroding while the rich only get richer. However, immigration has barely anything to do with this. The idea that immigration is a major cause on unemployment and wage stagnation is not grounded in any evidence, because what seems a causal connection is in fact a spurious correlation.

If only more Britons could read this passage. If anything, de Haas makes an argument that immigration is propping up the welfare state:

…instead of a threat to welfare states, the immigration of foreign workers is vital for upholding healthcare systems and providing care for children and the elderly, particularly in strongly liberalised economies like with UK and US and weak welfare sates like Spain and Italy.

Government’s aren’t at the helm of immigration demand

Every national government wants to show it has control of its own borders. However de Haas shows that migration demand is oftentimes outside of government control.

When economic growth is high and unemployment low, job shortages increase. This makes is more likely that migrants will apply for jobs, get hired and qualify for work visas if they need one. In this way, modern immigration systems have a built-in flexibility in which the numbers of legally admitted migrants automatically fluctuate with the state of the economy…

…The opposite is applies during times of recession. With labour demand going down and unemployment rising, migrants who lose their jobs are more likely to return home, while would-be migrants workers are more likely to postpone their emigration plans.

This explanation seems to fit for most advanced western economies, but it seems incomplete for a nation like Australia (and potentially Canada?) where international student migration doesn’t correlate neatly with the economic cycle.

De Haas rightly points out though that regardless of the real drivers, governments are quick to claim a win:

When the economy is tanking, fewer people will come. Of course, politicians will be keen to take credit and ascribe any change to their policies, but the next time immigration goes down, remember that rising unemployment is a more likely cause of the plummet.

Migrants look after themselves - if we let them

I’ve long thought that ‘settling services’ (loosely defined as state support for recent migrants) are critical to ensure effective social cohesion. That was until I became a migrant. In my experience once you have a job and a skeleton support network - you want very little to do with the state.

This experience seems to be common (if not obvious) for most migrants. De Haas makes a strong case that migrants integrate well into their new home countries as long as there’s no policy barriers:

Much more than official ideologies, what really seem to matter are bread-and-butter issues such as migrants access to education, work, and housing.

He continues:

The worst policies seem to be those that discourage or prohibit migrants and refugees from working. Nothing seems more detrimental for he well-being and economic contribution of migrants and refugees than to force them to remain in legal limbo zones for yeas because of administrative backlogs and appeals procedures.

This is a major win for policy makers. Rather than ramp up fiscal support, it’s the removal of policy barriers (which is far cheaper) than produces bang for the buck.

Migrants overwhelmingly pay their fair share

Many pundits comment on the ‘welfare narrative’ that migrants come and drain state resources with little financial contribution themselves.

De Haas explains that the evidence clearly doesn’t support this narrative. The ‘fiscal ratio’3 (fiscal contributions divided by fiscal costs) of foreign versus native people has been studied extensively over the past decades.

De Haas notes there’s some quite clear trends in just about every country:

Fiscal impacts change over immigrants’ lifetimes, typically following a U-shaped pattern - first positive, then negative, and then positive again as migrants get older. Recent immigrants tend to be net contributors to public finance, as they are generally young, employed, healthy and have no children. However, as migrants settle, marry, have children and become older, they may become a net cost to public finances, as they increasingly use public services like schools and healthcare. Fiscal impacts tend to turn positive again once immigrants’ children finish schooling and enter the labour market. However, when migrants grow into old age, their labour participation decreases and they are more likely to use health and elder-care facilities. This U-shaped pattern where fiscal costs change through the life cycle is not a particular feature of migrants, however, as it also applies to native workers.

What’s more, this impact multiples throughout generations. De Haas explains:

Things almost always improve as new generations grow up. As lower skilled immigrants and their offshoots climb the education and economic ladder, the average fiscal contributions of immigration improves as the new generations come of age.

Again, this shows the importance of removing barriers to work or education if we want to see an economic return from migration.

There’s no magic pudding in migration and growth

De Haas makes it clear that rich countries are the clear winners for migration. Taking the evidence they are net contributors to fiscal balances, and the economic theory they they will be self sustaining producers and consumers of goods/services - it’s a boon for the destination country.

While it’s true migrants do send large remittances home, the lost of human capital is equally as large. De Haas estimates the net effect on the origin country to be near zero:

While there can be little doubt that (freer) migration can considerably contribute to personal, national and eve global prosperity, the crucial question is how the benefits are distributed between origin and destination countries, as well as between richer and poorer members within societies. With regard to International inequalities, there can be little doubt that destination societies reap more economic benefits from migration that origin societies. This is because the labour of migrants primarily boosts economic productivity and profits in destination societies. Immigration benefits growth in destination countries by expanding the size of their population and economies and by stimulating innovation and investment.

With global remittances at an all time high (Afghanistan, Ukraine, CAR, etc), it’s good to remember that migrants are largely seeking better economic opportunities elsewhere for a reason.

Women on the move, with other women taking their place

De Haas describes an elegant narrative whereby western women (I think he really should just say ‘rich’ women) entering the workforce and having less children creates labour demand in the service sector:

So, the combined effects of increased education, women’s emancipation and falling birth rates have reduced the supply of local workers and increased demand for migrant workers in the service sector, who are increasingly women themselves. As Western families progressively moved away from the classic breadwinner model - with men working outside the house and women staying at home - the modern norm became that both partners worked. As a consequence, women have become less available to do the domestic chores such as cleaning, cooking, laundry and child rearing that were traditionally seen as ‘female’. This helped to fuel the growing migration of demand domestic and care workers.

This is a neat story, particularly if you add on senior and elder care workers.

Again, I suspect the inter-country differences here would be large.

Getting the crime causal effects correct

In both the US and the UK, the migrant/crime connection is bolstered by media and politicians alike. De Haas tackles this ‘myth’ head on:

A major survey conducted in 2010 revealed that one-third to a half of people in major Western immigration countries think that immigration increases crime, and between half and three-quarters think that illegal immigrants increase crime.

How valid is this perspective? It’s surprisingly difficult to measure: De Haas explains:

For instance, when immigrants settle in urban neighbourhoods that already had higher crime rates, correlations between immigration and crime can be spurious. As young men commit most crimes, an over-representation of migrants in crime statistics may just reflect the fact that many migrants are young men.

From the studies that do disentangle these relationships, what have they found?

Even though immigrants have on average lower levels of education and lower wages than the native-born population, most studies show that immigrants are generally less likely to commit crimes that the native-born. On average, neighbourhoods with high concentrations of immigrants have lower rates of crime and violence than comparable non-immigrant neighbourhoods. Another pattern is that, if immigrants are involved in crime, it is non-violent crime such as car theft and burglary - particularly among unemployed and poor immigrants - and they are heavily underrepresented in violent crimes such as aggravated assault, rape and murder.

This goes quite clearly against the mainstream narrative portrayed in both the US and UK media. So what about the ‘ethnic enclaves’ breeding poor behaviour narrative? Again, it’s been studied:

Another study that analysed trends of immigration and crime across England between 1971 and 2002 concluded that, in line with research from the US, neighbourhoods actually tend to become safer as more immigrants move in. It found that crime is significantly lower in ethnic enclaves where immigrants form at least 20 to 30 per cent of the population.

Key takeaways that have ‘updated my priors’

It’s the economy, stupid:

The reality of migration contradicts political narratives claiming that ‘we don’t need lower-skilled workers’. There is a real and consistent demand for lower-skilled migrant labour. In fact, the demand exists at all skill levels. Migrant workers are not as ‘unwanted’ as politicians claim. And they are generally not pouring in via desperate and irrational attempts to reach the Wealthy West. In reality immigration is primarily driven by labour demand.

Would be migrants would be taking a tremendously bad bet to rely on welfare in their new home:

The welfare magnet hypothesis is based on the assumptions that (1) many people migrate to live off benefits, (2) access to welfare is instantaneous and (3) immigration is free. All three assumptions are problematic.

‘Native’ born populations tend to be uncomfortable with ethnic enclaves (that they can easily see), however the evidence shows these communities are fundamental for migrants establishing themselves:

This crime-reducing effect was found to be particularly large in ethnic enclaves where immigrants from the same ethnic background concentrate, presumably because of the social control such communities can typically offer. This resonates with evidence reviewed in the previous chapter that, contrary to the ‘ghetto’ stereotype, the type of ethnic enclaves where new migrants tend to concentrate often provide community life that fosters informal social control, solidarity, and entrepreneurship.

Migration highlights where out existing social/justice systems aren’t up to scratch. When migrants become marginalised, they suffer from the same fate that all marginalised groups face:

Besides the problem of racial profiling, crime is strongly linked with economic marginalisation and particularly the socially, morally and mentally crippling consequences of racist discrimination and long-term unemployment. This shows these problems can only really be tackled if effective law enforcement is combined with policies that actively counter racial profiling and job market discrimination and give disadvantaged youth real opportunities for social mobility through education and work - regardless of their racial or ethnic background.

Migration isn’t going to fix plummeting fertility rates, simply because the number of migrants (which has been 3% of global population for decades) is simply too small, and migrants age just like the rest of us:

For the United Kingdom, keeping the support ratio constant would require more than 1 million immigrants annually - more than five times the average net migration in 2011-21 of around 200,000.

Final thoughts and unsolved questions

I now know that migration is a predominately function of economics and labour markets, despite what we hear from our politicians and media outlets.

Migrants, when they ‘settle’ well, are almost always net beneficial both socially and economically. It turns out the best things government can do to help migrants ‘settle’ is reduce any restrictions to starting work, and ensure kids can get to school.

These lessons seem basic in hindsight, but non-obvious when you observe out current political discourse.

After all this reading and thinking on migration, I’m left with many unresolved questions, to which I suspect there’s not a single answer:

• What is the ideal ‘level’ of migration to achieve sustainable growth and social cohesion? How much does this change country to country (e.g. Australia vs the UK vs China) depending on regional geography and historical societal connections?

• How should we manage migrants that are already in-country illegally (some for decades). Should the focus be on giving full work rights and access to education (the theory seems to say so)? However, this is terrible politics. I suspect the answer lies somewhere in the middle.

• What is the best way to process refugees and asylum seekers? That includes those that are found to be non-genuine (e.g. practice of returning to home countries or other countries). To what extent is the destination country preference of genuine refugees or asylum seekers the critical variable in the economic success?

• How has the political narrative on migrants become so devoid of the facts? Has this always been the case?

If anything, Hein De Haas’ How Migration Really Works really highlighted how much I still have to learn.

Subscribe now

Boom gone bust

In July, global oil giant BP quietly backed out of its majority share in a $55 billion green energy hub in Western Australia's Pilbara region.

The original idea was to build enough solar and wind to produce, in addition to abundant electrons, 10 million tonnes of the green ammonia each year1. For the unacquainted, that’s an absolute truckload and a half.

The clean2 fuel would then be shipped overseas via a new $550 million cargo facility being built at Lumsden Point (halfway between Broome and Exmouth) by the WA state government, crucially, with with federal funding.

Alas the project is now all but in the scrap heap alongside many other hydrogen pipe dreams.

Just 6-months earlier, in February, the Crisafulli LNP government in Queensland declined to commit a further $1 billion support of the Central Queensland Hydrogen Project, a similiar ambitious scheme, despite early funding from the Australian Renewable Energy Agency (ARENA). That project too now languishes on the precipice of cancellation.

Australia’s federal energy Minister, Chris Bowen, remains (somewhat) outwardly optimistic for hydrogen’s future - but if you read between the lines he sees it more as an industry policy or job creation scheme then a real part of the net zero solution.

"The Albanese government is firmly committed to seeing Gladstone's economy grow and creating new jobs for the region.

"Green hydrogen plays to Australia's unique strengths and we're unapologetic about pursuing an industry that is recognised as having an important role in the future of manufacturing and energy in Australia, and globally.

"Government support in developing hydrogen opportunities around the country provides additional certainty for projects, however how they progress ultimately remains a commercial decision for the parties involved."

The passive ‘that is recognised’ is the telltale sign here. A hedged bet that’s easily walked back.

So, have we reached the dog days of hydrogen evangelism?

The science is the issue, not the story

The physics of hydrogen production has never kept up with the policy narrative.

The National Hydrogen Strategy, originally written in 2019 by the Morrison Government, proposed hydrogen investments in residential, commercial, industrial, and transport infrastructure - many of which we now know will not come to fruition.

This analysis by the Grattan Institute explains why. Electrolysis to produce hydrogen (i.e. by splitting water into its hydrogen and oxygen molecules) is incredibly energy intensive.

Australia currently does not have sufficiently cheap electricity to produce hydrogen at a cost point to useful for most industrial purposes.

In the vast majority of residential and commercial instances, it is simply easier and cheaper to electrify appliances, rather than converting gas lines and vehicles to run off hydrogen blends.

Some use cases are clearly non-starters

Using hydrogen to power residential homes is a folly. Switching homes to hydrogen requires upgrading major parts of the current gas network, and changing all gas appliances to make them hydrogen-compatible.

Current gas appliances can tolerate only a maximum of ~13 per cent of hydrogen blended with natural gas. Domestic appliances that can use 100 per cent hydrogen (which exist only as prototypes at present) are likely to cost 20-to-30 per cent more than gas appliances do now.

The other fools errand is using hydrogen to power passenger vehicles. While hydrogen powered fuel cells are viable for electric cars (with a small but consistent market share), the technology is competing against traditional EVs that use lithium iron batteries - and their established and growing charging network.

Even excluding the efficiency argument (roughly 90 per cent for EVs vs 35 per cent for hydrogen), the capital costs of building hydrogen refuelling stations is enough to sink the technology. Despite valiant efforts by the likes of Toyota, it’s only a matter of time until EVs mop up their market share.

Even the best edge cases rely on heroic assumptions

Be it blatant industry policy, or a critical national security initiative, making green steel in Australia continues to be a tricky beast. Yes, primary3 steelmaking is emissions-intensive, and yes, hydrogen can replace coking coal as a reducing agent (producing water vapour rather than CO2 as a by product).

But in practice, the economics of this market are brutal.

A tonne of steel made with hydrogen rather than coking coal requires nearly 50 kilograms of the stuff - produced, ideally, by renewables that aren’t needed elsewhere in the grid.

Therein lies the rub. Every kilowatt-hour of electricity diverted to electrolyse water for hydrogen is one not used to directly decarbonise the grid. In practice this ‘second step’ (sunlight to electrons, electrons to hydrogen) is proving a step too far.

This also doesn’t consider the need for enormous battery storage capacity and very carefully designed PPAs to meet minimum supply requirements.

Saul Griffith has been banging this drum for a long while. When discussing hydrogen for steelmaking with David Roberts he explained:

It's just a very expensive way to do everything. We will do some hydrogen, and you need it for ammonia and you might need it for steel making, but only the redox component, not for the energy component.

Griffith goes on to explain it’s only the most rusted on evangelists that are still spruiking using hydrogen as a heat source for green steel:

The energy cost of coal into steel making is about a half a cent a kilowatt hour equivalent in heat. The energy cost of natural gas in steel making is about one cent per kilowatt hour in heat. The Australian stretch goal for hydrogen is $2 a kilogram, which is copied from America's $1 a kilogram, that's six or seven cents a kilowatt hour at the electrolyzer that's before you've transported or shipped it, it'll be 12 cents in any reasonable optimistic estimation.

So you're not going to make steel with a lot of hydrogen because you're going to make steel cost five or six times as much.

Modern electric arc furnaces (EAFs) skip the middle man - producing sufficient heat straight from electrons in the grid.

Export ambitions are really thought-bubbles

Nowhere is the big-H wishful thinking more acute than in Australia exporting hydrogen to the world.

It’s a convenient narrative. Japan has very little land for renewables, and is eager to import green energy. So, can Australia seize the opportunity?

Unlikely. There’s enormous obstacles that fall into two main buckets:

First, hydrogen export is thermodynamically inefficient. Energy is lost at every stage of the hydrogen lifecycle:

• Splitting water into hydrogen and oxygen through electrolysis consumes vast quantities of (ideally renewable) electricity, with losses of 25–30 per cent at least.

• Liquefying that hydrogen for transport - requiring chilling it to a frigid -253°C - costs another 10–20 per cent of the original energy.

• Re-gasifying it on arrival in Japan and converting it back into usable power for the grid incurs further losses of 15-30 per cent.

• By the time it reaches an end user’s power socket, only about a third of the original electrons survive the journey.

Second, hydrogen is expensive to move - especially all the way from Australia:

• It seeps through metal pipes, corrodes and embrittles storage tanks, and requires specialised cryogenic ships that do not yet exist at scale.

• Yes producers can convert it into ammonia (a far more stable compound) which can be shipped more easily. But then it must be cracked back into hydrogen at its destination. That process, too, is costly and energy-intensive.

• Australia isn’t exactly know for it’s abundant deep water ports with loads of free capacity (either on the port land itself for new infrastructure or spare berths for ships). Building (or expanding) new ports is a timely and costly exercise, even without clunky environmental approvals.

• Through this whole journey, the unit economics are fighting against LNG use cases, which require no new custom built kit.

Niche solutions met with vague policy

So which hydrogen use cases are left?

There’s 2, each full of caveats.

Firstly, producing green steel in a renewable powered electric blast furnace with hydrogen as the redox agent has solid science and a proven industrial track record.

The problem is the economics. The hydrogen has to come from somewhere, and diverting electricity from the grid to split water molecules isn’t all that profitable (especially at prices north of $2).

Secondly, some ammonia production (from hydrogen) is essential for the fertiliser trade, and the shipping infrastructure to transport it already exists - helped by the fact it liquefies at −33°C vs hydrogen’s −253°C).

The good news stops here. The physics of using this technology as a vehicle for energy export by converting the tonnes of ammonia back to hydrogen is unworkable. If Japan (or another importer) wants to must ‘crack’ the NH3 back into H₂, the net result is around 60–75% total energy loss from electricity-to-end-use.

In both cases, hydrogen plays a supporting role, not a starring one.

Despite ad-hoc capital contributions to pet projects, the Commonwealth government has largely hoped private sector business cases will prove up (any) viable pathways for a hydrogen industry.

The Commonwealth’s Hydrogen Production Tax Incentive (HPTI) provides a AUD 2 subsidy per kilogram of renewable hydrogen produced for up to ten years, between 2027 and 2040 for projects that reach final investment decisions by 2030.

Based on a similar scheme to the Inflation Reduction Act in the United States, it aims to sure up projects that have high capital costs today and long-tailed revenues.

Is this good policy? It’s tough to tell, but it likely won’t make a huge difference. The current lowest cost to produce green hydrogen in Australia for domestic use cases (e.g. redox agent or ammonia input) is between $4–6/kg. Best case scenario the HPTI equates to a 50 per cent subsidy, roughly putting the hydrogen use case on-par with LNG.

As more hydrogen projects go up in smoke, capital markets are reaching a quieter consensus: it is indeed easier to just electrify everything.

Heat homes with heat pumps, power vehicles with batteries, make steel with electric furnaces. Reserve hydrogen for what cannot be done more efficiently with electrons.

Let’s hope our trading partners soon catch on.

Australia, for all its sunshine and ambition, cannot rewrite the laws of thermodynamics. Some hydrogen use cases will persist - but these are well and truly the dog days of hydrogen evangelism.

Subscribe now

Subscribe now

I’ll set the scene.

My first London Climate Action Week. An eclectic mix of investors, bankers, academics, and scientists in the room.

The topic of discussion was on how to ‘widen the funnel’ for institutional investors to buy carbon credits1.

There were the usual suggestions. Better platform infrastructure for over-the-counter transactions, settlements through existing bank/FI relationships, and procuring long-term off take arrangements to ensure supply certainty.

Then one comment from a leading UK retail banker struck me:

‘We recommend that clients buy offsets up to their requirements - plus a margin of safety - just in case the quality isn’t quite there’.

Wait… what?

Here’s the logic behind that argument

Carbon credit quality is hard to notoriously hard to measure.

Despite best efforts, the inherent risk that carbon credits (to be used as offsets) have some fundamental quality issue mean the buyer should over-purchase their requirement.

In practice this could look like buying 110 per cent of a company’s annual offset requirements ‘just in case’ the credit stock winds up being (at least partially) junk.

We can see similarities to the venture capital investing model. Place many small bets and assume some won’t pay off, but risk will be mitigated by the ones that do.

Best case scenario you end up with a surplus of credits to hold over and use for the following year. Worst case scenario you have to write off some of your credit portfolio as junk, but any claims (e.g. offsetting) you’ve made against the use of those credits is protected.

Why I don’t buy it

Firstly, this attitude is blatantly defeatist.

Admitting carbon credit quality can be difficult to measure is of course true. Yet the extension to ‘we better accept some are junk and buy more just in case’ is sloppy logic. Here’s why:

• It undermines the crediting standards which already allow for ‘buffers’ (especially concerning additionality) in their own quantitative methodology modelling.

• It completely abandons the concept of procurement due-diligence and treats all carbon credits as roughly the same asset class, with the same risk profile, regardless of method. While all credits use the same quantity standard (i.e. per tCO2-e), there are vastly different methodologies to get there. The current price differential between credits (e.g. anywhere from USD $5 to $300) is a good indication of this.

• It incentivises opaque carbon credit production methodologies, as a risk buffer will be applied to the purchase regardless of how stringent the use and audit of MRV technologies.

A better way forward

Researchers at Oxford have put together the Oxford Offsetting Principles that step through a number of approaches for procuring, then using, carbon credits.

Principle 3 (page 18) aims to shift to removals with durable storage (low risk of reversal) to compensate residual emissions by the net zero target date.

It covers the risks of relying on biological storage methods (including plantings, soil carbon, mangroves) well. The authors maintain:

To reduce the risk of reversal of nature-based storage, these principles must be followed and projects with a high risk of reversal (e.g., due to bio-physical (including climatic) or political risks) should be avoided or approached with an appropriate risk reduction strategy, e.g. assurance to replace the storage in these projects should carbon be re-released.

This makes more sense on first principles. Buy 100% of the credits you need for 100% of the emissions you aim to offset.

If acquiring carbon credits that use biological methodologies, run sufficient due diligence that you can commit to their integrity. For the avoidance of doubt, also have a backstop that any assurance issues will be addressed (and if needed credits replaced) in the future.

How long is ‘the future’? Quite a while.

Principle 3 of the Oxford guidelines addresses it directly (on page 20):

…given the millennial lifespan of fossil carbon in the atmosphere, schemes or standards that only require monitoring and management of potential reversal on decadal timescales may undermine efforts to achieve and maintain net zero. For this reason, reversals must be monitored and addressed over timescales meaningful for net zero.

As I’ve written previously on the time value of carbon, the integrity of any carbon credit used for offsetting relies on a deep understanding on biological and physical sciences.

Planting methods are good for a few decades, soils for longer still. If the credits can’t stand up to scrutiny, the answer isn’t to buy more of them—it’s to buy better.

Subscribe now

For those interested, you can read more on time value of carbon here:

Subscribe now

We often hear that to fight climate change we need to decarbonise ‘everything, everywhere, all at once’. UN Secretary-General António Guterres’ even included the phrase in his opening salvo for an IPCC synthesis report.

This is a motivating war cry. But it’s not that helpful for businesses trying to reduce their emissions year on year.

While it’s true that innovative and novel solutions are needed in every sector of our economy, individual businesses need a clear strategy and action plan that they can work towards.

Let’s work through the steps in practice.

Step 1: Baselining and quantifying emissions sources

The first step for any business on their decarbonisation journey is to figure out the quantity of CO2e the business currently produces. I include the ‘e’ for equivalents, as impacts of greenhouses like methane (especially in the agriculture and waste sectors) are sometimes even more pressing than the CO2 component.

The more granular the analysis, the better. Splitting sources into Scope 1-3 is essential, and there’s a myriad of tools for almost all business types to give a good approximation based on expense data.

Estimates on a whole of division or BU level are good, but estimates based on the specific ‘activity’ (e.g. vehicles, flights, waste, energy) are better.

It's also useful for businesses with operations in different jurisdictions to have a site specific emissions estimate to track and measure.

Step 2: Designing emissions reduction activities

This is the hard bit.

Once a business knows its emissions sources, the next step need to figure out the decarbonisation activities that could reduce these emissions to zero.

The simplest way to reduce emissions is reducing (or eliminating) the activity that generates CO2. This could look like driving or flying less, or by running equipment (e.g. HVAC) for fewer hours in a day.

Some activities you can’t simply reduce. Instead, substitutes or new processes are needed. For example, a cafe might change suppliers to receive coffee beans with less packaging, reducing their waste emissions.

The tricky bit of designing emissions reduction activities is costing them. Often a ‘marginal tonne’ approach is useful here. The cost of reducing emissions from large infrastructure upgrades goes down the scale of the capital works increases.

The last resort for hard (read: expensive) to abate emissions is buying offsets1 - this is the ‘net’ part in net zero. This can rightly be part of a businesses transition strategy - however the cost becomes ‘baked’ each year as new credits will have to be purchased.

For any decarbonisation strategy, Step 2 will absorb the lions share of time and effort.

Step 3: Sequencing and financing emissions reduction

Once you know the approximate cost of an emissions reduction initiative, and the quantity of emissions that could be reduced - it’s time to plot them from least to most expensive.

Since the 1970’s economists tend to call these bar charts Marginal Abatement Cost Curves (MACCs).

Some businesses will have decarbonisation initiatives that save them money right now (the bars below the axis). These are the ‘no-brainers’.

For example, a food processing factory might upgrade to leasing more energy efficient fridges - reducing both their running costs and emissions.

Most businesses should then decarbonise based on their quick-wins - implementing the ideas that are cheap and fast.

It’s also important to invest in R&D for some of the more expensive (or larger) emissions categories. Better technology and operational efficiency brings down overall decarbonisation costs - reducing the need to rely on offsets in the long-term.

Step 4: Getting the most of your MACC

The MACC approach is the best first order response for most businesses, although it’s not without its critics.

Issue #1: MACCs are static

MACCs show decarbonisation costs at a specific point in time. They can’t be ‘set and forget’, as they ignore technological change and learning curves (e.g. input cost reductions or evolving regulatory landscapes). These effects can be both profound and rapid (e.g., solar prices dropping dramatically over time).

Regular input factor updates into MACC calculations (at least annually - ideally quarterly) are best to ensure the state of play is accurately represented.

Some suggest a forecasting exercise to develop a ‘MACC thesis’ on where costs are likely to increase and decrease over time. This is strategy is interesting but high risk - many have been burnt before by delaying action in the hope it’ll be easier tomorrow.

Issue #2: Lack of system dynamics

MACCs treat each emissions measure (i.e. each bar on the chart) as independent. This ignores interactions between initiatives (e.g. how increased electric vehicle adoption affects the grid and renewables).

Individual initiatives on the MACC can be a bit chicken and egg, with capital investment having to happen simultaneous for some projects,

The MACC also fails to account for path dependencies (i.e. how early choices shape future options). For example, if you’ve just invested in a brand new energy efficient office kit out that took years, you might be reluctant to move to new premises as the business grows.

Issue #3: Practical and equity blindness

Global businesses with several site locations or complex internal supply chains often run into curious predicaments if they produce a company-wide MACC. It’s often cheaper to reduce emissions in less developed countries2 than it is in the OECD.

Following the MACC approach, this makes perfect sense. Yet on a practical basis, it is unviable for a large multinational (think Coca Cola or Nestle) to invest in emissions reduction in SE Asia and South America while doing nothing in the U.S.

Regulators wouldn’t allow it - and customers would be outraged. The simplest solution here is to have country specific MACC tailored to addressing the local emissions targets.

The upshot of it all

Despite their limitations, MACCs are still one of the best tools we’ve got for decarbonisation prioritisation.

Grand slogans may stir hearts, but it is marginal abatement cost curves - plotted, updated, interrogated - that steer real progress.

For firms, the future will belong not to those who promise ‘everything, everywhere, all at once’, but to those who deliver the right thing, in the right place, at the right time.

That’s a less catchy but far more impactful rally cry.

I’m make that trade-off any day of the week.

Subscribe now

I’ve been thinking about Ronald Coase a lot this year. This might seem odd for an economist that died 12 years ago, and did his most seminal work in the 1930’s.

The thing is, if Coase were alive today I think he’d set the record straight on just how AI is going to rearrange our microeconomic furniture.

Let me explain.

Why Coase is on my mind

It all started when Dwarkish Patel published an What fully automated firms will look like. There was one paragraph which I couldn’t let go. Patel says:

Ronald Coase’s theory of the firm tells us that companies exist to reduce transaction costs (so that you don’t have to go rehire all your employees and rent a new office every morning on the free market). His theory states that the lower the intra-firm transaction costs, the larger the firms will grow. Five hundred years ago, it was practically impossible to coordinate knowledge work across thousands of people and dozens of offices. So you didn’t get very big firms. Now you can spin up an arbitrarily large Slack channel or HR database, so firms can get much bigger.

AI firms will lower transaction costs so much relative to human firms. It’s hard to beat shooting lossless latent representations to an exact copy of you for communication efficiency! So firms probably will become much larger than they are now.

This narrative hangs fine and seems logical enough. But hang on. In last sentence… what does much larger mean?

By headcount? Surely not. The tech layoff are here and don’t appear to be reversing. By revenue? If anything we’re seeing a disaggregation of the economy and less ‘winner take all’ firms.

So what do we know for certain on optimal firm size? Surprisingly little.

What Coase actually said on optimal firm size

For the record, Coase published The Nature of the Firm, an article (or more an essay really) in the newly founded economics journal Economica in 1937. He was 26 at the time.

The article offered an economic explanation of why individuals choose to form partnerships, companies, and other business entities rather than trading bilaterally through contracts on a market.

The core takeaway is that a firm will expand1 as long as internal transaction costs (i.e. organizing within the firm) are lower than the external transaction costs (using the market).

A firm will stop growing (i.e., reach its optimal size) when the marginal cost of organizing an extra transaction internally equals the cost of carrying out the same transaction on the market.

Therefore, by extension we expect that the optimal firm size is reached when the cost of organizing an additional transaction within the firm equals the cost of conducting that transaction through the market.

This makes sense in practice. If it's cheaper for a car manufacturer to outsource tire production (because managing a tire division would be more costly internally), it will do so. But if coordination, quality, or supply risks are too high, it may choose to produce tires itself.

Enter agents on both sides of the equation

This theory tracks well from the 1930’s to today, until we accept AI agents change the marginal cost of organising an additional transaction both within firms and for single-shingles in the external labour market.

So to answer the question ‘will AI expand or reduce optimal firm size’ we need to have a thesis on the question ‘will AI lower internal or external costs faster?’. Here it gets juicy.

Strong arguments that AI agents will lower internal costs faster:

• Automation of standard internal business processes (e.g., HR, accounting, logistics) reduces the costs of managing complex, multi-person teams inside the firm, encouraging bigger teams.

• Smart coordination tools (e.g., AI schedulers, predictive analytics, resourcing tools) make large-scale produce/service development more efficient, meaning more can be done in-house rather than through subcontractors.

While these points are reasonable, they also apply to the wider market too.

Strong arguments that AI agents will lower external costs faster:

• AI-driven platforms and agents (e.g., for hiring freelancers, managing supply chains) reduce the friction (and costs) of outsourcing. Smart coordination tools across firms (e.g. common tech infrastructure) is lowering boundaries for intra-firm collaboration (while maintaining security).

• As LLM’s mop up a lot of ‘grunt’ analytical/research tasks, it’s now easier to have a smaller number of highly specialised networked firms, rather than one large (and expensive) generalist firm.

Where that leaves us

I’m more bullish on the second argument.

I’ll happily accept the costs of organising within a firm are falling. It’s just that the cost of acquiring bespoke, on-demand labour from the wider market is falling more quickly. Applying Coasian arguments, average firm size should fall too.2

I haven’t seen many good papers on this thesis. I imagine it’s hard to disentangle the AI agent effect ripping through firms today from the highly volatile cost of capital firms have grappled with since 2020. We know interest rates are one of the primary drivers of firm aggregation or disaggregation within a macroeconomic cycle. When capital is expensive, firms:

• focus on core operations and divest or outsource non-core activities.

• borrow less for M&A

Both these factors may be propping up my Coasian thesis.

However I’d maintain that even in a low(er) interest rate environment, firms are incentivised to chase revenue and market share - not headcount.

Would Coase agree? I hope so. In his own words:

“Faced with a choice between a theory which predicts well but gives us little insight into how the system works and one which gives us this insight but predicts badly, I would choose the latter”

The result works - but you do really have to care about the ‘infrastructure’ layer of government to get through it.

Luckily for Freedman… I do.

The primary narrative of the book is that consecutive decisions throughout government have compounded over the past four decades, and the current paralysis is a patchwork of unresolved issues.

The book opens:

What I wanted to get across was that Johnson and Truss, and other recent self-inflicted wounds, were not the cause of our problems but merely symptoms. Over the past forty years a series of trends - centralisation, loss of effective scrutiny, and a superfast media cycle - have weakened our institutions to the point where any government, however full of clever or well-intentioned people, would struggle.

This is an important lead in. The Tories cop a pretty harsh critique by Freedman for most of their decisions throughout 2000’s. I kept coming back to thinking how much of this was ‘baked’ and how much was a series of poor choices. There’s a bit of column A for sure, but as we’ll find out, a hefty portion of column B.

The quirks of Westminster

It was news to me that the professionalisation of politics has been such a recent phenomenon in the UK. Freedman writes:

Yet, in reality, most MPs work far harder than they ever have before. Up until the 1990s, it was, for the majority, a part-time role, almost a hobby for some. Since then, it has transformed into a profession of its own, with extremely long and intense hours.

The days of long lunches and smokey cigar rooms with the odd nip into the House of Commons seem almost unimaginable. Trying to practice law in the mornings then fulfil the duties of a MP in the afternoon is almost comical to consider in 2025 - at least for any MP with Ministerial ambition.

Freedman notes this is probably a good thing:

Constituents get a much more engaged and active service, on average, than in previous generations. MPs work harder, drink less (though still too much) and have more of a sense of vocation.

Feeding The Grid

The infiltration of media management to the political process is ubiquitous across Western governments. However, Freedman explains in the UK there’s a formalised structure to this ‘random announcement generator’.

The Grid was first introduced by Alastair Campbell in 1997, as a day-to-day plan for coordinating government communications across different departments. But it quickly evolved into a way of controlling policy from the centre. A process designed as a way to sequence announcements to provide a steady diet of news quickly became a permanent hungry beast, demanding more and more policy from ministers and their advisors.

There’s a big part of this that becomes circular:

Increasingly the tactics have remained but the strategy has gone. It is a self-reinforcing problem. The more effort governments put into generating tactical announcements, and finding on-message ministers, the less scope there is for policy-making and long-term thinking. This panics central government into wanting more tactical announcements.

Freedman provides a personal anecdote that this policy on the run approach often comes with less than 12-hours notice:

These ‘night before’ phone calls from the Treasury are dreaded across Whitehall.

This chapter was probably my favourite of the book - for the sheer absurdity that everyone recognises, but fails to attempt to change.

What should we expect from local government?

When I first moved to the UK in 2025, I struggled to get my head around the role of local government. More specifically, ‘what on earth does Council do with all this tax they collect!?’. Turns out, a fair bit.

Local government has changed dramatically since Thatcher. Freedman writes:

The shift towards a much looser approach to managing the economy and the substantial reduction in state-owned assets created the space for central government to focus more on the delivery of public services. And here Thatcher, and most of her colleagues, felt local government was doing a bad job at high cost to the taxpayer.

This view was coloured by the increasing politicisation of local government. In 1965, only 50 per cent of councils were controlled by a party, or a coalition of parties, and independent councillors were commonplace. By the mid-1980s it was 84 per cent. In urban areas, those councils tenses to be controlled by Labour, and many cities, including London, were controlled by the Left of Labour. These authorities were explicitly included on her list of the ‘enemy within’ in a speech made to party colleagues in 1984.

The erosion of local government (often from a political agenda) is a recurring theme.

Incentives have become misaligned over time.

Running a large economy from Downing St is a fools errand that Whitehall is yet to acknowledge. Unfortunately I don’t see a revamp of the responsibility (or fiscal powers) of local government anytime soon.

Productivity (…ahh that little thing)

Calling the UK economy ‘unproductive’ may be true in a relative sense, but it fundamentally misses the spatial problem. Freedman explains:

It is well known that the UK’s productivity is lower than other large developed countries and has been for some time - it’s a common concern aired by politicians and commentators as a leading cause of our woes. It is less well known that the south-east of England is more productive than most other countries but the rest of Britain is far less so. In 2015 a worker in the south-east was 7 per cent more productive than the average German, but elsewhere they were 22 per cent less productive.

The how and why of this is complex… but Freedman assigns (partial) blame in a novel direction.

The absence of any regional or sub regional tier of government with the ability or incentives to develop local economies is a major cause of Britain’s economic weakness compared to our competitors.

Freedman continues:

But it all links back to centralisation. With local government given so little room for manoeuvre, highly unstable funding and little incentive too grow, due to their inability to create local taxes, they can’t do much to grow their economy. Nor can they improve their public services, which can draw investment and people to their area.

Blaming Thatcher era policies for today’s productivity woes seems a stretch too far. Although… there is certainly some evidence in that direction.

Outsourcing and privitisation continues (at a cost)

When punters refer to a ‘broken UK’, often their first namedrop is a privatised state service. Think hospitals, prisons, and immigration to name a few.

To be fair to the economics behind privatisation theory, Freedman develops principles for when the model can work:

The core principle of contracting out is that competition will lead to innovation and lower cost than doing it yourself. This is true enough when certain conditions are met. Firstly, there needs to be many organisations, public or private, who can provide the service to generate genuine competition on both price and quality. Secondly, it needs to be relatively easy to measure where the service you want has been delivered to an acceptable standard. Thirdly, government needs to be able to hand over most of the risk of failure - if the taxpayer is still on the hook for picking up the costs if things to wrong then the premium paid to private firms is not justified.

The trouble is, the exact things governments want to outsource (which are normally both difficult politically sensitive) don’t fit these rules. Freedman gives an example:

Take a specialist activity like running immigration processing. There is no existing market. That means there are few plausible bidders outside of a handful of huge multinational outsourcing firms whole sole purpose is to win new contracts, almost regardless of what they are for.

The above issues are compounded by the private equity model:

By 2019 three of the four largest providers - representing by themselves 10 per cent of all care home beds nationally - were private-equity backed.

And unsurprisingly, this model has a certain rhyme and rhythm:

The standard model for these private equity firms is to load up companies with debt, slash costs, and then sell on to another buyer.

This is one area where the UK needs urgent, long term, and ideally bi-partisan policy reform. Many of these services either support or protect the nations must vulnerable.

Feeding these contracts into a profit extraction machine is not only bad procurement policy, but alarming social policy too.

Subscribe now

There's a new trend to report - often with a healthy dose of alarm - that we're in a 'per capita recession'.

In Australia, the ABC among others has been banging the drum (see here and here) arguing that living standards for the average citizen have gone backwards since last quarter.

The ABC writes:

On one key measure for households, Australia is already in a deep recession comparable to the early 1990s. That measure is GDP per capita, or how much economic activity there is per person.

But hang on… is that really a ‘key measure’?

Technically, the measure is no good

It's true economists have long been fond of dividing things: income per household, productivity per worker, and of course GDP per capita. The last of these, though, can be a statistical sleight of hand.

GDP per capita is a great proxy for wealth and living standards when comparing countries over time. However, it's a terrible measure of the change in output of an economy in any given quarter.

When a country's population surges, normally due to migration, GDP per capita may fall—even as total GDP rises and economic output grows. In the vast majority of developed economies, population growth in any given quarter is a policy choice.

We saw this in the easing of post pandemic travel. In 2023, Australia's net overseas migration surged to record highs—over 500,000 people in a single year—as the government loosened post-pandemic visa rules to address labour shortages and welcome back international students.

While Australia's total GDP continued to grow between at a rate of 0.2-0.5 percent per quarter, Australia's population - driven by migration - grew faster. As a result GDP per capita fell for two consecutive quarters, and journalists rushed to publish 'per capita recession' articles across all the major mastheads.

It's a bit more complicated than that

In most of the 'per capita recession' journalism, there's an undercurrent of 'although GDP is growing, there's not enough to go around'. This angle gets clicks, but it fundamentally misunderstands how economic development occurs.

Changes in population do not immediately translate into commensurate changes in output - and certainly not in the same quarter. There's a mountain of published literature saying newcomers do not instantly produce goods and services; they need time to find jobs, acquire skills, and integrate into the labour force.

There's ways to speed this up - namely tightly controlled (and often skill based) migration programs, as well as extensive settlement and welfare services when migrants arrive. However even for the best and brightest migrants, their impact will take longer than three months to hit the national economic accounts.

It's true our economic fabric feels pressure—on housing, transport, and services—when population grows quickly. But calling it a "recession" muddies the water. These headlines confuse short-term arithmetic with the long-term economic outlook.

We must be careful not to mistake a shifting denominator for a shrinking economy.

Subscribe now

“What’s the point in us even trying to reduce emissions?”

The argument is as familiar as it is seductive. Why should we impose costs on ourselves, regulate our industries and potentially hobble our economies, when China’s emissions dwarf everyone else’s?

Since the early 1970’s, China’s emissions have escalated quickly. Bypassing the US in the mid-2000’s, China’s emissions of around 12 billion tonnes annually are now a clear outlier.

By contrast, largely thanks to Paris agreement commitments, U.S. emissions have generally trended downward since 2007 albeit with significant fluctuations (financial crisis, COVID, etc.).

The most recent data suggests US emissions are 5.6 billion tons of CO₂. This roughly 24% below 2005 baseline levels under the Paris Agreement targets.

So while other countries are flatlining or reducing emissions, China is racing away producing more CO2 than ever… right?

The right tool for the job

China's emissions aren't all that high on a per capita level. On average, a Chinese citizen emits roughly 8.5 tonnes of CO₂ annually. That is on par with Europeans and well below Americans, Canadians, and Australians who emit more than 14 tonnes per person.

The emergence of the Chinese middle class in the past 30 years is nothing short of an economic miracle. Hundreds of millions have been lifted out of poverty. World class health services and education became almost ubiquitous in a single generation.

This development has allowed China to become a research powerhouse, exporting IP (through licensing or market adoption) globally.

Further, China is taking significant climate action right now, even though you can't see it in the chart:

• It is the world's largest investor in renewables.

• It produces the vast majority of the world's solar panels and dominates global battery manufacturing.

• It is home to the largest and fastest-growing electric vehicle market on earth.

• It has committed to peak emissions before 2030 and to reach carbon neutrality by 2060.

In fact, Chinese solar energy capacity took off during the pandemic and hasn’t stopped. In 2023 alone China installed 200GW of solar. That’s more than the total solar capacity of the United States at the time.

But… coal?

The naysayers at this point usually point out that China has been on a coal rampage for the past 20 years. This is true, and consistent with the widely documented global narrative for economic development.

For most developing economies, coal begins as the energy source of choice. It is cheap, widely available, and well-suited to the demands of baseload power generation. These qualities make it particularly attractive to governments focused on rapid industrialisation and energy security.

Only as countries grow richer do they typically acquire the fiscal space, institutional capacity, and technological expertise required to scale up (and firm-up) cleaner alternatives such as solar and wind.

These low-carbon sources, while increasingly competitive, still demand substantial upfront investment and in a functional and reliable energy grid - luxuries often beyond the reach of lower-income nations.

The narrative of ‘burn coal first, then transition to cleaner sources later’ becomes even more obvious when we add in some time series comparisons with developed economies.

Notice how in the chart below China and India have similar shaped curves?

So it’s true a huge component of global emissions come from coal consumption in China. Again, this is an expected part of the development narrative.

And yet, things are changing quickly.

The latest estimates for 2024 have China adding another 280GW of solar to the grid (roughly 30% year on year growth). There was another 80GW of wind power added on top of this.

And for coal? China added only around 40 GW of new coal power capacity in 2024. In fact, approvals for new coal projects dropped significantly: just 10.3 GW approved in H1 2024 (a nearly 80% decrease from H1 2023).

So, where does that leave us?

Here's a better way to think about the problem

The worst (but probably most common rebuttal) to the question of “what’s the point in us even trying to reduce emissions compared to China?” is that “well every bit counts”.

While that answer is technically true, it misses the point that the question itself is wrong. China emits around 30 per cent of global emissions. For this they provide energy for about 18 per cent of the global population.

On a per capita level, this is a great deal compared to almost all developed economies. What’s more, the unprecedented renewables installation rate in China is making the deal even better year on year - for China, and everyone else.

Using China's currently high* emissions as a reason for inaction in the UK or Australia:

• Is fundamentally ignorant of the standard economic development narrative

• Ignores basic moral and geopolitical leadership in driving global momentum through collective agreements

• Misses the economic opportunity for advanced economies to install increasingly cheap renewable power into fully functional energy grids

So instead of asking "what's the point of changing anything here at home?", perhaps the better question is: “how on earth are we going to keep up?”.

Subscribe now

In March 2023 Ezra Klein introduced the concept of the 'everything bagel', where policies aim to fix just about everything adjacent to their core problem, as well as the problem itself.

Seen by an uninformed audience as 'consideration' or even worse 'consultation', this approach is actually highly counterproductive to getting things done.

The idea expands to explain the solutions of one era can become the problems of the next. A common example is that the same rules that were designed to help the environment in the 1960s now harm progress in the 2020s.

‘Focus’ policymaking at least sounds nice

Joe Hill’s piece Everythingism: An Essay, published in March 2025 builds on Klein’s idea, outlining the merits for a far simplified version of policy making - especially in the UK.

The piece doesn’t pull it’s punches, running with a subtitle that describes everythingism as the pathology holding back the state. There’s even a snazzy diagram to explain what ‘focus’ policymaking could look like.

Hill’s essay draws out some of the absurdity currently observed in the UK:

Everythingism is why the Government can't have a policy that delivers safe and cheap nuclear power quickly. Britain has some of the highest energy prices in the world, and why increasing geopolitical instability energy security is a key government priority. But because of Everythingism moving to shoehorn geopolitical security, 'green' goals, and cheap nuclear power is impossible unless it also promotes the Welsh language, enhances biodiversity, and upskills the local workforce.

Everythingism is the belief that every proposal, project or policy is a means for promoting every national objective, all at the same time. Trade policy is not for producing cheaper goods, but for reshaping the global economy. Planning policy is not for getting high-quality housing near the best jobs, but creating local jobs, hiring Trade Union members, and enhancing biodiversity. And Government procurement isn't for the best products and services at the cheapest price. It's about promoting "social and economic value", reshaping the British economy to promote Net Zero, regional economic development, and diversity and equality training.

Because of Everythingism, we never do any one thing well, we do everything badly. Housing policy becomes the main route for fixing the nitrogen imbalances in local rivers, and creating more social housing the main way of subsidising the welfare state. Trains must look after bats. Climate policy is to support the services sector.

Hill follows up:

If every policy is about every other policy, it's even easier for policymakers to get distracted and forget to 'keep the main thing the main thing'. Too often, policy levers aren't pulled with equal enthusiasm, despite the evidence suggesting that in most areas, a small number of the causes generate the biggest effects. James Q Wilson identified this "goal ambiguity" as a central problem of bureaucratic organisations.

This rolls off the tongue well and seems to be correct, but as with anything, it’s useful to consider the edge cases.

Enter Ravi Gurumuthy

Ravi Gurumuthy hits back eloquently in his piece In defence of everythingism saying that this approach is flawed in some particularly critical areas - law and justice, energy, and immigration are top of this.

History is important here too. Ezra Klein isn’t the first left leaning intellectual to discover progress really does matter, as well as process. We’ve been in this cycle before.

Gurumuthy writes of the UK:

Despite the inevitable need to balance competing goals and constraints, governments have attempted in the past to push back on everythingism. In 1982, when Michael Heseltine was setting up Urban Development Corporations, he said he wanted them to be 'single-minded' and 'focused like a laser beam' on the problem they were charged with. Next Steps agencies that separated policy from delivery and the growth of quangos and arms length bodies were also part of this effort.

But agencies with a singular focus often create externalities, dumping costs onto other departments. They are usually defined by a particular function, when the reality of people's lives is that they interact with multiple professions and needs. The fragmentation of services created by 'new public management' led to the push for 'joined up government' in the 1990s.

This makes intuitive sense - peoples lives are complex and ever changing, and programmes design to support health, education, housing, and incomes will all affect each other. It is genuinely difficult to find ‘shortcuts’ where ‘focus policymaking’ actually works.

International development may be one of the true cases where simpler is better. Gurumuthy explains:

Some forms of joining up do not add services but subtract them, creating much greater simplicity. In international development and humanitarian interventions, there has been a growing focus over the last twenty years on conditional and unconditional cash transfers. Randomised trials have shown that cash transfers are not only more cost-effective than providing aid recipients with food, clothing or shelter, they also increase enrollment at school, children’s nutritional status, and reduce intimate partner violence. There is a good case for donors and aid organisations with single mandates - such as the World Food Programme - to pool their efforts to create effective cash safety nets and meet multiple outcomes together much more cost-effectively than they can do alone.

Getting everythingism right is practice

I think Ezra's Klein’s core idea that there is no real ‘everything bagel’ is true - yet it needs a slight tweak.

Progress requires regulation of the regulation itself. We must hold a dual focus on introducing policy where it is the only solution, while stripping away old rules and regulations that are inhibiting progress today.

This is hard in practice. It's unpopular to take away a regulation if it increases the risk of a single person getting harmed, whereas new policies only need to stack up in a cost-benefit analysis to get a business case fully funded. Up and coming Ministers also don’t win hearts and minds (let alone votes) selling dry reform initiatives.

Some say that sunset clauses are the answer to this problem - policy with an expiration date. This makes legislative review an active rather than a passive nice to have. Personally, I don’t see this as likely in any Westminster system anytime soon.

So what is to be done?

Firstly, we need a culture that values and prioritises progress. The tools for progress change over time. Policies of old aren’t the bedrock of intellectual knowledge and social welfare - they were means to end for the issues of the day.

The tools we need today are different, just like the tools we’ll need tomorrow. Scrapping old policies is progress, and it’s equally as important as introducing new legislation.

Secondly, ‘focus policymaking’ where one policy targets one outcome is a worthy goal - but it’s not gospel. The Theory of the Second Best is a whole field of welfare economics dedicated to explaining this. Policies have very human consequences, and not all eggs can be unscrambled.

We should avoid everythingism as a general approach to policy making. Instead, we need a new culture that values progress as just as it value humility One that knows when to stop. Or better yet, when to let go.

Subscribe now

Yet the rules themselves have changed with almost every chancellor, and their brittle design is beginning to look less like prudence and more like self-harm. Particularly grievous is the way today’s rules are strangling long-term investment in infrastructure, just when Britain’s creaking economy needs it most.

How did we get here?

he UK’s formal fiscal framework began under Chancellor Gordon Brown in 1997, following Labour’s election victory.

Two key fiscal rules were introduced:

1.  The Golden Rule: The government would only borrow to invest over the economic cycle, not to fund day-to-day spending. This was meant to ensure that current expenditure was covered by current revenues.

2. The Sustainable Investment Rule: Public sector net debt (PSND) should remain below 40% of GDP over the economic cycle.

These rules were monitored by the Treasury itself, with no independent oversight, and applied over a rolling cycle. Over the years, the goalposts were moved to hit a moving target.

For example in 2005, with public borrowing rising and concerns mounting that the government was about to breach its own Golden Rule, the Treasury under Chancellor Gordon Brown revised the official dates of the economic cycle. Instead of running from 1999 to 2005, as previously understood, the cycle was redefined to span 1997 to 2006.

This seemingly technical change had major political and fiscal implications. By extending the start of the cycle back to 1997, the Treasury could include early surpluses from Labour’s first years in office, which offset subsequent deficits. This allowed the government to claim it had met the Golden Rule—that borrowing was used only for investment across the cycle—without altering actual spending behaviour.

The financial crisis of 2008 sent borrowing through the roof (now the responsibility of Alistair Darling), as the government propped up banks and pumped stimulus into a collapsing economy. With little fanfare, Brown’s original fiscal scaffolding was dismantled.

By the time the Conservative-led coalition arrived in 2010, a new orthodoxy had taken hold: austerity. The casualty of this was public investment. Now Treasury orthodoxy drew no distinction between borrowing for a new road and borrowing to pay nurses. In its place came a singular focus on deficit reduction and debt control, not as a means to an end, but as ends in themselves.

Because the rules cap overall borrowing, capital spending often gets the chop when savings are needed. The Office for Budget Responsibility (OBR) is not fooled: it repeatedly warns that public investment is “lumpy” and prone to cancellation when fiscal targets bite.

What’s more, the official forecast period—five years—is used to judge compliance, but the deadline is always rolling, giving the illusion of discipline while allowing convenient wiggle room. This has encouraged chancellors to tinker with taxes and squeeze spending in order to hit targets in a forecast, even when reality diverges wildly.

Indeed, we’ve seen a recent case study in self-inflicted injury. In October 2023, the government scrapped the northern leg of HS2, Britain’s flagship high-speed rail project, citing affordability. Never mind that the cost would have been spread over decades or that private capital was queuing to co-invest. The fiscal rules—as interpreted—declared it unaffordable. So the bulldozers were sent packing. Other projects, from renewable energy infrastructure to flood defences, faced similar fates.

What is to be done?

A better system is not hard to imagine. One option is to revive a version of Brown’s golden rule, adopting a dual framework model: one rule for day-to-day spending, another for public investment, with the latter subject to different borrowing limits and longer time horizons. This would reflect economic reality—roads and railways last for decades—and align public accounting with how businesses and households think about investment.

Another is to create an independent body to assess the long-term value of investment projects, shielding them from fiscal haircuts. Countries like Sweden and the Netherlands have adopted such mechanisms with success, maintaining high levels of infrastructure investment without spooking bond markets.

What has emerged is less a set of durable rules than a ritual of annual reinvention. Each chancellor redraws the lines to suit the moment, offering the appearance of discipline while preserving political room to manoeuvre. The result is a framework that is both rigid and fragile—quick to constrain long-term investment, yet endlessly malleable when short-term politics demand it.

This is a false economy. Britain’s productivity is anaemic, its roads potholed, and its energy grid ageing. The country needs more infrastructure, not less. Borrowing to invest in such assets—especially when interest rates are lower than the economy’s growth rate—is not reckless. It is rational. But the current rules treat all debt as equal, regardless of its purpose or return.

Britain’s fiscal rules were meant to protect the economy from short-termism. Ironically, they have become a source of it. If the country wants to build for the future, it must first unshackle itself from its own accounting games.

Subscribe now

This was my starting question in January 2025 ahead of my London move. I didn’t have the narrative at all clear in my head, but had heard various whispers that ‘things had changed’. I was searching for the best, most recent book on the topic, and soon found Torsten Bell’s research.

Bell’s latest instalment Great Britain? (published conveniently as he was elected as a first time Labour MP) provides a succinct overview of the issues to date. What’s more, he dedicated a fair amount of time to the treatment rather than rehashing the diagnosis.

There’s one stanza that captures the essence of the conundrum the UK faces:

The time has passed when the biggest generational threat to our social contract was pensioners being left behind as the economy grew. Acknowledging this is essential; not so as to blame older generations for the plight of the young, but to emphasise where we are falling short. Today we are in fact in an era of twin generational challenges: our failure to improve economic outcomes for the young, and our struggle to maintain our welfare state for the old.

One of the core themes of the book is that the ‘idea’ of Great Britain is so different to folks of different generations, geographies, occupations, and cultural heritages. There’s no longer a unifying economic narrative, meaning there’s multiple ‘truths’ competing for budget priorities.

My diagnosis of the current woes boils down to 4 key ideas.

Idea 1: Fix the Treasury rules

If there’s one place I’d start to fix Britain’s economic woes, it’s the HM Treasury investment rules. Webb writes:

The rules limit borrowing (how far spending can exceed taxes raised) in a given year and require public debt (the stock of borrowing) to be falling in give years’ time. The problem is that both rules treat investment spending, such as building a new council house, as indistinguishable from day-to-day spending, such as paying teachers’ wages - ignoring the fact that the former usually creates an asset, improving the government’s books (the public sector balance sheet). No company would ever think in those terms, but that is how we oversee the nations finances.

For those still trying to get up the curve (myself included), the current approach to British fiscal rules started under Gordon Brown, who as Chancellor in 1997 introduced the "Golden Rule" and "Sustainable Investment Rule":

• Golden Rule: The government should only borrow to invest, not to fund day-to-day spending — over the economic cycle.

• Sustainable Investment Rule: Public debt should be kept at a "stable and prudent" level relative to GDP.

There was nothing wrong with this approach, until the rules were butchered in 2008.

After the 2008 financial crisis, the UK saw a huge rise in public debt due to bailouts and stimulus. In response, fiscal rules became more focused on absolute debt levels and annual deficits, not on the quality of spending.

The Coalition government led by David Cameron (2010–2015) shifted toward stricter rules on overall borrowing and debt reduction. The causality of this approach is investment. Because the rules cap overall borrowing, capital spending often gets the chop when savings are needed. This incentivises governments to delay/cancel major projects to meet the rules.

For a country with anaemic productivity, potholed roads, and an ageing energy grid, fixing the Treasury rules seems like an obvious place to start.

Webb provides some ideas for rule changes (e.g. tax revenues to cover day to day spending). There are likely many others equally as suitable.

Idea 2: It’s the investment rate stupid

The UK investment rate as a percentage of GDP over the past 20 years has been truly shocking. This is true for both public and private investment:

Reflecting on this table, Bell writes:

Over time, this scant investment risks turning corporate Britain into a museum. A look at the top five firms in the FTSE 100 today should be cause for concern. They are old firms, in old industries.

The reasons why are complex, especially for the private sector. Bell notes that:

British managers are too often bad managers

The idea stemming from:

In British firms, managers rarely are in charge, and they face a highly unusual lack of pressure to invest for the long term, from both above (the owners) and below (the workers). Today, owners of listed firms are too remote. Foreign ownership increased from just over 10 per cent in 1990 to over 55 per cent in 2020, and is too dispersed to meaningfully monitor executives. Among rich economies, the UK has the lowest proportion of firms with ‘blockholder’ shareholders - owners with a sizeable stake, and therefore enough power and skin in the game to bother ensuring their firm has a strategy for growth, not just for profits today. Without that pressure, chief executives, who have short tenures and face immediate public scrutiny when profits fall short, will focus on the here and now.

This narrative seems about right - but there’s no simple (or immediate) solutions at hand.

Idea 3: Focus on more than just London

Regional differences, most notably been north and south have characterised UK society for centuries. Bell shows this distinction is still alive and well in just about every economic metric you can find:

Between 2016 and 2021, London’s service exports were up 47 per cent, totalling 152.2 billion. Yet service exports in Greater Manchester and Birmingham grew by just 11 and 3 per cent respectively. Productivity in Greater Manchester remains 12 per cent below the UK average. Birmingham, its competitor to be England’s second city, is even further behind (14 per cent). This is not an economy with all its engines firing.

Having genuine, highly productive ‘second cities’ needs to be a fundamental priority of the UK’s hub and spoke model.

We see this urban-regional divide play out in infrastructure. Bell writes:

The highest-profile attempt this century to boost the economies of Birmingham and Manchester was a high-speed train line (HS2) connecting their city centres to each other and to the capital. But after years of mismanagement the Manchester leg of the project was scrapped in September 2023, with the line now set just to link Birmingham to a London suburb - a hugely expensive project that no one would have set out to build, and which even the government department responsible for it recognises ‘offers very poor value for money to the taxpayer’.

It will take generational efforts to truly bring Manchester and Birmingham ‘up the curve’ - and the time to start is now.

Idea 4: Get realistic about housing

The horse has well and truly bolted on the housing market - especially in Greater London.

Webb hits the nail on the head regarding affordable housing:

Once the yuppies arrive (or anyone else, for that matter) then housebuilding is part of the solution, not the problem. But we must not forget the rest of the answer: for many poorer households, the market will never deliver housing they can afford. The need to more social housing requires both the higher levels of public investment that we discussed in Chapter 5 and cheaper land acquisition for councils. Britain’s stock of affordable homes is large relative to most of Europe, but relative to the number of families it has fallen by around 40 per cent since the 1980s, as few were built and many sold off.

Many highly dense, global cities have accepted this fate and started working on proper solutions (New York, Hong Kong, Singapore). London needs to throw off the shackles of old and accept that housing isn’t a standard market - it’s a fundamental human right.

Subscribe now

Karl Marx taught us in 1867 that those who control the means of production control the future. Throughout the 20th century, the factory was the furnace of economic power, and the bourgeoise were those who owned the machines within.

Since the mid 1950’s the office gradually replaced the factory, fading workers’ collars from blue to white, and filling the pockets of shareholders along the way.

In 2025 the means of production are no longer smokestacks nor spreadsheets. Today, they’re GPUs, propriety models, training data, and the torrents of electricity that run them.

The arms race is on. Open AI, Microsoft, Meta, Amazon, X, and a handful of Chinese counterparts are not merely competitors - they’re landlords of tomorrow’s economy.

Labour (or more specifically ‘human capital’ for the purists reading), once central to value creation, is becoming peripheral. The marginal cost of reproduction—for code, copy, even R&D cognition—is near zero.

AI model development has every sign of being a winner takes most style market. Building a LLM currently features:

• High fixed costs: training large models can cost well over USD $100m.

• Network effects: plugins and APIs enforce uniformity across services.

• Exclusive partnerships for distribution: immediate access to to hundreds of millions of existing customers (think Grok on X, or Meta AI via WhatsApp).

Lock-in is inevitable. The firms that control these models do not just sell services, they set terms. Building new modern day technology products already relies on the infrastructure of these select companies. This is not mere licensing, it is feudalism by API.

I’m certain this time really is different. The origin stories of Facebook, Amazon, and Google, whereby a savvy founder from humble beginnings makes it big with a grand idea and a truckload of conviction — will not be repeated in the AI economy.

Instead, the model owners will ‘take most’, with everyone else set to scrap over ‘layer 2’ products applications. We saw similar developments in crypto and web3. The layer 1 blockchains - namely Bitcoin and Ethereum) have lasted the test of time and appreciated significantly, whereby the layer 2 protocols and products have been frothy… at best.

Keeping with the trend of Russian economic thinkers: What is to be done?

The ship has largely sailed on genuine competition. Incumbents are rapidly approaching the point where their R&D can generate further R&D autonomously. The window to compete on ‘layer 1’ AI models is effectively closed.

Perhaps unsurprisingly, governments are playing catch-up. Their regulatory instinct are rooted in past paradigms: consumer harm, misinformation, job displacement. But if they are serious about AI, they must think like Marxists, not moralists.

That brings us to the question of ownership. Public companies owning models (Alphabet, Meta, Amazon to name a few) have in-built checks and balances for governance and consumer safety. Crucially, they distribute value across broad shareholder bases.

Private companies — OpenAI, Anthropic, Stability AI — operate in far murkier waters. With few constraints and almost zero transparency, they hold an unprecedented amount of power unlikely to be challenged soon.

After all, those who control the compute will control the future.

Subscribe now

RCCD 1st edition was originally published in 2020 as a side project during the COVID-19 pandemic in Melbourne.

As I wrote in the midst of lock down:

I use R a lot for work and for side projects. Over the years I’ve collated a bunch of useful scripts, from macroeconomic analysis to quick hacks for making map legends format properly.

Historically my code has been stored in random Rpubs documents, medium articles, and a bunch of .Rmd files on my hardrive. Occasionally I feel like doing things properly - and upload code to a repository on github.

It doesn’t take a genius to realize this isn’t a very sustainable solution - and it also isn’t very useful for sharing code with others. It turns out 2-years of lock down in Melbourne was enough incentive to sit down and collate my best and most useful code into a single place. In the spirit of open source, a book seemed like the most logical format.

RCCD1e had surprisingly good (and long-lasting reviews). Alas, five-years on, a lot has changed - in both the R community, and the world more broadly.

As such, RCCD2e includes significant revisions. The most major of these is a restructure to make the chapters flow more logically. Additionally:

• The Australian specific chapters have also been grouped together (e.g. election data and ABS economic indicators).

• There are new chapters on functions, packages, as well as more advanced econometric techniques (Bayesian approaches and causal inference).

• Worked examples have been updated to reflect best (or at least better) practice.

• Many a bug and typo has been fixed.

Subscribe now

I have always loved weather. I’m also lucky enough to live in Australia where weather data for thousands of locations across the country is recorded and uploaded by the Bureau of Meteorology every day.

When thinking about living in different towns or cities, I’ve always struggled to picture what the conditions are like on the ground — particularly at different times of the year. Sure there’s snapshots of the heatwave or the morning frost, but what about between these events?

I think I’ve figured out the best way to visualise climate variables for a year-round view. Claus O. Wilke’s fantastic ggridges package includes a function called geom_density_ridges.

The result looks like this (for my new home-town of Melbourne).

These charts show all 365 data points (in this case for the temperature variable) — but it’s the simplicity that I think drives the message home. The curve peaks are the most common temperature, but the outliers and the seasonal change are all still there.

The best part comes in comparing these charts to other cities. Here is the temperature chart for where I grew up on the Queensland and NSW border.

A few things stick out when compared to Melbourne. Firstly, the entire distribution is shifted right (temperatures are a lot more moderate on the coast). Secondly, the weather is more consistent with the size of the ‘tails’ in each month tending to be shorter than in MEL.

The best part of these charts is that they can be used for any monthly variable. Rainfall, snow, sunlight hours — it’s a pretty neat way of getting the ‘vibe’ of what the weather is like in a new place.

Code for these charts is on my github here.