Should we maximise equality or utility?

Posted on May 23, 2026 by gezwinstanley

What matters more: maximising general well being or reducing inequality?

We seem to live in a world that tends to equate well-being with wealth and very much favours the former answer, sometimes to an extent that seems reckless and ruthless.

To protect wealth, amid our polycrisis, it is increasingly argued that the most vulnerable and the poorest should be abandoned. Austerity programmes are adopted; wealth taxes are pilloried; fossil fuels are to be extracted down to the last drop, with no regard paid to the climate vulnerable, even though whole parts of the world could be rendered uninhabitable.

Is there any moral justification to any of this?

In the earlier post “Is there a universal morality” we explored how respecting the wants and needs of others as equal to your own, provided they reciprocate, can be the only basis of a universal morality.

What does that principle imply when we come to decide whether to prioritise equality or utility?

So what do we mean by utility?

By utility we mean something similar to the “greatest happiness principle” of Utilitarianism, only instead of talking in terms purely of “happiness”, we are going to try and go a little deeper, and maintain that what sentient beings want, by definition, is satisfaction or “fulfilment”.

How does this relate to the “purpose of life”? Life can have no meaning or purpose that has worth to you unless it satisfies your nature at some level, unless you find it “fulfilling”. Would you accept a purpose imposed on you from without, which would negate everything you consider to have worth, and that would therefore leave you utterly unfulfilled?

Now it is possible that life has some deep purpose that would satisfy, and be fulfilling, for all minds. This would be the equivalent to the “Purpose” or “Meaning of Life” as normally conceived. However, we can not be certain that there is any such common, ultimately satisfying, purpose. Only your sense of what has, and has not, worth can give your own life purpose.

Seeking fulfilment is fundamental, but, when we interact with others and organise ourselves into societies, how should fulfilment be distributed?

Equality

The need to respect the wants and needs of others as much as one’s own, provided the favour is returned, unavoidably entails a strong commitment to equality, as everyone must accord equal respect to the wants and needs of everyone else.

Therefore no member of a moral community should, through action or even inaction, increase their own fulfilment at the expense of any other member of the moral community who already benefits from less fulfilment.

Therefore society must be organised so as to maximise equality, by ensuring no one gains fulfilment by reducing the fulfilment of anyone with less fulfilment than themselves.

This means that society should be organised to maximise the fulfilment of whoever then gets to be least fulfilled, noting that who specifically ends up least fulfilled would very likely vary as the exact social arrangements change. Otherwise we would not be respecting the wants and needs of whoever then gets to be least least fulfilled as equivalent to our own.

This principle also equates to “fairness” and “social justice”. It is not, it must be noted, a mandate for any form of “free loading” or indeed for any form of exploitation at all. Where work or effort in any way detracts from fulfilment, that work or effort should be counted against those workers’ fulfilment, meaning that, when assessing equality of fulfilment, those workers should enjoy sufficient total fulfilment to compensate for their sacrifices.

So what are are the implications of this notion of equality for the maximisation of utility?

Note that this commitment to equality does not require absolute equality of outcomes. It is radical but not “masochistic” egalitarianism. For example, it does not constrain the fulfilment of the most fulfilled in a moral community, provided those extra heights of fulfilment are not, in any way, obtained at the expense of the less fulfilled. For example, in so far as wealth contributes to fulfilment, it is not a mandate for everyone just being “equally poor”, unless, of course, being poor is what gives people fulfilment.

Where there are many ways our community could be organised that accord with the principle of equality, the one that should be adopted is the one that maximises open-ended fulfilment, where open-ended fulfilment is fulfilment that does not close off the possibility of obtaining still higher forms of fulfilment.

One source of fulfilment, which is both difficult to share and can result in some people being much more fulfilled than others, is the fulfilment we attain through our closest relationships. In many instances where we treat people better than others, it is because we have a duty of care, or are in a relationship of tight reciprocation, or some combination of both. Both of these types of relationships mean we have additional obligations to others but do not mandate, and morally cannot mandate, that, if there is a conflict of interests between those involved in such relationships and those who are not, but who are still respecting our wants and needs as much as their own, that we should prioritise the interests of the former over the latter.

Furthermore, the existence of these relationships, that involve treating others with extra care, is often pro-social because it generates general benefits. For example, conscientious child rearing on the part of parents both relieves the rest of society of the burden of child rearing and is likely to create more capable and prosocial adults. Such tightly bonded relationships can also serve as a broader exemplar or model to society as a whole: for example, the concept of “fraternity” very obviously derives from notions of kinship. Finally, as a source of comfort and fulfilment in their own right, such relationships can hugely improve well-being. A society where people are denied the fulfilment derived from close relationships would restrict the opportunities for fulfilment for all, including the least fulfilled. Therefore, these close relationships therefore do not violate our commitment to general equality.

This is, however, a radical egalitarianism, but one that follows inescapably from the principle that, provided the favour is returned, we should respect the wants and needs of others as much as our own, in other words, not be selfish. Amongst the implications of this principle of equality are that it is indeed true that, as stated by Nelson Mandela in “Long Walk to Freedom”, “A nation should not be judged by how it treats its highest citizens, but its lowest ones.” Any arrangement of living together in a moral community that benefits the least well off members of that community is to be preferred to other arrangements that do not help the least well off, even if those other arrangements would leave everyone else better off. Otherwise, the interests of the least well off are not being respected equally to those of the better off, thus violating this principle of universal morality.

It is also not the maximisation of total well-being at the expense of equality, thus avoiding some of the paradoxes that can arise with Utilitarianism. The most notorious of these conundrums involve justifications for slavery, such as those proposed in the Antebellum South of the US, including the “Mudsill Theory” that there must always be an underclass to support everyone else, where utility is maximised because the benefits to the non-enslaved outweigh the suffering of the enslaved. If more equality makes whoever would be worst off better off, even at the cost of reducing total well-being, then that increase in equality is moral.

So should we maximise equality or utility? The answer is neither or both. Really, we should increase equality by maximising the fulfilment of the whoever gets to be least fulfilled, while also maximising total fulfilment provided it does not in any way compromise the former commitment.

Why, if this moral imperative for equality is so simply derivable from basic morality, is it repeatedly ignored?

Most large-scale human societies, for all of history, have not maintained order primarily through a mutual respect of all for all, but rather by a combination of brute command, George Orwell’s image of a boot stamping on a face forever, and mass deception, from the divine mandate of ruling classes to “trickle-down economics”.

Nevertheless, the principle of universal morality, which is central to this explication, is essentially the Golden Rule of all mainstream religions. Formulations of this principle date back at least to the Ancient Egyptian story of “The Eloquent Peasant”, which is arguably also one of the oldest accounts of class conflict. This is unsurprising: this universal morality is, after all, the only way we can freely build populous societies of relative strangers.

However, the full implications of this universal morality challenge the selfish side of all of us, as well as the established power structures of nearly every human society that has ever existed. That the principle has repeatedly been muted, ignored or deceptively re-interpreted is tragically inevitable.

A time of crisis, though, is when we should seek guidance from our most foundational moral principles. Here the lesson is clear: the last people we should sacrifice are the poorest and most vulnerable.

Is there a universal morality?

Posted on December 31, 2025 by gezwinstanley

In Cixin Liu’s acclaimed “Three Body Problem” trilogy all civilisations in the Universe are conceived as living in a “Dark Forest”. In this dog-eat-dog, or alien-exterminates-alien, vision, all cower in the blackness terrified of even being detected by each other, and always attempting to eliminate their rivals if they do. In this seemingly never-ending genocidal war of all against all, ferocious weapons have been developed that bend the laws of physics itself.

In particular, species attempt to literally crush their rivals by reducing the dimensions of their space. Step by grim step, a once spectacular n-dimensional Universe has been collapsed down to our mundane three dimensions by the expanding side-effects of these devices. Now weapons have been used that are rapidly reducing even our own meagre three dimensions down to two.

Is there a way to avoid the cold, ruthless, logic of the Dark Forest? In a reality where humans, a single species, and one that is among the top 5% of the most genetically uniform primates, are often lethally vile to each other, murdering and exploiting on scales from the individual to the tens of millions, like an army of evil but mutually loathing clones, could sentient beings with no shared origins ever find a way to live together?

Is there, in short, a universal morality?

Focussing down further on this question, what would it mean for a morality to be “universal” in the first place? We are going to assume that, in this context, “universal” means a morality capable of, at least conceivably, being freely and knowingly accepted by any being with any capacity to understand it. After all, would we accept a morality as universal if it didn’t have those characteristics?

By contrast, a non-universal morality means mere moral relativism, and then inevitably descending back into the Dark Forest, as then there are no principles regulating how participants in the different “moral relativisms” should behave towards each other. Therefore, if we seek a general or ultimate morality at all, no other form of morality is even possible. The only fundamental morality that can be proposed at all is one that could apply to all beings: a “universal” morality as just described.

Note that this morality isn’t god-given, for these beings won’t share the same gods, or any gods, nor is it something evolved, or hard-wired, and inherent, because these beings may not share the same origins. Rather, a universal morality has to be derived from reason alone, in the same way as a proof in game theory.

So what is this universal morality?

In a way, we are likely to know the answer already.

What is “moral” is to respect the wants and needs of others as equal to your own, provided they reciprocate, i.e., to not be needlessly selfish.

Otherwise we have to posit entities that would knowingly and freely submit to a system of morality that would not accord their deepest wants and needs, meaning whatever it is they hold most dear, the same respect as the wants and needs of other adherents of the same system of morality. That involves those beings having to act “freely” but also in a way that does not follow from their own motivations, which is internally contradictory.

As implied, respecting the wants and needs of others as equal to your own, provided they return the favour, may sound familiar. As humans have come to live together in ever larger groups, it is roughly the morality that we have hit upon semi-accidentally many times before. It is very close to the Golden Rule, with an extra emphasis on reciprocation, associated with many mainstream religions and forms of humanism.

For example, it is equivalent to the principle of “We must treat others as we wish others to treat us”, adopted as a “Global Ethic” by the Parliament of the World’s Religions, or the “treat others as you treat yourself” of the first millennium BCE Mahabharata. That’s because, in the final analysis, the “treatment” we expect of others is to respect our wants and needs as much as their own.

It shares with the Categorical Imperative, “act as if you are a universal legislator”, of the philosopher Immanuel Kant, as it encourages us to act as we want others to act, rather than acting in a way that advances our own interests but at the expense of others.

It shares with Utilitarianism the principle of equivalence of the well being of all. In other words, that I can not value my own well being higher than that of others, though with a stronger impetus towards equality rather than merely maximising total Utility.

In these times of “polycrisis” we need to rebuild our variously failing social systems from the bottom up. We need a strong moral foundation to do that. We must re-discover this universal morality, as otherwise it is the “Dark Forest”, in some form or other, for all of us.

Is Britain Broke?

Posted on November 22, 2025 by gezwinstanley

Introduction

These days we are plagued by questions that roughly amount to “is Britain broke?”, and closely related inquiries such as “where did all the money go?”, or assertions such as “the government has run out of money.” Often, these statements are intended to frame a debate trying to justify yet more austerity, and are therefore, at a time when public services are creaking, destitution is rising and many are still suffering the effect of the cost of living crisis, very politically charged indeed.

Laying the groundwork

The below is an attempt to address these concerns in a rational, evidence-backed, and moral way.

To be clear, the moral assumption here is just that if we accept the need not to be selfish, meaning respecting the wants and needs of others as equally deserving as our own, provided they return the favour, then we are obliged to organise our society, including our economy, in such a way that whoever then gets to be the worst off is still better off than the worst off would be if we organised ourselves in any other way. After all, as the saying goes, there but for the grace of God, or simple dumb luck, go you.

Now, let’s refine the question. For starters, we should drop the alliteration: we mean the UK, specifically the people of the UK, not Great Britain. But what does “broke” mean?

Being “broke” means not having enough money to buy the things we need. Quantifying “need” is clearly open to debate, and that’s where moral considerations come in: to not be selfish, we have to treat the needs of whoever our social arrangements cause to be worst off more seriously than any other needs.

To become broke we need to run out of money, meaning that we have exhausted any savings, or wealth that can feasibly be converted into money. Therefore, another question arises: not just are we broke right now, but are “eating into” our stores of wealth? In other words, even if we aren’t broke, are we becoming broke?

Finally, money is an easily storable and exchangeable claim on resources: we use it to buy stuff. What matters if we run out of money is really that we then lose access to the resources we need. If we had direct access to all the resources we needed, such as place to live, food, clean water, energy and a capacity to manufacture goods, then we wouldn’t need money, and going broke wouldn’t matter. So the deeper question here isn’t is the UK broke, or going broke, but is the UK running out of resources.

To assess how our capacity to meet our needs is changing, especially as “need” is debatable, it is useful to have a point in time with which to compare. A moment where we are likely to agree that we had, as a society, a capacity to meet our “needs”, if not perfectly, for perfection is ever elusive, but adequately.

We are going to use 2008 as that reference year, because this was just before the Financial Crash, an event that still casts a long shadow over everything, as we shall see. Whatever, it’s faults, 2008, pre that Crash, pre-COVID, pre the war in the Ukraine, can seem like something of a golden age compared to our contemporary perceptions of creeping dystopia.

We will normally use 2023 as the end point of our analysis, simply because at the time of writing, 2025, that’s often the last year for which reliable statistics are currently nearly uniformly available across multiple datasets.

Also note this is a “whole economy” approach. What share of our combined resources we choose, or not, to devote to communal activities, through public spending, is up to us to decide democratically. It is a choice. Therefore the question “has the government run out of money?” is irrelevant: how much resource to use publicly, and administer through our different levels of government, is up to us. Provided we aren’t actually running out of real resources, of course.

Are we running out of resources?

Of course, we won’t be able to efficiently produce all the resources we need here in the UK, so some of the resources we produce here will have to be exchanged with resources produced elsewhere. So, as resources will often be produced for sale, i.e. converted into money, rather than used directly, it is the market value of resources, rather than “use value”, that we will need here. “Use value” is a meaningful and quantifiable concept (see gezwinstanley.wordpress.com/tulips-and-cryptocurrencies-is-speculation-theft for an example of how to quantify use value) and it isn’t the same as market value. For example, many essentials can be accessed freely, such as breathable air, meaning something with a value as great as life itself has zero price. Also, items caught up in asset bubbles can temporarily acquire hyper-inflated market values, as demand for the item is driven by buyers seeking to speculate on the market value of the item increasing still further, driving the price up yet more and feeding the bubble. However, the value of resources when we are extensively trading in resources is, by definition, market value, so we’ll have to stick to that for this analysis, even if market values often reflect the price of something and the real value of nothing.

So how has the market value of the UK’s total resources changed between 2008 and 2023?

First we should consider what’s called our net capital stock: the market value of the total supply of “made” resources, from homes to factories to intellectual property such as software. The “net” means that this figure attempts to allow for depreciation: that things wear out. In 2023 prices this value had increased from £4.8 trillion in 2008 to £5.8 trillion in 2023, so by roughly a trillion and 20.17%.

Secondly, we should consider natural resources, the market value of the total supply of non-made resources, those available just in the environment, which the UK was then able to utilise. This stock can increase because previously unavailable resources become accessible, e.g. we can exploit more renewable energy, or decrease due to resources being used up or as a result of environmental degradation. In 2023 prices, this value had increased from 1.2 £ trillion in 2008 to 1.7 £ trillion in 2023, so by roughly half a trillion and 42.78%. However, these types of “ecosystem services” statistics are still under development, not just methodologically but, to the extent that some of the component figures for 2008 and 2023 were missing, and had to be imputed on the basis that anything missing would form the same proportion of the total as it did in 2022, when complete figures are currently available. Therefore, we should use some caution here.

Then of course we have people: the total market value of the work that we could do. This has been estimated by taking the working population in 2008 and 2023 and multiplying by the corresponding average wage figures, converted to 2023 prices. The annual figures are £1.3 trillion in 2008 rising to £1.4 trillion in 2023. It would be useful to be able to simply add this figure to net capital and the total for natural resources. Technically, however, this would be difficult to do, as the annual value of our work is not a stock or a “level”, it is a flow or a rate. To calculate the market value of the total stock of labour we would need to derive at least an average figure for the value of work to be generated for the remaining working life of each worker. Then we would need to assert a “discount rate”, to turn future years potential earnings into what is technically called “net present value”. This is because, and with apologies for making this sound like a slave market, anyone “buying” the value of the worker’s work won’t be willing to pay as much for, say, the work 10 years in the future as this year’s work, because of all the uncertainties involved. This all involves so much guesswork that your conclusions will depend on the assumptions you make far more than anything “objective”. So, sorry, there is no complete total of UK resources being stated here, as there is no straightforward way to add the value of the labour force, as a resource, to the values of the stock of made and unmade resources. The trend in the value of the UK’s total labour resources is more meaningful, and this implies a real term growth of 5.57% from 2008 to 2023.

The larger our population then, provided that population is working, the more people we have to create resources. We also, though, have more people to supply, more “mouths to feed”. Therefore, we should look at what these changes in UK resource levels mean on a per capita basis. Economic figures that attempt to assert a population’s well-being on anything other than a per capita basis are at best incomplete and at worst not to be trusted.

From 2008 to 2023 the UK population has increased from 61 823 800 to 68 265 200, or by 10.42%.

That means that when we divide the total market value of labour resource by the population, we actually get a small decrease, of 4.39%. That’s because, from 2008 to 2023, the increase in that value has failed to keep up with the population increase, mainly because we had to use real-term average wage figures to estimate the market value of labour and real-term mean earnings (not median: it’s usually the median that gets quoted) fell by 2.46%, largely due to COVID and high inflation after 2021, but also because the proportion of the population that was of usually non working age (16 or under or 65 and over), or of working age but inactive, increased by 1.99%. Most of the latter increase was due to the ageing population, with the number of over 65s increasing by 30.63% from 2008 to 2023 (from 16.00% to 18.93% of the total UK population), where-as the number of 16 years old or younger increased by 6.56% and the number of people of working age but “inactive” increased by 0.99%, i.e. just under 1%.

The increases in both net capital stock and natural resources, however, more than kept pace with the population increase. In 2008, but in 2023 prices, per person “made stuff” had a value of £78 259. In 2023 the same figure is £85 168, so the per capita value has increased by 8.83%. Similarly, the per capita share of natural resources has increased, in 2023 prices, from £19 686.40 in 2008 to £25 455.50 in 2023, or by 29.30%.

Overall, there is no strong evidence that the “real resources” of the UK have shrunk from 2008 to 2023: on the contrary it is more likely these have increased, though the changing age composition of the population is an emerging challenge to which we should remain alert. We do not seem to be going broke in the sense that we are gradually running down our stock of resources, which is unsurprising given the propensity of “produced items” to accumulate faster than they depreciate, and continuing technological advancement.

However, these “stock of real resource” statistics are still under development, so we will also look at answering these questions in two other, more conventional, ways: how has our nation’s wealth and income changed since 2008?

(For methodology and sources see: Note 1 ).

Are we running out of wealth?

The data source we have used here defines “wealth” as: “Net national wealth is the total value of assets (cash, housing, bonds, equities, etc.) owned by the national economy, minus its debts. The national economy – in the national accounts sense – includes all domestic sectors, i.e. all entities that are resident of a given country (in the sense of their economic activity), whether they belong to the private sector, the corporate sector, the government sector. “. This therefore differs from real resources because it only includes things that are “owned” (ecosystem services are often not owned) and also includes financial assets.

In 2008, in 2023 prices, the net national wealth of the UK was £9.389 trillion, rising to £10.233 trillion in 2023, an increase of 8.98%. However, as we can see, this is lower than population growth, so per capita wealth falls slightly from £151882, in 2023 prices, in 2008, to £149904 in 2023, or by 1.3%.

Again, this fall was largely due to high inflation post COVID and Russia’s invasion of the Ukraine, plus the higher interest used in an attempt to suppress that inflation. That meant the price rises of many assets did not keep pace with inflation and the face value of bonds fell. If, by money, we roughly really mean wealth measured in market value terms, that may be a small part of the explanation for our experience that the wealth has “gone” somewhere.

Of course, these are often “paper” values, which can fluctuate without the underlying “use value” of any “real resources” which those “paper” values represent changing. For example, a “house” doesn’t become less useful when house prices fall because interest rates have risen and choked off demand. There is one crucial exception to this pearl of wisdom, though. The “paper” value of something, it’s market value, is the value for which that item can be sold, with the returns then possibly used to buy real resources. Therefore levels of market value, and often purely financial, wealth have huge implications when it comes to being able to access real resources.

When we look at how the distribution of this wealth changed, though, we can further address the question of “where did the money go?”. From 2008 to 2023, in 2023 prices, the average per capita wealth of the wealthiest 1% of the UK population increased from £3 068 016 to £3 192 955.2, so from roughly 3.1 to 3.2 £ million, or by 4.07%. By contrast, from 2008 to 2023, in 2023 prices, the average per capita wealth of the poorest 50% of the population decreased from £19 137.13 to £14 090.98, or a fall of 26.37%. That means that in 2008 the average member of the top 1% had about 160 times the wealth of the average member of the bottom 1%, but by 2023 that factor had risen to about 227 times.

(For methodology and sources see Note 2)

Are we running out of income?

Another way we could be going broke is if our income is falling. Here we’ll measure national income in a fairly conventional way, as Gross Domestic Product, which can be interpreted as the sum of the income derived from all domestically produced goods and services in a given year. Note that Gross Domestic Product, unlike wealth or the sum of resources, is therefore a flow or rate, rather than a stock or a level.

UK GDP rose from 2.3 £ trillion in 2008 in 2023 prices to 2.7 £ trillion in 2023, or by 18.85%. In per capita terms that meant a rise from £37 000 a year in 2008 in 2023 prices to £40 000 a year in 2023, or by 7.64%.

Therefore, far from our national income falling, it has increased. Since the Financial Crash that increase has undoubtedly been slower than in previous historical periods, due to austerity, COVID and most recently energy price shocks caused by Russia’s invasion of the Ukraine, with an added dose of Brexit, of being the first nation to impose sanctions on itself. However, an increase, in real, per capita terms, is still an increase: it is not going broke.

We should also ask if, as with wealth, the income figures provide any answers to “where did the money go” style questions. Now GDP is an aggregate figure for a nation as a whole: therefore there are no “GDP” figures for individuals or families. Therefore we will need to look at actual individual income statistics. We will use the same source as we did for wealth, which defines “Pre-tax national income” as “the sum of all pre-tax personal income flows accruing to the owners of the production factors, labour and capital, before taking into account the operation of the tax/transfer system, but after taking into account the operation of pension system .” Working from that source, the average per capita pre-tax income, in 2023 prices, for someone in the top 1% of earners, was £444 512 in 2008, rising to £445 728 in 2023, or increasing by 0.27%. For the bottom 50% of earners the corresponding figures were £12 691 in 2008, rising to £13 678 in 2023, or by 7.78%. This means that in 2008 the average person in the top 1% had an income 35 times that of the average person in the bottom 50%, while in 2023 the corresponding ratio was 32.6, so a slight 7% improvement in income inequality, likely mainly due to the “triple lock” alleviating pensioner poverty.

Looking again at the figures, that means that, on average, those in the top 1% received a rise of just over a thousand pounds from 2008 to 2023, while those in the bottom 50% received a rise of just under a thousand pounds, meaning that, very roughly, the increase in the national income since 2008 was distributed evenly between the two groups, slightly alleviating income inequality in relative terms. On the face of it, income inequality has not increased and there therefore isn’t any kind of answer here as to “where did the money go”. How do we then reconcile this with the very significant increase in wealth inequality? Shouldn’t people getting generally richer correlate with them earning more, both because higher earnings mean they can save more, and more riches generate higher earnings?

Now, as earnings increase the rate at which an individual can accumulate wealth tends to increase disproportionately: if they were able to save, say, 10% of their income before an increase, then, after a raise, and even if they spend much of the increase, they will have more surplus income than before and be able to save, not just 10% of their increased income but, say, 12%, so proportionately more and therefore potentially significantly absolutely more than before. Consequently, high earners can save much more than low earners, who very often can save nothing and are likely to be in debt. For any income inequality ratio the corresponding wealth inequality ratio is therefore likely to be much higher, so what we see here, a top 1% compared with bottom 50% income inequality ratio of 33 in 2023 but a wealth inequality ratio of 227 is unsurprising. However, by these considerations alone, that wealth inequality ratio, though very significant, should at least stay the same. The fact that high earners can save much more doesn’t explain how the wealth inequality ratio could increase from 160 in 2008 to 227 in 2023, unless income inequality was rising in proportional terms, which, as we have seen, it, generally and on average, wasn’t, at least when we compare the top 1% to the bottom 50%.

The likely explanation here is that wealth creates wealth, as most famously analysed by the economist Thomas Piketty, meaning that wealth tends to grow exponentially. There are two ways this can happen. One is pure “asset appreciation” where the “use-value” of owned resource stays constant but there is a “paper” increase in market value. This can happen because assets are in an asset price bubble, or it could happen because there is a fixed or very slowly growing supply of some asset, and shortage is driving up the price, or some mixture of both factors. The obvious example here is housing. Note that wealth can easily beget wealth in such a situation because the richest can hoard the assets, increasing the shortage and therefore further driving up the price to purchase or rent those assets. In this way, the wealthiest can enrich themselves while impoverishing everyone else: hence the claim that “the rich will own your Nan’s house,” made famous in the UK by anti-inequality campaigner Gary Stevenson. The other way wealth can beget wealth is more straight-forward: resources are acquired and invested to create still more resources, for example setting up a factory and increasing production in some way. Repeatedly, the wealth arising from either asset appreciation or expansion of production can be further invested to create still more wealth.

Nevertheless, we still have to ask, if wealth is creating wealth in this way why isn’t it showing up as an increase in proportionate income inequality? The answer is that the wealth created by wealth will tend not to show up as earnings within limited time-frames. If an asset simply increases in value that will not show up as a capital gain unless the asset is sold. If a company makes more profits those will not be paid out as dividends if the profits are re-invested: instead the value of the company will increase and, if it is listed, so normally would its share price. The wealth generated by many investment vehicles, such as private pension funds or stocks and shares ISAs where dividends are re-invested, will not show up as individual income until many years after it was generated.

Therefore, there is no inexplicable contradiction here between roughly flat income inequality over this fifteen year period but increasing wealth inequality, and therefore no reason to doubt that increase in wealth inequality, or that wealth inequality could go some way towards answering questions along the lines of “where did the money go”. Essentially, the average increases in the wealth of the richest currently includes a lot of the “deferred” increased income of the richest.

(For methodology and sources see Note 3)

How are our needs changing?

At this point it seems we are not, as a nation, going broke in any meaningful sense. Our stock of real resources and wealth seems to be reasonably healthy, and our national income has grown in real and per capita terms. If anything, we may start to suspect that the issues we are experiencing are more to do with maldistribution of our resources than any absolute shortage of resources.

However, there is another way in which we could, as a nation, be going broke, which is if our needs have grown so fast that our income hasn’t kept pace.

This begs the question of how to measure our changing “needs” as a nation. Population is part of that, which is why, in assessing all this, measures in per capita terms are essential. Beyond that, though, how are our needs, as a population, changing as the characteristics, and not just the size, of our population changes, and how is the cost of what we need changing? Are we, in short, having to spend more than we have coming in, using up more resources in real market value terms than we are creating?

The review of the stocks above: resources and wealth, would imply that the answer to that question is likely to be “no”, because if the answer was “yes” those would be running down consistently. However, we did identify a small likely per capita fall in the value of our labour resources and in our per capita wealth, though that seems attributable mainly to recent inflation in both cases, so let’s try and assess our needs in more detail to see what the challenges really are in terms of our changing needs.

To do that, we used the Treasury’s high level breakdowns of public spending, for 2008 and 2023, with everything converted to 2023 prices, and also, to get a handle on our private per capita average adult income, took GDP, subtracted the total public spending, and then divided what’s left across the population, in this case across everyone aged 20 or over. We will further assume that in 2008, as asserted when we were laying our groundwork for this analysis, we were roughly adequately meeting our needs. That means we need to assess how our needs have changed since 2008.

In 2008 total public spending was 886.83 £ billion, in 2023 prices, or 38.88% of GDP. When we follow the above methodology, we find that, in 2008, the per capita income left after deducting total public spending from GDP, in 2023 prices, was £29 735.64, rising to £30 356 on average in 2023, so a real terms increase of 2.09%. Also note, that these are the average per capita income figures after we have subtracted public spending, they are free of all taxes, fees and fines, free of everything from direct taxes such as income tax, national insurance or council tax through indirect taxes such as VAT or stamp duty to fees, such as the cost of passport renewal.

Again, then, we see a pattern of low growth, but still very much not, as a society, going broke.

However, the fifteen years after 2008 and the Financial Crisis were years of austerity. We will define “austerity” here meaning not simply real term cuts in public spending, but instead changes to public spending that mean that services can no longer meet needs as effectively as they did before. If needs for a public service are increasing faster than general inflation, say due to changes in the size of a component of the population, or cost pressures rising more than general inflation in this area of activity, then a budget can increase in real terms, but still be an austerity budget, because needs that were previously being addressed are now being left unmet, causing harm to the general public. In the case of what has happened since 2008, various studies link austerity to hundreds of thousands of excess deaths (for example, see this British Medical Journal report: jech.bmj.com/content/76/12/1027).

So we are left with having to assess how needs, quantified in terms of the spending required to meet those needs, has changed from 2008 to 2023. To do this, we again took the Treasury’s high level breakdowns of public spending for 2008, converted to 2023 prices. Then we increased each of the budgets to allow for population growth, except for two particular areas which we increased even more. Those budgets where we applied higher increases are health and social care. For healthcare we applied the average figure for real term increases in NHS spending for the first 70 years after its foundation, which is 3.7% a year, for fifteen years, to the 2008 figure for health spending in 2023 prices. For the social care (personal social services) component of the overall social protection budget we scaled the budget not by general population growth but by the higher growth in the number of people aged 65 or over. For each sector of spending, we then took whichever was the higher figure: the actual figure for 2023 or our projected figure.

This resulted in a projected, non-austere total public spending for 2023 of 1197.16 £ billion, or 44.16% of GDP. For comparison, the actual total of public spending in 2023 was 1114.3 £ billion, or 41.10% of GDP. For more context, according to the OECD, “General government expenditures across the OECD averaged 42.6% of GDP in 2023,” and “On average public spending in OECD countries that are also EU members was at 49.3% of GDP in 2024” (See : https://www.oecd.org/en/publications/2025/06/government-at-a-glance-2025_70e14c6c/full-report/general-government-expenditures_395dfea8.html). Neither of these GDP shares is therefore unusually high.

We may ask what are the reasons for the 3.7% annual long-term real term growth figure for health spending. If spending had followed that trend from 2008 to 2023 that would have meant a 72.5% real term increase over 15 years, instead of the actual 42.48% real term increase, and 29.04% real term and per capita increase, that did occur. There are many reasons, though, why it is very likely that underlying need for healthcare increased at rate closer to the long term 3.7%. These include, as well as an increasing population, an ageing population, with higher incidence of co-morbidities, and more expensive drugs. In more recent years, we could add the bed-blocking caused by over-stretched social care, the aftermath of COVID, under-funding of capital investment, and staff shortages leading to expensive back-filling with private resources, such as agency doctors and nurses, and what could be an emerging mental health crisis. Of course, by putatively fixing social care through our hypothetical non-austere public spending, we should have solved the “bed blocking” problem, and the NHS staff recruitment and capital investment issues, but those were only recent factors contributing to long term increases in health spending, which extend back to periods where the social care wasn’t in crisis and the NHS was adequately funded. We also wouldn’t have neglected mental health spending for as long as we actually did. Though those are examples of how austerity is counter-productive, serving in the longer term to increase rather than decrease costs, overall 3.7% is a reasonable figure to use here, in the absence of any other.

The ageing population is especially notable. Any glance at projections of how the composition of our population will change will create the impression that we will have to cope with an ever increasing proportion of older people indefinitely. However, it is vital to understand that the proportion of older people in a population tends to change for two main reasons: increasing life expectancy and the ageing of a higher birth cohort who were actually born decades ago. In the former case, provided we triage provision of healthcare more to quality of life rather than basic life extension, increasing life expectancy should also mean an increasing “healthspan”. In this situation, we can sustain a larger older population by retiring later, while still enjoying as many years of actual and active retirement as previous generations, or perhaps even more years, if we retain the duration of retirement as a proportion of life. The second case, where an ageing population is due to births over a certain period being higher than later equivalent periods, and then those people simply growing older, is more difficult, as there is no reason to believe that those elderly people will be any healthier than previous cohorts of the elderly, as there is when life expectancies increase. However, this challenge, which is obviously the one we face with the “Baby Boomer” generation born in the couple of decades after the end of the Second World War, can be temporary. The simplest, and most responsible, way to deal with this is to accept the need for higher spending and higher taxes while the “Baby Boomer” generation passes through its old age.

Now, this is back-of-the-envelope, but having roughly factored out austerity we end up, after we have subtracted these increased, non-austere, totals for public spending from GDP, with a per capita, for everyone aged 20 or over, after tax income in 2023, in 2023 prices, of £28 781.

Note that this is slightly lower than the equivalent 2008 figure of £29 736, a decrease of 3.21%, but in exchange for fixed public services, including a fully functional NHS, social care and social housing being provided at 2008 levels, continuing NHS dentistry and fewer potholes. As public spending provides most of us with benefits in kind, or “transfer payments” in the jargon, it is possible that would, for most people, more than compensate for the implied £955 annual loss even in pure cash terms. Moreover, without austerity public services are likely to have been provided more cost-effectively: for example, better healthcare means lower benefit payments to cover long-term sickness and more people able to work and pay taxes. Therefore, it is more likely than not that the non-austerity picture being painted here is over-pessimistic, and without austerity we could have both preserved public services and still had the same or more disposable income left-over than in 2008. As we have seen, austerity is usually penny-wise but pound foolish. In terms of public services, and the economy as a whole, it is like trying to save money by avoiding regular car maintenance, only to have the car then explode.

Also, of course, the extra public spending would, in and of itself, directly add to GDP. However, resources are finite, so unless the public spending is tapping into resources that are currently under-utilised, such as people who are under-employed, or workless, or relatively unproductive due to under-investment, resources would end up used by the public sector that would otherwise have been used by the private sector. So the public spending would net add to GDP where it mobilises otherwise under-used resources, or where it ultimately uses those resources to produce more market value than the private sector. Again we’re talking market value here: as the public sector tends to supply the most essential services, it is likely that a given amount of public sector investment would produce more use value than the same amount of private sector spending, on average. The public sector responds to needs that are expressed democratically, with the requirements of every voter, in theory at least, meant to be given equal weight, where-as the demands that are best served by the market are those of the customers with the greatest ability to pay. However, GDP, for good or ill, isn’t measured in use value, so for now use value isn’t relevant.

The usual assumption is that the public sector will use any resources it acquires less efficiently than the private sector, in terms of market value output for market value input, because private sector activity is honed by competition. This isn’t, however, necessarily the case. The private sector has to competitively work for a profit, and continually seek to maximise that profit. That means not only that there’s an extra margin to pay for any services provided on a profit-making basis, but that businesses have to compete ruthlessly, to grow or die. That makes enterprises focus relentlessly on the just mentioned efficiency, but it also means that they are amoral, and will exploit customers, workers and suppliers if given the chance. Private enterprise has to prioritise its own short term concerns: co-operation and long term planning, or inadvertently conferring advantages to competitors, have to be avoided as impediments to the unforgiving necessity of continuously having to outgrow rivals. So whether to go for public or private spending depends on the context. If the market value to be generated by some spending is of a general benefit that cannot easily be monopolised and there-by monetised, and a public good such as street-lighting is the example often given here, or form an intervention that is part of a wider or longer-term plan coordinating different aspects of the economy, or utilise resources than the private sector would otherwise ignore, such as employing the workless and so bootstrapping an economy out of a slump, then public spending can be more efficient, in terms of bang for the buck in increased GDP, than private sector spending.

A telling example here is the development and deployment of the COVID vaccines. As knowledge is very difficult to monopolise and therefore monetise, and as Big Pharma did not want to waste its resources on preparing for a pandemic that might never happen, all the basic research here was done with public and charitable funding. When the pandemic hit, governments allowed various pharmaceutical companies to use that research to manufacture and distribute specific vaccines at pace, which they did efficiently and effectively, saving at least an estimated 20 million lives (according to an Imperial College London study cited here: www.covidvaccineresearch.org/news/covid-19-vaccines-prevented-20-million-deaths-worldwide-first-year-estimates-imperial). As part of the deal, though, the pharmaceutical companies were allowed exclusive patents on the vaccines. As it did not suit Big Pharma’s profits to allow poorer countries to manufacture their own versions of these vaccines, the pharmaceutical companies lobbied their governments to reject calls for a waiver of their patent rights. As a consequence millions more died unnecessarily (see for example www.globaljustice.org.uk/news/over-10000-people-die-every-day-the-uk-blocks-covid-vaccine-patent-waiver , which arrives at a rough estimate of 3.7 million). The private sector can be very effective at commercial production, but completely ineffective at managing any form of public good or addressing the needs of those who are unable to pay.

Let’s do a thought experiment. Conventional wisdom would have it that before we can, say, turn someone into an additional nurse, we have to grow the rest of the economy first, to generate the taxes needed to pay for the nurse. So, for example, a government, through the creation of the right incentives, including often tax cuts and deregulation, could encourage a stationary supply company in Slough to increase its paper-clips sales. That company would then perhaps hire more workers, or pay them more, thus allowing the government to receive more tax from those workers, or from the profits of the company itself, and through the VAT paid by customers when they buy the paper clips. Then, and only then, can the extra taxes pay for the additional nurse.

This is, of course, nonsense, that gaslights us into thinking of a whole range of public services as just “overheads” supported by a real, market economy, rather than as essential services and vital “economic goods” in their own right. The government should just go right ahead and pay for the nurse. It doesn’t even need to wait for tax receipts first, as logically governments have to create money before anyone can pay that money back to governments as taxes.

But is training the person in question to be a nurse a less efficient use of resources than, say, them becoming a manager at our stationary supply company? That seems very unlikely indeed. Clearly, in terms of improving our health, the nurse has considerable use value, but what about market value? Here we should note that the nurse helps to produce a succession of fixed patients, many of whom can return to being economically productive in their right.

On balance, it is very likely that less austerity would therefore have meant a higher GDP, allowing both increased tax returns and more disposable income, making our figures above pessimistic. Nevertheless, even making those very pessimistic assumptions, we would still be able to reverse austerity and meet our needs roughly as well as in 2008. It’s tight, and some economic growth, of the right kind, would be welcome to make it less of a squeeze, and to shield us against future shocks, but our needs have not rocketed away from our means.

Despite this we have chosen not even to meet our most basic needs, as we shall see next.

(For methodology and sources see Note 4)

Rising destitution.

Remember that when setting out the groundwork for this analysis we made the moral assumption that we need to avoid being selfish, meaning that we are obliged to respect the wants and needs of others as equally deserving as our own, provided they return the favour.

The upshot is that, at any time, we must prioritise the needs of the worst off. This moral assumption compels us to strive for equality by levelling up, because we have to regard everyone’s needs as equally deserving as our own.

Through this lens, our austere economy has failed miserably, because, since 2008, the worst off have become considerably worse off. This effect, though, isn’t immediately obvious if you look at the long term standard statistics on poverty, such as the percentage of households with absolutely low income or relatively low income, measured as people living in households with income below 60% of median income. In the absolute case, the median income used for comparison is from a base year, currently 2010/11, while the relative poverty figures use each year’s median income. In both cases, though, all figures are adjusted for inflation using the Consumer Price Index.

To be specific, in 2008/9 18% of individuals were receiving relatively low income, and 17% receiving absolutely low income, as compared to 17% and 15% respectively in 2023/4. ( www.gov.uk/government/statistics/households-below-average-income-for-financial-years-ending-1995-to-2024 ). On the face of it, that means poverty is improving, and that is consistent with what the income figures we cited above showed for the poorer half of the population too.

How then do we then explain that, according to the Trussell Trust, the number of emergency food parcels they distribute has gone from mere thousands in 2008 (barely showing up in this graph in section 2.2 of the House of Commons report here: researchbriefings.files.parliament.uk/documents/CBP-9209/CBP-9209.pdf) to 3.1 million in 2023/4, consistent with the general impression that food banks were barely “a thing” back then, but are now commonplace? Similarly, there has been a dramatic rise in homelessness over the same period, with the number of households in temporary accommodation increasing from 67 480 in Q4 2008 to 112 620 in Q4 2023 (according to the UK government homelessness figures here: assets.publishing.service.gov.uk/StatHomeless_202503) , a rise of 66.89%, or still a rise of 51.15% even after we remove underlying population growth?

This is the issue of increasing destitution, which has now drawn sufficient attention that attempts are being made to define and measure it more generally. For example, the charity the Joseph Rowntree Foundation now assesses “destitution” regularly, which it defines as “Destitution denotes the most severe form of material hardship. People are considered destitute if they have not been able to meet their most basic physical needs to stay warm, dry, clean and fed.” The charity estimates that “over the five years between 2017 and 2022, the number of households facing destitution more than doubled, with even higher increases for people overall and for children.”

These statistics on destitution do not extend back to 2008: hence our use of food parcel and homelessness figures above. It means, despite all the data above that we are not worse off as a nation than in 2008, and overall very likely slightly better off, our economy is doing a worse job in 2023 than in 2008 in terms of ensuring everyone’s basic needs are met, such as shelter and food. Our awareness of this issue is another reason why we often wonder “where did all the money go”? So how can we reconcile this with the apparent improvements in the low income statistics above?

The answer is that the inflation experienced by the poorest is different. During the post COVID and invasion of the Ukraine period, that has been very obviously the case with heating and food inflation usually outstripping most other items. In terms of another essential, which is rental costs, this has been a trend for far longer. As these are all necessities for survival, and the spending of the poorest has to be focussed primarily on necessities, the poorest are affected by high inflation in the “basics” more than anyone else.

The poorest almost always have to rent their own homes. So how did rental costs change between 2008 and 2023? The ONS Price Index of Private Rents only extends back to 2015, but the increase from Jan 2015 to December 2023 implies an average annual increase at a rate of 3.18%, or a total increase over fifteen years of 160.03%.

Similarly, the ONS collates statistics for household cost indices by income decile, and the “all items” index for the poorest tenth of the UK population, by income, shows costs increasing by 162.18% from April 2008 to April 2023. For comparison, general inflation in the economy from the financial year 2008/9 to 2023/4 was 143.14%.

The poorest are clearly experiencing much higher inflation. To illustrate what that means, taking the UK official government earning figures for households with below average income, to derive an annual income for the poorest tenth, this income, in terms of those poorest households’ actual purchasing power, in 2023 prices, fell from £13668.75 in 2008 to £12220.00 in 2023. That is a decrease of 10.60% and easily explains why people, already struggling to keep a roof over their heads and who may be falling ever deeper into debt, are being driven to food banks and into homelessness.

Above we discovered that, for someone in the top 1% of earners, in 2023 prices, average income rose from £444 512 in 2008 to £445 728 in 2023. That is with 2023 prices adjusted for general inflation across the economy. As the richest have been much less affected by the price increases in life’s basics, and instead often see the flip side of that, such as receiving rather than paying those higher rents, we have likely seen a situation where, in terms of effective purchasing power, the rich have got richer and the poorest, those most in need, who should morally be society’s priority, have got poorer.

Despite the fact that there is no evidence the UK is, in any sense, going broke, with us, instead, having had the capacity, as a society as a whole, to end austerity at any point, we were less willing to feed and shelter the entirety of our population in 2023 than in 2008. This was clearly not a matter of economic necessity, but a political choice. Not only is there a strong economic case for ending austerity, but the moral case for ending austerity measures such as the housing benefit freeze and the two child benefit cap, seems overwhelming.

(For methodology and sources see Note 5)

Could we go broke in the longer term?

So, in terms of public spending, our society clearly has critical unmet needs. In practice, we were never, as a society as a whole, broke and could have resolved that problem with some combination of money creation, increased public borrowing to invest and higher, and more progressive, taxes. As austerity is self-defeating, our economy would then, almost certainly, have been healthier as a result, plus we would have properly functioning public services and much less desperation.

We have seen, though, that our needs are likely to grow faster than our population for the foreseeable future, because of our ageing population and the increasing costs of healthcare. Some per capita growth is therefore required if we want to avoid continually increasing public spending as a share of GDP, including addressing the concern that such a trend is not sustainable indefinitely. However, it has to be the right kind of growth. We need growth directly targetted to meet real human needs and that lowers the carbon intensity of our society. The simple answer to the growth or “de-growth to save a habitable planet” argument is that it depends on exactly what is growing..

What pace of growth are we talking about here? As just stated, let’s assume that we want to keep public spending as a constant proportion of the economy. In practice some temporary growth in proportional terms may be justifiable as the Baby Boom birth cohort passes through old age, but let’s keep this constraint as described, at least for the purposes of illustration. Then let’s calculate the average economic growth needed to match the growth in public spending from 2008, in 2023 prices, to 2023, but not to actual public spending in 2023, which was highly deficient, but to our projected, and arguably pessimistically high, non-austere public spending. This means our GDP would need to grow by 2% a year to keep pace with our needs. We should remove population growth from that figure, as otherwise we could cheat and “grow” our economy just by importing more people. When we do that, we find we would need to grow our per capita GDP by 1.35% a year. That is not historically very high, but it is nearly three times the actual average annual growth in per capita GDP over this fifteen years, which is 0.49%.

So how can we grow our economy? It would overwhelm this article to even attempt to make detailed plans here, but a few points should be obvious.

Economic growth is about using existing resources as efficiently as possible to create new resources. Though this means focussing on what is termed the supply side of the economy, use and misuse of resources often happens because of issues of demand. The right level and pattern of demand can stimulate resources to be used in a way that is broadly beneficial across the population, Both austerity and inequality, two phenomena currently so causally inter-related it is sometimes difficult to prise them apart, damage demand in ways that discourages an efficient use of resources. A concentration in spending power among the very richest can stifle demand and result in under-utilisation of resources, as the richest save a much larger proportion of their income than the poorest, and often do so off-shore or in safe havens such as gold, or in speculation on the stock markets, rather than as genuine investment in the domestic economy. The spending the richest actually do can also cause a misallocation of resources, perhaps directing productive capacity towards luxuries rather than essentials or meaning that critical resources, such as housing, get hoarded. Similarly, austerity can mean that investing in essential “public goods”, such as the health of the population, gets neglected, or that coordination problems, such as mobilising resources in such a way to grow the economy as a whole, don’t get solved.

So reversing austerity, and reducing inequality by levelling up, starting with the poorest, in itself will promote economic growth, and most likely growth of a more responsible kind that better meets general needs. Everyone would feel the benefits of such growth, and not just the richest.

The main solution was half-accidentally already arrived at in the mid 20^th Century, with the New Deal in the USA and especially with the subsequent war economies, which ended the long stagnation of the Great Depression and gradually initiated a long period of economic boom. We need an “industrial strategy” and the stimulus of public investment targetted to achieve that strategy.

That “industrial strategy” should concentrate on addressing fundamental resource bottlenecks, and do so with direct public investment rather than tinkering with incentives and deregulation or re-regulation in the hope we can cross our fingers and let markets pick up the slack. Very often we are talking here about coordination problems and public goods issues that markets can never address.

Firstly, we should reduce our crippling energy costs. Breaking the link of energy prices with gas prices is an obvious first step. Then we need to expand our energy generating capacity. Not wasting the second highest tidal range in the world by building a tidal reef across the Bristol Channel is something that is long overdue. Some of the energy generated could then be used to supply the nearby steelworks of south Wales with cheap, green hydrogen, allowing a renaissance of British steel making in a sustainable, carbon neutral, form. Expanding wind power: the cheapest form of energy generation in the UK, is equally obvious. We should also consider more geothermal energy, including newly emerging possibilities for deep geothermal.

Then we should get on with building new homes. Here we should consider using new approaches to make efficiency improvements in construction, including automated 3D printing and, as we did after World War Two, greater use of prefabrication. There is no reason to carry on building homes like we did in the Middle Ages, so there is considerable scope for improvements in productivity and efficiency. The new homes should also be cheaper to run, with better insulation and built in energy generation such as solar or heat pumps, and rainwater capture to reduce water charges. We could also ensure that these new homes are better adapted for multi-generation use, allowing people to stay in their homes longer and so reduce demands on social care, for example by specifying that all staircases should be designed in such a way as to allow cheap retro-fitting with stair-lifts as and when needed.

Similarly, we need to get food inflation under control by modernising food production and better ensuring food security. Livestock farming uses considerable amounts of land and water, and is carbon intensive, so we could, for example, encourage more precision fermented, factory-produced, food by mandating that all public sector catering contracts have to replace meat with these products, giving this sector of our economy a huge boost, while livestock farmers could be financially incentivised to shift their land over to use for carbon sequestration, recreation and nature restoration.

We should resolve our multiple public health crises investing more in the NHS and taking other steps too. We would train more medical staff, and restore NHS capital investment, equipping the NHS with technologies best able to drive its efficiency, such as AI to assist with patient monitoring and diagnostics and more digitally-enabled services that could be accessed from home. We should also create a publicly owned pharmaceutical and biomedical treatments company. That can do multiple things for us: boost still further our already world-leading biotech industry, reduce our reliance on often US based Big Pharma, make cheaper drugs and treatments available to the NHS and concentrate efforts on finding cures and extending “healthspan”, rather than simple lifespan, something that Big Pharma will tend never to prioritise, as creating long-term dependency on their products is much more profitable. Such an enterprise should also dramatically improve our defences against the next pandemic and antibiotic resistant infection, other areas Big Pharma inevitably neglects.

We should create a National Care Service, ending bed-blocking in the NHS and saving multiple local authorities from bankruptcy. This should be cheaper to run than the current almost entirely privatised system, because of economies of scale and the elimination of profit taking, plus reduced property related costs, as well as improving standards of care.

We would restore community recreational facilities and per capita funding for youth services and mental health, while, with an actual vision of a liveable future backed with real investment, perhaps, with some hope restored, the mental health of all us can improve, reducing long term sickness, welfare costs and pressure on the NHS.

All of this requires the right skill gaps to be plugged, which brings us onto another economic fundamental: to create new services providing vocational training, both to young people and adults seeking retaining. This would increase the value of one of our most key resources: ourselves.

What we should not do: continue with austerity, scapegoat migrants, abandon our climate change commitments or attempt to deal with the temporary problem of heightened elder-care by having more babies.

After 15 years it is totally obvious that austerity hasn’t worked: in fact it has crippled our public services and scarred our economy.

We could blame migrants, except for the fact that, according to the experts in the field such as the Oxford Migration Observatory, the overall long term fiscal impact of migration is most likely roughly zero (migrationobservatory.ox.ac.uk/resources/briefings/the-fiscal-impact-of-immigration-in-the-uk). Whenever we increase the population we increase the demand for everything our economy provides, from houses to medical services, but it’s often forgotten, or deliberately overlooked, that we also increase supply: the number of workers, say builders and doctors, and the number of tax-payers. Given that migrants tend to be young adults, in the short term, before they have their own kids and themselves grow old, their economic impact is likely to be positive. If we are to reconstruct our economy, migration, along with retraining those of us already here, it will also be essential to quickly plug skill gaps.

We could abandon all efforts to decarbonise our economy. There are a few issues with that one. Firstly, it may not save any money: for example with wind power now the cheapest way to generate energy in the UK. Secondly, we fail to build the industries of the future, so becoming uncompetitive on the global stage in the longer term. Thirdly, we betray the rest of humanity and make civilisation destroying climate catastrophe much more likely. However small our emissions are now compared to, say, China’s, as the first nation to industrialise we have used up, relative to the size of our population, a disproportionate amount of humanity’s carbon budget. We are also still, despite our difficulties, one of the richer nations on Earth. If we fail to keep our carbon commitments, why should the rest of the world even attempt to do so? If everyone does the same, and “defects” (we are back to coordination problems here) then we are quite simply doomed. The problems we are already experiencing, storm damaged buildings, floods, wildfires, heatwaves and water-shortages, then compound with massive movements of people, as some regions of the planet become uninhabitable, with harvest failures, lethal heat-waves and mega-droughts, leading to global unrest and escalating warfare that sooner or later turns nuclear. All of which would be rather apocalyptic for economic growth, as well as everything else.

We could have more babies, to improve the ratio of younger people to older people. Again there are issues here, not least of which is who it is who actually has to have the babies?Then there’s the environmental cost of more humans in a relatively high consuming country. Finally, as a “solution” this actually adds to our “needs” for a couple of decades: the “babies” don’t turn into productive adults for many years. So, far from being a solution, this actually adds to our challenges. If, instead, stopping fertility rates falling further is the concern, it should be noted that people may choose not to have children when struggling to keep a roof over their heads, or when they have no hope for the future.

How do we raise the money to “bootstrap” our economy back into responsible, moderate growth?

In the end, the measures above pay for themselves by creating further and fair economic growth, but where do we get the initial “seed capital” from to fund those measures in the first place? How can we, as it were, “pull ourselves up by our own boot-laces”.

Firstly, we can no longer simply create the money, like we did with Quantitative Easing (QE) when the Bank of England created money and used it overwhelmingly to buy back government debt. Originally, QE was meant as a time limited intervention just to prop up the banking system and keep money flowing after the crash. In the end, to deal with first the Financial Crisis, then COVID, and a bit of Brexit, nearly a trillion pounds was created in this way over more than a decade, with QE being used to directly finance government spending, as government debt was immediately bought back after issue. Unfortunately, though, the current situation is inflationary, where-as those crises were deflationary. Right now, at a time where we are still significantly missing targets on inflation, increasing the money supply in this way would risk undermining confidence in sterling and fuelling yet more inflation.

However, recently the Bank of England hasn’t been doing QE, but it’s opposite, Quantitative Tightening (QT). Now that means selling the government debt it owns back to the private sector, in other words effectively borrowing the money again, as then the public sector no longer owns its owns debts, and cancelling out the money so received.

Governments like to pretend that isn’t really increasing the debts, because the debts bought back by the Bank of England were retained within the total of government debt. Arguably, QE is best thought of as “deferred borrowing”.

At the end of the day, though, it is likely that there was so much “deferred borrowing” that the consequences of turning all that currently Bank of England owned debt back into real debt will be disastrous, and we’d be better of keeping that debt deferred infinitely, perhaps forever, and not cancelling out all that money. After all, it isn’t as if the crises ever stopped happening, and, as we said above, unless inflation is again replaced by potential deflation, re-creating the funds destroyed by doing further QE can’t happen because it would be too risky.

Sadly, right now the Bank of England is adhering to a target of turning £100 billion of deferred borrowing into real borrowing each year. By June 2025, the government debts that were still owned by the Bank of England were down to 590 £ billion. (https://www.bankofengland.co.uk/asset-purchase-facility/2025/2025-q2)

This means a few things that have extremely deleterious effects on the public finances and tend to lock in austerity. The most obvious is that, once the debts are returned to the markets, we have to pay interest on those debts. The second is that, as the real debt piles up, the government becomes less credit-worthy, so less able to borrow to invest, rather than simply borrow to cancel out QE. That, and the simple amount of debt on sale, pushes up the interest rates the government has to pay, so not only do we have to start to pay interest on debts that were previously effectively interest free because they were held at the Bank of England, but we also have to pay a higher interest rate on any new borrowing. (Note that new borrowing also includes what the Bank of England needs to raise to pay off the principal of government debts it owns that mature, with the money borrowed then again simply being cancelled out). Finally, because the interest on government debts, or “gilts” as they are known, is usually fixed, the higher the interest rates, the lower the re-sale of the gilts. Therefore the Bank of England is also making significant losses when it sells these government debts, bought by QE, back to the markets. These losses are already in the tens of billions, and the Office of Budgetary Responsibility estimating that the losses associated with QE and QT will eventually total roughly £63 billion (https://obr.uk/box/the-lifetime-impact-of-quantitative-easing-and-quantitative-tightening).

This is a crazy, dogmatic policy that starts to look like a deliberate attempt to undermine the public sector and force “permanent austerity”, on the part of a Bank of England far too close to our financial institutions and the richest in society. Nor is the Bank of England doing this to fight inflation, as it says of its own QT programme that “the impact on activity and inflation is also likely to have been small” (Quantitative Tightening: House of Commons Committee Report publications.parliament.uk/pa/cm5804/cmselect/cmtreasy/219/report.html)

We should stop QT immediately, if it isn’t already too late. Creating another approximately £0.5 trillion of real public debt just to cancel out QE is cruel and irresponsible.

The other thing we can do is to reduce the interest we pay on the commercial bank reserves at the Bank of England, by tiering the interest rates in the way already done by the European Central Bank (ECB). This measure can be linked to stopping QT, because one of the arguments made for doing QT, and reversing QE, is that all the money created by QE really sits in these reserves. These reserves are actually the ultimate “electronic money” backing all other money in our economy, as it is these reserves that the commercial banks use to “settle” with each other at the end of each day of banking. Reversing QE therefore reduces the size of the reserves. That can have some risks, like increasing the chance of a run on the banks, but it does mean the Bank of England has to pay less interest on the reserves. That means that provided the interest we now have to pay on gilts sold back to the markets is lower than the “base rate” we pay on those commercial bank reserves, and that is often the case as the interest on many of those gilts was fixed at a time when interest rates were lower, then we can actually reduce our interest payments.

That argument loses its cogency when we realize that we don’t have to pay interest on those reserves to the commercial banks in the first place. Arguably, we could even be justified in charging the banks for the use of this facility, just like commercial banks now charge customers for some current accounts. The Bank of England only started paying interest in reserves in 2006, and does so for two reasons: to increase the stability of the banking system by encouraging each bank to hold at least a minimum amount of reserves, and to influence interest rates charged generally, by setting a floor for rates. Both those objectives could be achieved, though, by paying zero or a lower rate of interest on a base level of reserves, and a higher rate of interest on any reserves above that base, i.e. tiering the interest rates in a similar way to already done by, for example, the ECB and the Bank of Japan. According to the Institute for Fiscal Studies in 2022, tiering the interest rates would have have saved “between around £30 billion and £45 billion over each of the next two financial years ” (ifs.org.uk/publications/quantitative-easing-monetary-policy-implementation-and-public-finances).

Again, we can only deduce that the Bank of England is far too close to the City of London.

According to the same Treasury breakdown of spending we used above, since 2008 interest payments on the public debt have been the faster growing component of the budget, increasing, in 2023 prices, from 70.24 £ billion in 2008/9 to 126.6 £ billion in 2023/4, or by 80.25%. By contrast, health spending increased by 42.48% and social protection spending, which includes the bulk of welfare, by 25.09%. This reflects increasing public sector debt as a result of a succession of crises and, arguably and ironically, austerity itself. The state of the public finances is another reason why we so often ask “where has the money gone”, even though, as a society as a whole, we are, as we have seen, arguably richer than in 2008, and even though the bulk of UK public debt is held domestically by institutions such as pension funds, that is where it is not still actually owned by the Bank of England, so broadly we kind of owe the money to ourselves. We have to bring future interest payments down by tiering the interest we pay on the reserves, stopping QT and ensuring that when we do decide to borrow it isn’t to cancel out QE, but to invest, so should generate a return greater than any future interest payments. (To find out more about QT and QE go here: gezwinstanley.wordpress.com/2024/10/28/infrequently-asked-questions-about-quantitative-tightening-that-you-were-never-even-meant-to-ask ).

Then there is the obvious question of a wealth tax. As we have already noted, there has been a significant growth in wealth inequality in the UK since the Financial Crash. A lot of that increase in wealth inequality is down to increases in valuations of certain assets, such as property and shares, itself partly due to the money pumped into the economy by Quantitative Easing, so really something of a “windfall”. Now, wealth as a quantity is a “level” not a “flow”, so we can’t repeatedly tax wealth without, eventually, running out of wealth to tax. Of course, wealth begets wealth, so we could counter here that we can in fact tax wealth repeatedly provided we do so at less than the rate of real growth of wealth. The complication here is that growth in wealth, as we have seen, takes two forms: profit, which is a flow not a level so therefore can be taxed repeatedly, as income, and increases in how assets are valued. Increases in valuations can’t be taxed repeatedly because the the only way to realize that gain is to sell assets, and the underlying assets aren’t actually increasing: therefore, again, if we repeatedly taxed those assets we would run out of assets to tax.

It could be argued, though, that the point of a wealth tax isn’t just to fund the repair of public services and the economy generally, but to directly reduce extreme wealth inequality as a goal in its own right. To that it should be countered that there are less crude ways to do that than a wealth tax, which simply attempts to siphon off the value of assets above a certain level continually, and that would be to share the ownership and control of the mechanisms that generate wealth, for example, through democratically managed public ownership and workers’ cooperatives. It is simpler, and less illiberal, to ensure that wealth is fairly distributed as it is created, to do “pre-distribution”, rather than allow wealth to become ever more concentrated in the hands of the richest and then repeatedly have to attempt to prise it out of their politically well-connected and media controlling hands using “redistribution”.

Right now, though, a one-off wealth tax seems like the most sensible way to raise money to invest (for example, as proposed by the, in part, LSE and Warwick University funded UK wealth tax commission: www.ukwealth.tax). Note that the sums raised will have to be invested, to improve how effectively our economy can meet our needs in the long term, and must not be used on day to day spending. If we ever did that we would just fritter the money away, not solve our underlying problems, and then be back in exactly the same mess again tomorrow.

To meet our day to day needs, and to keep a lid on inequality, and possibly raise some more money to invest, we would align taxes on “passive income”, the income generated by wealth that we touched upon above, with those on income generated by work: for example making the tax rates on capital gains the same as those for income tax. We would make our tax system less regressive (meaning poorer people can end paying a higher proportion of tax than richer people) by increasing the lower rate of national insurance paid on income above £967 a week to match the higher rate paid on weekly income below that amount (something most people don’t even realize exists). Richard Murphy, the Professor Emeritus of Accounting Practice at the Univeristy of Sheffiled and a tax campaigner, has a range of such proposals here taxingwealth.uk. In 2024, Professor Murphy estimated that the national insurance change alone could raise £12.5 billion a year.

Though indefinite taxation of wealth as a stock is not sustainable, there is a case for taxing use of uncreated wealth. Anyone who is exclusively exploiting natural resources, such as the land, should compensate everyone else in the community for the deprivation of the same opportunity, really as a basic upshot of the moral requirement to respect the wants and needs of others as equal to our own. This applies even if the rest of the community could not currently exploit the resource in question because they lack access to the necessary investment or specific technologies, as to rule otherwise would then simply compound pre-existing inequalities. Therefore, the users of uncreated wealth should pay a form of rent to everyone else. This means that initiatives such as a land value tax should be seriously considered, Such a tax would also allow the community to better ensure that its uncreated wealth is used in a way that is generally beneficial, by adjusting the tax to be paid appropriately, potentially even turning the tax negative to subsidise food security or environmental restoration.

We could use a land value tax, or possibly a local income tax, to raise more money than the currently regressive council tax. Council tax bands scale far less rapidly than property prices, so council tax tends to bear down heavier on the poorer. Similarly, a local corporation tax could be considered to reduce the business rate burden on smaller companies.

Ideally we could raise more, and do so more progressively, by shifting taxes to the companies making the highest profits. To facilitate that, we could work more closely with like-minded nations to create economic blocs that stop the race to the bottom in taxes such as corporation tax, and force transnational corporatisations to pay more tax. This is another argument in favour of cooperating more with overseas partners such as the EU, on top of those relating to undoing some of the damage done to our trade by Brexit. As all of this requires regulatory alignment with the EU, and we are better off helping to make those regulations rather than being a mere rule-taker, we should seriously consider negotiating a deal to rejoin in some form, and working again from the inside.

So where did all the money go?

If we take that question to mean where did all the economic growth go, compared to what was happening prior to 2008, so instead we got very slow growth instead, well that was down to de-industrialisation coupled with over-reliance on finance capital, austerity, rising inequality, Brexit and the after-effects of COVID and Russia’s invasion of the Ukraine, especially our over-dependence on global fossil fuel markets.

If we take that to mean resources and wealth, then the “money” hasn’t gone anywhere, with the total value of real resources most likely growing in per capita terms. Wealth, though, denotes who can “own” those resources and here wealth has gone to the already richest.

If we consider income, then overall income per head has increased and income equality has actually improved slightly, but if we look at purchasing power then inequality has increased significantly when we compare the richest to the poorest.

If we take “where did all the money go” as a lament about our public finances, then the issue is the richest lobbying and conniving not to pay more taxes at a time when repeated crises and increasing underlying need has put pressure on public spending, not beyond our capacity as a nation to support but beyond the willingness of the richest to pay.

Perhaps the problem with low growth is not so much the low growth in itself but the way that the super-rich, desperate to still out-grow each other, then become even meaner and more ruthless: ever less willing to pay taxes, ever keener to shrink the public sector and ever more determinedly promoting the idea that the only way to restore growth is for everyone else to let their greed run amok. Basically, when growth is negligible it is no longer possible for the super-rich to hide increasing inequality, where they can take a larger, and even possibly growing, share of the expanding pie but most people don’t notice, because their share is getting a little bit bigger too. That means they have to resort to other distraction tactics, such as scapegoating the sick, the poor, the foreign and the culturally non-conformist, all the while attempting to divert the anger caused partly by their own actions in a way that ironically boosts their wealth and power still further. There’s even a new term for that: “pluto-populism”. The upshot is a threat to democracy itself.

Conclusions

The UK is not broke. The UK is not going broke.

Per capita economic growth has happened, but slowly. The main issue with our resources and wealth is not currently the depletion of either but increasingly unequal distribution.

In future, our increasing needs, per head of the population, could start to outrun such a slow rate of economic growth, and we do need to buffer ourselves from the inevitable crises to come, so driving some modest, responsible and fair per capita economic growth would be sensible, especially in the health sector, to better meet basic needs, such as housing, and to protect our security of supply when it comes to energy, food and some areas of industry.

Finally, we should reject, not only the lie that the UK is broke or is going broke, but the very framing that, were the situation to arise, what we should sacrifice first are the needs of the poorest and most desperate, that the very first things we should cut are publicly funded healthcare or the welfare safety net. As we have seen, morally that is abhorrent. Those are the very last needs we should abandon. When we next encounter genuine shortages, the moral thing to do is to triage and ration, just as we did in World War Two when the Nazis were attempting to starve us by sinking our Atlantic convoys. It’s a very uncertain world, and we should never accept the notion that we solve a resource crisis by letting the poorest perish and freeing up the greed of the rich to get yet richer.

***

Methodology and Sources

Note 1:

Office of National Statistics (ONS) Capital stocks and fixed capital consumption, UK: https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/bulletins/capitalstocksconsumptionoffixedcapital/2024

Office of National Statistics (ONS): Weekly Earnings (defined as “wages and salaries”): https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/methodologies/averageweeklyearningsqmi

GDP deflators used to convert money values into 2023 prices unless others stated: UK government Statistics: https://www.gov.uk/government/statistics/gdp-deflators-at-market-prices-and-money-gdp-june-2025-quarterly-national-accounts

Office of National Statisitcs, Population Figures: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland

UK Government Statistics: Inactivity Rates: https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/timeseries/i46d/lms

Note 2:

Wealth figures: World Inequality Database (WID): https://wid.world/country/united-kingdom/

Note that the WID is the product of the World Inequality Lab, which “is composed of about forty people (co-directors, coordinators, research fellows, research assistants, staff). It is based primarily at the Paris School of Economics and the University of Berkeley, California. The World Inequality Lab works in close coordination with more than 200 researchers – the WID Fellows – based in institutions around the world”.

WID accessed in September 2025, at which point all GBP figures were given in 2023 price terms.

Note 3:

GDP figures were sourced from here: https://www.ons.gov.uk/economy/grossdomesticproductgdp/timeseries/ybha/pn2

GDP deflators used were as cited above.

Per capita income was taken from https://wid.world/country/united-kingdom/, multiplied by the ONS population figures, cited above, to derive UK income totals for 2008 and 2023, with that total income then being apportioned to the top 1% or bottom 50% of the population according to the income shares also sourced from https://wid.world/country/united-kingdom/. Then those top 1% and bottom 50% income shares were then divided by 1% or 50% of the relevant UK population totals to derive the per capita incomes cited. Note that, as the per capita income figures given in the World Inequality Database are for all people aged 20 or over, the UK population totals used in these calculations were all consistently those for individuals aged 20 or over.

https://wid.world/country/united-kingdom/ was as accessed in September 2025, at which point all GBP figures were given in 2023 price terms.

Note 4:

The HM Treasury breakdowns of public spending are from https://assets.publishing.service.gov.uk/media/6874fa6f92691289bdb7d393/Public_Expenditure_Statistical_Analyses_2025.pdf, table 4.2

The 2008 personal social services spend was taken from table 5.2 of Public expenditure statistical analyses 2009 – GOV.UK .

The population and GDP deflator figures used were as cited above.

Source of average long term figure for annual growth in NHS spending: https://pmc.ncbi.nlm.nih.gov/articles/PMC6616184/

Note 5:

ONS Price Index of Private Rents: https://www.ons.gov.uk/economy/inflationandpriceindices/datasets/priceindexofprivaterentsukmonthlypricestatistics

ONS Household Costs Indices (HCI) for UK household groups: https://www.ons.gov.uk/economy/inflationandpriceindices/datasets/householdcostsindicesforukhouseholdgroupsreferencetables

A weekly income figure by decile were taken from here: households with below average income, for financial years ending 1995 to 2024, at https://www.gov.uk/government/statistics/households-below-average-income-for-financial-years-ending-1995-to-2024, and multiplied by 52 to derive annual figures. The specific table used was “2.1ts Decile” from income-values-and-inequality-measures-hbai-1994-95-2023-24-tables.ods.

Infrequently asked questions about Quantitative Tightening that you were never even meant to ask

Posted on October 28, 2024 by gezwinstanley

Worried about public sector debt levels? Worried about never-ending austerity? Worried about lack of investment and economic stagnation? Worried about the lack of finance to deal with the climate emergency? Then let’s stop doing this….

Or infrequently asked questions about Quantitative Tightening that you were never even meant to ask!

What is Quantitative Tightening (usually abbreviated as QT)?

QT refers to the sale of debts a central bank has previously purchased back to the financial markets, with the money then received being cancelled out (that’s the “tightening” bit). More usually than not, and overwhelmingly in the case of the UK over time, and currently totally in the UK as all other types of debt have already been sold, those debts are those of a central bank’s own government, in other words government “bonds” (also, in the case of the UK, called “gilts”). Selling public sector debt “back to the markets” therefore amounts to the public sector re-borrowing the money.

Why were central banks buying debts in the first place?

Central banks bought debts, and, as we have seen, usually the public sector’s own debt, in order to do “Quantitative Easing”.

What is “Quantitative Easing” (usually abbreviated as QE)?

QE is the process where a central bank creates money and uses it to buy up debts, usually the central bank’s government’s own debts. QT is the reverse of QE, and that’s why the money received when debts are sold back to the markets is cancelled out, as that destroys the money created by QE.

What is the point of QE?

The point of QE is meant to be to avoid economic contraction, by increasing the supply of money, attempting to reduce interest rates and driving more lending. Part of this is about avoiding deflation. Deflation means falling prices, and is more feared by economists than inflation, because deflation can send an economy into a downward “deflationary spiral”, when people will attempt not to buy today what they expect to be able to buy more cheaply tomorrow.

We have also seen QE used for other purposes though: to prevent a collapse in the value of government bonds by buying bonds up (during the Truss Crash, in order to prevent pension funds going bust) and, even more importantly, to finance government spending during various crises from the Financial Crash onwards.

Was QE really used to finance spending?

This is controversial, as central banks don’t like admitting to it, but yes., with, for example, the Financial Times noting in 2020 that the the Bank of England’s asset purchases perfectly tracked central government’s borrowing needs.

Government vastly exceeded what would have been its normal capacity to borrow money, because it wasn’t really borrowing it at all: it was creating it.

Though it was never actually described as such at the time, this means that QE took on the nature of deferred borrowing, leaving us naturally to wonder how over a decade’s worth of crisis spending, from the Financial Crash to COVID, will impact our economy once that money is genuinely borrowed, rather than those debts being owned by the public sector itself.

Why was QE done in such a contrived way?

It is natural to wonder why central banks didn’t purely just create money, and bothered with this contrivance of buying back debt at all. However, a central bank creating money and say, lending it to its government, direct financing as it is called, is seriously frowned upon: it is even illegal in the EU under Article 123 of the Lisbon Treaty. The reasons for this, in turn, are concerns over creating inflationary expectations.

Debt therefore had to be bought back to obscure any appearance of direct financing. Also, using the created money to buy debt means that QE comes inbuilt with the means of its own reversal. The purchased debts, now held as assets at the central bank that did QE, can just be sold back to the financial markets, in the process that has come to be known as Quantitative Tightening, with the proceeds from the sale of the debt simply being cancelled out, destroying the original QE created money.

In the meantime, the public sector debt owned by the central banks, which are part of the public sector, is still recorded in measures of public sector debt, even though the public sector now owes this money to itself.

It could be argued that this obscures the reality of the debt. Debt that an entity owes to itself effectively has no economic significance. Debt owed to third parties, however, very much does, in that it could affect the credit-worthiness of the entity in question, meaning its capacity to borrow more and the cost of any future borrowing. Keeping QE purchased debts in the totals of debt means that the significant act of selling that debt back to the markets, in reality re-borrowing the money, does not change those totals, dangerously obscuring the economic reality of what is going on.

How can central banks “create money” in the first place? Isn’t creating money just trying to get something from nothing?

The idea of central banks creating money can seem offensive to common sense. It feels like getting something for nothing.

There are two points to make here. Firstly, the central banks themselves do not deny that the money used for QE was created. Secondly, this is what governments do all the time: consider, money has to be created and spent into an economy first before any government can tax it back.

More fundamentally, though, it is important to be clear about the real nature of money. Money is not in itself a resource. It’s when we think of money as a resource that the whole notion of money creation feels so bizarre, because that would then amount to creating a resource out of thin air. Rather than being a resource, money is really a transferable claim on resources, That claim is backed by a government mandating that the money it issues has to be accepted to buy goods and services in the economy under its control. To confuse money with resources is actually an example of an error in thinking called “reification fallacy”, when we conflate the representation or abstraction of something with the concrete thing itself.

Once we think about money as a claim on resources it’s clear that nothing is really being created here so much as redistributed. If the value of the total supply of money in circulation is the value of all the goods and services it could purchase, then what money creation really does is redistribute purchasing power. When a government creates some money it adds to its own purchasing power, but by reducing everyone else’s a little bit, because each unit of money is now worth slightly less. The governments enhanced purchasing power then spreads through the economy as it spends its new money.

How much QE and QT has been done in the UK?

Since and including the Financial Crash, then a little bit of Brexit, and finally COVID, the Bank of England carried out £875 billion of QE. This was a scale of QE far greater and persisting for far longer than was originally envisaged, when QE was implemented, as explained above, just as a way to keep money flowing in the economy in the wake of the Crash. Even by 2012, in the words of the Bank of England: “the scale and duration of QE was far in excess of that originally envisaged.”

The Bank of England started QT in February 2022, initially just not purchasing new bonds to replace those that had matured (passive QT), but then also starting to sell-off its own bond holdings (active QT) in November 2022, re-borrowing and then cancelling out £118 billion up until January 2024, with a target of doing £100 billion worth of QT each year from then until, presumably, the QE created money is eliminated.

Isn’t this all just conventional “financial engineering”, really rather technical and nothing to worry about?

No: QE is a relatively recent invention and QT even more so. The first QE was initiated by the Bank of Japan in March 2001. Back in 2009 QE was still regarded by the Bank of England as an “unconventional measure” We are still working out the full implications of unexpectedly large-scale QE programmes, meaning that the effects of any extensive QT are even less well understood, as QT itself has only begun recently. In the words of the Bank of England itself, as of February 2024, QT is “ innovative for all central banks”.

Other countries are also doing QT: basically every country that has done QE with the odd and ironic exception of Japan itself. Note, though, that the US central bank, the Federal Reserve, is planning to leave it’s own “balance sheet”, basically the debts it owns, dominated by US government bonds, which are called Treasury Securities, indefinitely. In the Fed’s own words, it wants to “achieve in the longer run a portfolio that consists primarily of Treasury securities”. That means, it isn’t planning to fully unwind its QE.

The European Central Bank (ECB) is still doing QT but before it started faced a revolt of more than a hundred economists, politicians and campaigners from across Europe, including Thomas Piketty, arguing that the EU debts then owned by the ECB should effectively be forgiven, either cancelled or converted into a zero interest never-maturing debt. The ECB’s response was essentially that this would violate Article 123 of the Lisbon Treaty.

What is the scale of QE/QT?

Though it doesn’t match exactly, as it isn’t the only cause of variation in the size of the central bank balance sheet, QE/QT has been the main driver for change to the balance sheets for all countries doing QE and QT. Indeed, sometimes QE or QT has exceeded the size of the change in the balance sheet, but been somewhat offset by changes in the opposite direction. Looking at those balance sheets, does however, allow us to perform an approximate comparison of QE and QT internationally. (Note that here we mean a balance sheet in accounting terms, where the rules of double-entry book-keeping mean that every entry for an “asset” is balanced by an entry of a “liability”, so that the balance sheet should balance at zero. A change in the size of the balance sheet is therefore formally a change in both directions: growing or shrinking assets and liabilities. Also note that loans, including government bonds, bought by a central bank are counted as assets because they represent money owed to the bank).

So, in the UK the Bank of England balance sheet grew from £93.2 billion in August 2008, and peaked at £1.12 trillion (1 120 billion) in January 2022, but is now down to £885 billion in October 2024, meaning that it has shrunk from its peak value by 20.1%. To get an idea of the scale of these figures the UK GDP in 2023 was £2.27 trillion.

For the European Central Bank. the balance sheet grew from €1.45 trillion in August 2008 to €8.83 trillion in June 2022, and then shrunk to €6.43 trillion by October 2024, so by 27.1% from its peak.

The size of the Federal Reserve balance sheet starts at $910 billion in August 2008, peaks at $8.95 trillion in March 2022 and shrinks down to $7.04 trillion in October 2024, so 21.1% since its peak.

Major Western governments have basically financed their way through over a decade of crisis by trillions in money creation.

Will the Bank of England make a profit when it sells the bonds back?

No. Government bonds normally pay a fixed amount of interest each year: say £2 for a £100 gilt. When interest rates generally rise this therefore makes the resale value of bonds fall: for example, if interest rates are now 4%, it would only then make sense to buy that bond that pays £2 a year for £50 rather than £100. That’s because, as £2 in £50 is 4%, that means that the bond is again “yielding” the “going rate” of interest, which also illustrates the reason why, as bond resale values fall, what are called “bond yields” rise, and why rising bond yields are bad news for institutions already holding bonds. Now, the Bank of England bought bonds when interest rates were near zero, so bond prices were high, but is selling them back to the markets after interest rates have risen, so the resale value of those bonds has fallen. In February 2024, regarding these bond sales, the Bank of England noted that “annual losses in each of 2023 and 2024 will amount to £40 billion”.

What are the justifications for QT?

Multiple, i.e. differing, reasons for doing QT have been given over the past three years or so, which is itself questionable. All of these are Bank of England cited ones:

a) To reduce the interest the Government has to pay. Basically, all the QE created money ended up in the reserve accounts held by the commercial banks at the Bank of England. This is the “ultimate” form of money in the economy, essentially the electronic equivalent of notes and coins, and it is the way the banks clear their debts with each other at the end of each banking day. From 2006 on, the Bank of England has been paying interest on all that money to the commercial banks, who have therefore repaid being saved by QE in the first place by being allowed to make hearty excess profits on these increased reserves. Nevertheless, initially, when the interest rates in the broader economy were lower than the fixed rates on the bonds, paying interest on the reserves rather than the bonds saved the Government money. Now, however, with interest rates increasing, this position has reversed, and selling back the bonds, receiving some of the money from these reserves in payment and then destroying it (i.e. QT) can reduce those payments. However, there is an alternative, as interest is only being paid on the reserves in the first place to encourage the banks to keep enough reserves to reduce risk of their failure, and as a way to set a floor for more general interest rates. Both of these objectives could still be achieved, while reducing those payments, if the interest rates on those reserve accounts were tiered, as is already done by European Central Bank and the Bank of Japan, and has already been suggested by various UK pundits and think tanks. So, this is arguably a non reason.

b) To curb inflation (QT shrinks the money supply directly by cancelling out reserves, while also, by increasing demand for borrowing, possibly pushing up interest rates). There are two counters to this. QE stopped years ago now, but had been going on and off since the Financial Crash. Generally, it didn’t cause net inflation because it offset deflation, and where it did those effects have already happened. As general deflation wouldn’t be allowed to happen, the best QT could therefore now offer is to reduce current inflation, the rate at which prices rise, a small amount. Here, though, the Bank of England itself regards the effects of QT on inflation and on increasing interest rates as minor. The Bank of England regards its setting of the base rate as its main weapon against inflation, not QT. In August 2023, reviewing QT to date, the Bank of England said this: “the impact on activity and inflation is also likely to have been small”,

Also note that attempting to mitigate inflation caused by supply side factors by reducing demand, which is how QT would have an effect, can do nothing to address “cost of living” pressures. Consider: would anyone think the cost of living problems for their own household had been improved, even if prices were rising more slowly, if that fall in inflation was also associated with an increase in their mortgage payments or less government spending, so that, for example, they have to spend thousands because they can’t find an NHS dentist?

(c) To allow more scope for QE in future. This reason is nonsense. QE, creating money, can only be done in a situation where there is potential deflation, as otherwise it could be inflationary. Economists fear deflation more than inflation. Deflation can stop an economy dead: because few will buy today anything that they could possibly buy cheaper tomorrow. In such a situation no one would care how much QE a nation had done: instead the demand would be to let the (electronic) printing presses roll, as it were, and get on with creating new money.

A more fundamental reason, but one that is rarely discussed explicitly, perhaps for fear that it could invite the very thing it is concerned about, is to maintain faith in a currency, or, in other words, to avoid creating “inflationary expectations”. The clearest statement of this comes from a report commissioned by the European Parliament: “Since the beginning, the ECB has been uncomfortable with QE (which led to a delayed start compared to other central banks) because it sees it as blurring the separation between monetary and fiscal policies which could constitute a threat to its independence. Launching QT would then be a way to reduce the risk of fiscal dominance and reaffirm monetary dominance in the euro area. It would confirm that QE was not a permanent monetary financing of deficits (as it was sometimes perceived). This would also help ensure that inflation expectations remain well anchored, especially at a time when inflation is significantly above target.”

Is QT necessary to stop inflation as economies recover and confidence returns?

QE did not cause general inflation when it was carried out, most likely because, as intended, it offset deflation that would have happened otherwise. However, could QE, money creation in other words, cause delayed inflation that would only occur as economies re-open? Delayed inflation that can only be dampened down by doing QT?

After all, if money is just a claim on resources then each unit of a state-issued currency is just a share of the value of the sum of all the resources that can be bought using that currency as legal tender. An economic crisis might reduce the money supply, as people try and keep their money somewhere safe rather than spending or investing. Once the crisis ends though, all that money could return to circulation, resulting, if the QE created money isn’t cancelled out, to an increased money supply relative to total purchasable resources, and therefore inflation.

This model gets two things wrong. Firstly, it doesn’t grasp that, even during a period of reduced economic activity, perhaps as measured by something like GDP, which is really tracking the rate at which goods and services are bought and sold and not the level of total “wealth”, the total stock of purchasable resources could still increase, just most likely more slowly than in a period of prosperity. Even more importantly, though, the model doesn’t reflect the actual nature of the money supply. Most money in circulation is created by commercial banks lending out more money than they hold. That money circulates for a while and is then cancelled out again when the loans are repaid: let’s call it “virtual money”.

Now the level of virtual money in circulation will depend on the rate at which it is created as loans versus the rate at which those loans are repaid. An economic recovery is likely to increase both those rates. However, the amount of money that can be created as loans is is primarily influenced by the demand for loans, which itself tends to fall as borrowing becomes more expensive, i.e. as interest rates rise. Therefore, central banks’, including the Bank of England’s, main tool for managing the money supply is setting a base interest rate, the “bank rate”. Does that mean that base interest rate would have had to be set higher without QT? That seems unlikely, as QT itself increases the demand for loans and therefore actually raises the cost of borrowing.

It is therefore unsurprising that, as stated above, the Bank of England regards the bank rate as its primary tool for managing the money supply and says that the impact of QT on inflation is “is also likely to have been small “.

What are the downsides of QT?

QT means reducing commercial bank reserves, so increases risk of bank failure.

QT causes bond prices to fall, so can threaten institutions that hold lots of bonds, such as pension funds.

Together the above two effects amount to a heightened risk of a new financial crash, as arguably nearly happened with the three bank failures in the USA in March 2023 (Silicon Valley Bank, Silvergate Bank and Signature Bank). In the UK, the Bank of England announcing £40 billion worth of QT also contributed to the “Truss Crash” of September 2022: in the end the Bank had to change course completely and initiate £65 billion worth of new, if short term, QE, to support bond prices and save many UK pension funds. These are therefore very much not purely theoretical risks.

By increasing demand for borrowing, QT can push up interest rates.

Very likely, though there’s a bigger risk: continued austerity and constrained public sector investment, as what the government can borrow to invest is crowded out by the money being borrowed through QT, i.e. there are too many government bonds on sale. Furthermore, there is a risk that, as notional debt is turned back to real debt, the government is seen as less credit worthy in the longer term, increasing its costs of borrowing and constraining its borrowing possibly for generations, as roughly 40% of a year’s worth of GDP is added to its real debt.

What could we do instead?

Arguably, all the government needs to do is to tweak the rules on the Bank of England to prevent it cumulatively expanding or shrinking the size of its balance sheet by more than say £10 billion cumulatively over the course of a year, without having to seek the prior approval of the Chancellor. With-holding that approval would prevent the Bank of England borrowing £100 billion a year just to cancel out the money received, expanding the space for the government to borrow to invest.

We should also consider reducing the interest payments on the reserves held by the commercial banks at the Bank of England by tiering the interest rates payable on those reserve accounts.

What could we do in the longer term?

We could convert the government debt still held by the Bank of England into a zero interest never-maturing debt. Note that this is preferable to simply cancelling out the debts for two reasons. Firstly, the reserves (of QE created money) appear as the liabilities on the central bank balance sheets, balancing out the value of the purchased bonds as assets. Now, these reserve accounts are very odd liabilities, because the reserves are the ultimate form of money, so this “debt” could never technically be settled (the only way to pay to “reduce” the reserve would be to pay into the reserve to increase it, a clear contradiction). Furthermore, a case could be made that it is the commercial banks that should be paying the central bank for services provided rather than vice versa, because without these reserve accounts the commercial banks could not function. Still, it is an unavoidable fact that these reserves are, by international convention, accounted for as liabilities, and that therefore cancelling out the debts owned by the central banks while leaving these reserves in existence would make the central bank’s balance sheet go negative, breaking accounting norms and having unpredictable consequences. Simply “freezing” these debts avoids that.

Secondly, by leaving the debt in existence, as an asset owned by the central bank, it would still be possible, if this policy didn’t work as expected, to reconvert that debt back into bonds: freezing rather than cancelling the debt makes this QT stopping initiative reversible.

Once the debt the public sector owes to itself is frozen, removing the value of the debt from totals of public sector debt would also more accurately reflect the debt position of government than maintaining any fiction that debt owed to oneself is equivalent to debt owed to third parties.

What can we conclude?

The real choice here is to decide between heading off creating potential “inflationary expectations” by being seen to destroy the QE created money, but at the cost of adding roughly 40% of a year’s worth of GDP to real public sector debt, or leaving the QE created money in existence, indefinitely, and not increasing real public sector debt by that huge amount, primarily to maintain government’s capacity to borrow to invest in the longer term.

Considering all the nations who have done QE worldwide the dilemma is similar: avoid possible “inflationary expectations” but at the cost of destroying trillions in potential government purchasing power, at the very time when the democratically accountable, and amply funded, collective actions of governments are needed more than ever to deal with crises such as the climate emergency.

No one can really know for certain which of those risks is greater, so the best thing to do is to follow a course of action that can be fully or partially reversed if the needs arises.

Undoing QE through QT is not easily reversible unless the economic situation becomes clearly deflationary, because of the risks that money creation is inflationary. Once it has been destroyed, any QE created purchasing power is likely to be gone for a very long time. However, provided the public sector debt purchased through QE still remains in existence, in some form held at the central bank, either as a frozen debt that could be converted back to the original bonds or just because QE keeps being rolled over as bonds mature, keeping the size of the balance sheet constant, then the policy of stopping QT is easily reversible. Therefore, stopping QT is the logical course of action.

“Quantitative Tightening” is being done purely to wipe out the “Quantitative Easing” created money. There is no need to destroy that QE created money at the moment, and may never be.

Sources and further reading

Bank of England, the original position on QE: https://www.bankofengland.co.uk/-/media/boe/files/quarterly-bulletin/2009/quantitative-easing.pdf

Bank of England plans for QT, as explained to Parliament: https://publications.parliament.uk/pa/cm5804/cmselect/cmtreasy/219/report.html

For the Fed: https://www.federalreserve.gov/monetarypolicy/fomcminutes20240320.htm

The open letter to the ECB demanding debt forgiveness: https://www.euractiv.com/section/economy-jobs/opinion/cancel-the-public-debt-held-by-the-ecb-and-take-back-control-of-our-destiny

(Unfortunately that has now been pulled behind a paywall, but the core message can be seen here: https://puheenvuoro.uusisuomi.fi/amosahola/on-cancelling-ecb-debt-open-letter-to-100-politicians-economists-and-campaigners/ )

The ECB’s response: https://www.youtube.com/watch?v=Z9nRKRVuAu0

The report to the European Parliament on Quantitative Tightening: https://www.europarl.europa.eu/cmsdata/267598/Final%20Compilation%20topic%201.pdf

Central bank balance sheet figures.

UK: https://tradingeconomics.com/united-kingdom/central-bank-balance-sheet

Eurozone: https://tradingeconomics.com/euro-area/central-bank-balance-sheet#:~:text=Central%20Bank%20Balance%20Sheet%20in%20Euro%20Area%20averaged%204097919.14%20EUR,Million%20in%20February%20of%201999

USA: https://tradingeconomics.com/united-states/central-bank-balance-sheet

Financial Times, on QT as “direct financing” of spending: https://www.ft.com/content/f92b6c67-15ef-460f-8655-e458f2fe2487

One people bitterly divided?

Posted on March 27, 2024 by gezwinstanley

Many people will have seen news images from the West Bank, of Israeli settlers attempting to steal the land of Palestinian farmers. The Israelis, having adopted the same farming lifestyle in the same terrain, are clad in similar clothes to the Palestinians, also speak a “West Semitic” group language that is indistinguishable from Arabic to anyone who speaks neither Arabic or Hebrew, and have even been toned to the identical tan from working outside in the same climate. If they are religious, they will worship the same Abrahamic god.

An outsider could be forgiven for seeing them all as the same people.

Of course, if you have any recent ancestry from anywhere near the Mediterranean Sea, or Europe, then you are Jew-“ish”, including being descended from 1st Century Hebrews living in Palestine at the time of the Great Revolt against the Roman Empire and the ensuing most infamous chapter in the Diaspora. The 1st Century population of North Africa, West Asia and Europe is estimated at about 70 million. 2000 years is roughly 60 human generations, so we each have 2 to the power of 60 possible ancestors back then, or, thinking about it in another way, 2 to the power of 60 “spaces” in that line of our family tree 60 generations ago. Now the chance of any specific individual out of that 70 million occupying a specific space on that family tree, 60 generations ago, is 1 in 70 million, so the chance that any specific individual back then does not occupy a specific space on that family tree is 69 999 999/70 million.

However, remember there are 2 to the power of 60 possible ancestors, or spaces on that tree to be filled, 2000 years ago, and that’s a huge number, vastly greater than the number of people actually alive then, so the chance of any one of those 70 million avoiding every single one of those possible spaces on the family tree is therefore 69 999 999/70 million raised to the power of 2 raised to the power of 60. That number is basically ZERO, and it stays basically as ZERO even if we change these numbers and assumptions a lot.

The upshot is, over very broad areas of the planet, you don’t have to go back far in time to find a point where everyone alive then who has ANY descendants alive today is the ancestor of EVERYONE with any ancestry in that broad area of the planet, and will, in fact, be an ancestor many times over: in other words would appear in multiple spaces in that line of our family tree 60 generations ago, dear cousin.

So Jewish Israelis are indeed descended from 1st Century Hebrews, but also Arabs, Greeks, Romans and every other “ethnic” group then in existence. However, so are Palestinians.

Partly that’s because the Diaspora was never total, and was instead a series of migrations out of the Levant, offset at times by migrations back into the Levant, starting before the Great Revolt and continuing after it. At times, even after the disastrous Jewish revolts of the 1st and 2nd Centuries, and as late at the 7^th Century, the majority of the population of the Levant is estimated to have been Jewish, but that population is estimated to have declined to 5000 at the start of the Ottoman era in the 16^th Century .

In antiquity, Jews were distinguished from Arabs by culture, especially religious and political affiliations, rather than ancestry. Both terms have been applied to certain groups living in the Middle East, including the Levant, for nearly three thousand years: since the 8^th Century BCE in the case of the label “Jew”, most likely originally meaning a member of the kingdom of Judah, and from the 9^th Century BCE for “Arab”, so the “labels” are of roughly the same age. Possibly the best way to think about this is in terms of a number of closely related tribes, speaking similar languages and interacting with each other, but with some forming settled communities on the more fertile land, who over time developed different identities for themselves around culture, religion and politics, including, for example, the Jewish and the Phoenicians, while the “Arabs” initially designated the tribes who had to follow a nomadic life style because they occupied the desert areas, but who also eventually started to found kingdoms of their own, such as Nabataean Petra in the 4^th Century BCE or Iturea in the 2^nd Century BCE. Of course, in terms of Jewish and Islamic tradition, both groups see each other as descended from Abraham, and, regardless of the literal truth of that claim, the ancestral origins of the ancient Jews and Arabs are clearly basically the same. Modern genetic analysis bears out “a remarkable degree of genetic continuity in both Jews and Arabs” (REF 1).

Meanwhile, Diaspora Jews sometimes intermarried with the populations living in the places to which they moved. This could cause issues, such as the notorious mutual intolerance between Sephardi and Cochin Jews in Kerala. Sephardi were the Jews who lived in the region ruled by the various Islamic Caliphates, from Iberia in the north-west, through North Africa, to the Middle East itself, though these days Middle Eastern Jews often adopt a separate identity, such as Mizrahi. There was a further Diaspora of Sephardi people following the completion of the “Reconquest” Iberia by the Christians and the subsequent instigation of the Spanish Inquisition . Sephardi fled first to Portugal, and then, in 1536, when the Inquisition was introduced there too, to North Africa, but also to outposts of the Portuguese Empire, and other places in Europe, especially Holland, where the Dutch Republic had declared freedom of conscience. Kerala, in south east India, was one such Portuguese outpost, but, in the 17^th century, also became a prime destination for the Dutch East India Company, with which many of the Sephardi, who had sought refuge in Holland, also became involved. The result was Sephardi movement into Kerala, where they encountered an already existent Jewish population, the Cochin Jews, who had settled there much earlier, most likely in Roman times. Having been dispersed into different regions for so long the Sephardi were intermixed with other groups, especially Iberians and other southern Europeans, while the Cochin were also extensively Keralite (specifically “Malayali”: the label given to the main ethnic grouping in Kerala). Tragically, whatever joy was felt by some at re-uniting long scattered brethren was soon replaced by xenophobia, followed by a bitter dispute over who was more “authentically Jewish”, the White Jews or the Black Jews, that went on for centuries. (REF 2).

Therefore, “Arabic” Palestinians started off with the same origins as Jews and continued to mix with Middle Eastern Jews even after the conquest of the Levant by Islam and its incorporation into Arab caliphates. Indeed, many “Arabic” Palestinians are descended from “Jewish” Palestinians who converted to Islam, so it isn’t just a question of inter-mixing, but also of members of one group actually becoming members of the “other” group. Genetic studies have found that “The closest genetic neighbours to most Jewish groups were the Palestinians, Israeli Bedouins, and Druze in addition to the Southern Europeans, including Cypriots” (REF 3). Basically, the Palestinians are a population with mixed ancestry, deriving roughly (and possibly mainly) both from Jews who stayed or migrated back to the Levant from the Diaspora and from Arabs who migrated to the area from the Byzantine era onwards. Some of the founders of Israel were proponents of the Jewish ancestry of the Palestinians. The first president of Israel, Yitzhak Ben-Zvi as well as former Prime Minister David Ben-Gurion, wrote several books and articles on the subject (REF 4).

The upshot all of this is that both Jews and Palestinians are descended in part from 1st Century Hebrews, and both populations have clear genetic links to the Middle East. Nevertheless, if we compared randomly chosen Palestinians and Israelis, and were really able to draw up their family trees all the way back to the 1st Century, Hebrews would most likely repeatedly occupy the slots in the 1st Century line of Palestinian family trees more often than in the same line in the family trees of Israelis.

In turn, this means that the Zionist argument of a Jewish right of return to the land of Israel based on ancestry is nonsensical: anyone with West Asian, North African or European ancestry is descended in part from ancient Hebrews, the Jews more than most others but the Palestinians usually more so than most Jews. It is also, by definition, unreasonable to assert a right of return based on faith, as an article of faith is, also by definition, not based on reason. Someone might claim that God had granted a land to themselves and their descendants, but anyone else could equally claim an entirely contrary divine revelation, and there is no rational way for non-believers to choose between either of those articles of faith.

This leaves one viable argument for Zionism. Given the state-sanctioned, and often state-assisted, persecutions to which the Jewish people have been repeatedly subjected, especially but not entirely in Christian Europe, including the Inquisitions, the Pogroms and, most horrifically of all, the Holocaust, it is perfectly reasonable for Jews to seek the protection of a state power of their own, in other words to control as a people their own state which would have, as per Max Weber’s definition of the word “state”, a monopoly over the use of violence within a certain territory. The repeated mass murder and torture of Jews implies that only this would really allow them a home where they could live safely. That means it is desirable to grant anyone who identifies as Jewish, or could be likely to be labelled as Jewish by anyone who might use that as a pretext to do them harm, the possible sanctuary of a state of their own. Another irony here, at a time of cheap and widely available genetic testing, is that the group of people so defined would now be likely to also include most Palestinians.

That argument is credible. However, “n” wrongs do not ever sum to a right. If Palestine was really a “land without a people, for a people without a land” all would have been fine, but it was not: the land was already the home of the Diaspora Jews’ cousins, and expulsion of those people, who up until then had generally lived peacefully with Jews and had not been the cause of their persecutions, was not morally justified. The situation was further exacerbated when the Arab states responded to the expulsion of the Palestinians, the “Nakba” by expelling their own Jewish populations, who were then encouraged to migrate to Israel, creating the tragically ironic situation where the resentment of the Palestinians is now often most extreme amongst Sephardi and Mizrahi Jews, as compared to the rest of the Israeli population, even though those are the very Jewish groups most closely related to the Palestinians.

Now a new state of Israel exists, and a way must be found to allow the Jews to retain a homeland while also addressing the plight of the displaced Palestinians, through some combination of restoration of lands, rights and compensation, whether that is a Two State solution, or a single State solution with strong and equal protections for all, or some federal system combining both approaches. That is a complex issue not only requiring the advice of experts but also extensive consultation amongst the Israelis and the Palestinians, the people most affected.

Ancestry, as we have seen, is nearly inextricably complex and inter-mingled, and identities are fluid, usually a matter of habit, often one of convenience. Of course, any of us who believe in any vaguely Golden Rule style morality, religious or secular, are obliged to try to treat each other with kindness, or at least not be wilfully selfish. Identity shouldn’t matter. It shouldn’t matter whether others are of “our” group or not: our moral obligations extend to at least anyone who, given the chance, could return the favour. Such moral obligations could even extend to non-human entities. Nevertheless, feelings of identity are significant. Crimes against humanity are typically preceded with dehumanising the “other”. Therefore, remembering what we have in common is important.

At the time of writing, what the International Criminal Court has deemed the “plausible genocide” in Gaza is continuing to unfold. Over 32 000 Gazans have been killed as a result of direct Israeli actions: tens of thousands more will be dead under the rubble or have perished from indirect effects. At least 12 000 children have been killed: higher than the number killed globally in all conflicts from 2019 to 2022 inclusive. There is a terrible irony in that too, given that the Judaism’s strongest appeal at the time of its foundation is likely to have been opposition to child sacrifice. Right now, the remaining population of Gaza seems to be facing devastating famine and the prospect of a final brutal Israeli land onslaught with nowhere liveable left to flee. This has to stop. It has to stop now. It has to stop because, as written in the famous poem adapted from the work of the Persian poet Hafiz, it really is true that “there is just one flesh to wound”. (REF 5).

Salaam/Shalom

Notes and Sources

Stop the Quantitative Tightening Scam

Posted on March 11, 2024 by gezwinstanley

From November 2022 until January 2024 the United Kingdom borrowed £118 billion, not to invest it in any way, but simply to “remove it from circulation” or, in simpler words, destroy it, cancel it out of existence.

£118 billion is a huge sum of money. Just think what else we could have done with that vast amount if we hadn’t just cancelled it out. £118 billion is Labour’s £28 billion green spending pledge restored. You can add in the whole of the government’s now scrapped Health and Social Care levy, which was meant to fix the NHS and social care by investing £36 billion from 2021 to 2024. It’s the extra £7 billion that Tom Tugendhat wants us to spend each year on defence, now that we are in an active confrontation with Putin’s Russia. That still leaves tens of billions for building new council houses. Possibly we could even fill in some pot-holes….

Moreover, the £118 billion is just the start. The Bank of England plans on borrowing a further £737 billion over roughly the next seven or eight years, just to cancel it out of existence.

Those sums are so large that those amounts don’t even seem real. Unfortunately those hundreds of billions are very real indeed. Why, at a time of multiple, interlocking crises, is the UK doing such an apparently crazy thing?

This is all about destroying the money created by Quantitative Easing (QE) to deal with first the 2008 Financial Crash, and then the pandemic (plus a bit more for Brexit). That further begs the question of why the UK feels the need to cancel out that QE created money in the first place. The justifications for all this have varied over time, but the latest, as at February 2024, are given in this report by the Treasure Select Committee: https://publications.parliament.uk/pa/cm5804/cmselect/cmtreasy/219/report.html.

Unfortunately, given the extreme importance of this issue, that report is full of holes, copiously uses technical jargon and euphemisms in a way that looks as if it is at least unconsciously intended to deceive, and has been produced essentially by two institutions, the Conservative dominated Treasure Select Committee, and financial industry influenced Bank of England, whose motivations have to be called into serious question.

Firstly, we may think that the main rationale for borrowing money, for cancelling hundreds of billion of pounds out of existence, is to fight inflation. That is not, however, the case: the Bank of England regards its setting of interest rates as its main weapon against inflation, and sees the likely affects of Quantitative Tightening (QT) on inflation as being both difficult to quantify and most likely not significant. In August 2023, reviewing QT to date, the Bank of England said this: “the impact on activity and inflation is also likely to have been small” .

Instead, the primary justification for QT, as stated in the Treasury Select Committee’s report, now seems to be “in order to create space for future interventions,” in other words for a future resumption of QE.

This claim is so disingenuous it would best be described as preposterous. Now money is just a claim on resources, not actual resources, so governments (at least those with what are called “fiat” currencies) can genuinely conjure it into being at will, and then force people to accept it as payment by mandating that money as “legal tender”. However, there is one obvious constraint on a government’s ability to create money: inflation. Adding to a money supply, without increasing the resources that money can be used to buy, reduces the purchasing power of each unit of money.

The reason why the UK was able to create £875 billion of additional money, and then use that money primarily to address the effects of the Financial Crisis, including saving our banking industry, and then dealing with COVID, was because those periods were deflationary. The potentially inflationary effects of Quantitative Easing were largely offset by prices that would otherwise have been falling. That indeed was one of the main points of QE in the first place, as, however, unpleasant inflation feels, economists are actually more scared of deflation. In a situation where prices are generally falling, people will tend not to buy today what they anticipate being able to buy more cheaply tomorrow, and the whole economy can stall and spin downwards into a “deflationary spiral”.

Our reality now, and for the foreseeable future, is, however, obviously inflationary, due to a sequence of supply shocks, most lately the disruption to shipping, arising from the Gaza crisis, affecting the route through the Red Sea and Suez Canal, and climate change induced droughts reducing the water available to operate the locks in the Panama Canal. With the ongoing and intensifying climate emergency those supply side pressures are, of course, likely to continue.

There will be no prospect of sustained “future interventions”, in other words more QE to reverse the public funds destroyed by QT, unless inflation is likely to fall significantly below the Bank of England’s inflation target of 2%. To pretend otherwise is absurd and very deceiving. Moreover, if we were to actually find ourselves in another deflationary situation, no one would care how much QE you had done in the past: all they would want is for the Bank of England to quickly create new money to stop the deflation. The implied assumption that there is a finite limit on QE, that applies across all time, is nonsensical. The whole argument that we are doing QT to allow more QE to be done in future is insultingly absurd.

The £118 billion destroyed so far is gone. Once the whole £875 billion is wiped out it will also be gone. There will be no reversing those increases to the real public sector debt burden any time soon. Also, as we saw in October 2022, it isn’t easy for governments to borrow another odd hundred billion, to say address the multiple crises we now face, to compensate for the hundred odd billion borrowed so far just to cancel it out, without causing other financial instabilities, such as driving up interest rates or trashing the value of government debts already held by institutions, such as pension funds, potentially making them bankrupt. Just as government money creation is constrained by inflation, government borrowing is constrained by the risk of destroying the value of government debt already held by institutions that have, for example, invested workers’ life savings into purchasing that debt, and making government’s future borrowing, and indeed everyone’s borrowings, much more expensive.

That in itself is another of the huge issues with the Treasury Committee’s report. Though it talks at some length about the losses to the public purse being caused by QT (mainly down to the fact that QE bought UK government debts, i.e. “gilts” in the jargon, at prices above that at which those gilts are now being sold back to markets, under QT) it never even mentions how borrowing money, just to cancel it out, will make it much more difficult for governments to borrow money to invest our way out of all our current difficulties.

QT is essentially the cause of our new rounds of austerity, and that fact needs to be stared fully in the face. Also, if we accept that public investment will be needed to facilitate sustainable economic growth, it condemns us to continuing stagnation, so any talk about growing our way out of our problems is, again, total nonsense.

It is also important to understand how the UK did QE, as otherwise we can’t grasp that, at a time we are repeatedly being told that there is no money for public spending, we really are borrowing hundreds of billions just to cancel that money out. Even though governments create money, “direct financing”, meaning just creating that new money and spending it, is seriously frowned upon among central bankers: indeed it is actually illegal for members of the EU. The new QE created money was therefore used to buy back government debts (gilts), but those debts were not cancelled out: instead they were held at the Bank of England. The plan was to eventually reverse the QE either when the debts matured, or by later selling the debts back to the markets: those methods of reversing QE are called “passive” and “active” QT receptively. Now passive QT involves having to pay off the capital of a government debt, now effectively held by government itself though the Bank of England, which is done by reborrowing the money and then cancelling the money so received out of existence. Active QT is similar: it involves “selling” the debts back to the “secondary markets”, in other words reborrowing the money again, and then again cancelling the money received out of existence.

This is yet another failing in the report. Nowhere does it state that QT involves borrowing money just to cancel it out: instead we get all those euphemisms. Reborrowing the £875 billion becomes “selling the Bank’s stock of QE gilts in the secondary market”. Cancelling money out of existence, becomes the much “softer” “removes from circulation”. This is very dangerous, as even the report itself says this: “Given the uncertainty over its effects, the Bank should also update Parliament and the public on QT”. How engaged, so you feel, dear reader, by the Report’s jargon? Were you even aware of this issue and it’s implications? Of the real reason for our never-ending austerity? When did you last hear this discussed on the news? Alternatively, does the way this topic is being covered when it is even discussed, and how it is so little discussed at all, make you feel as if you may be being “gas-lit”?

Also, when QE was started, there was no anticipation of the huge sums that would eventually be involved. It was intended just as another intervention to keep money flowing once there was no capacity to reduce interest rates below zero. Instead it evolved into almost a trillion being created, with, at times, government debts being bought back as soon as they were issued. Rather, this was direct financing in all but name, and it worked: it allowed us to survive a decade of crises. Now, however, the plan is to turn all that created money into real public debt. That will cripple us for generations: remember that, infamously, it wasn’t until 2015 that we finally paid off the debt incurred through the compensation we paid to slave owners for emancipating their slaves.

There is an alternative to QT. As Ford Prefect famously challenged the Vogon guard in “The Hitchhiker’s Guide to the Galaxy”, just before said guard threw him out of an airlock, we could just “stop doing it, of course”. The immediate way to do that is to simply “roll-over” government debts owned by the Bank of England when they mature: in other words, when a Bank of England owned government debt matures, and more money must be borrowed to repay the capital of the debt, immediately buy back the new debt with more QE: this leaves the accounting position exactly as it was before the debt matured and therefore means no net additional QE or QT. Active QT can be stopped simply by ceasing sales of the debts back to the markets. A more permanent plan is one outlined by Thomas Piketty, and more than a hundred economists, politicians and campaigners from across Europe, as part of a similar conversation they had with the European Central Bank (https://www.euractiv.com/section/economy-jobs/opinion/cancel-the-public-debt-held-by-the-ecb-and-take-back-control-of-our-destiny): that is to convert the government debts still owned by the Bank of England into a zero interest never-maturing debt, which avoids destroying money while leaving the Bank’s balance sheet still perfectly healthy, according to current international central banking accounting conventions.

To finally accurately record the fact that this “debt” is in no way owed by the UK public sector to anyone else, we should remove the value of this Bank of England owned public sector debt from all measures of that debt, better reflecting the UK public sector’s actual credit worthiness.

Whichever of those approaches we take would still leave us with the issue of the “reserves” the commercial banks hold at the Bank of England. Basically all the QE created money ended up in those reserve accounts, and, from 2006 on, the Bank of England has been paying interest on all that money to the commercial banks, who have therefore repaid being saved by QE in the first place by being allowed to make hearty excess profits. The way to solve that one is to reduce those payments by tiering the interest rates paid on those accounts, as is already done by European Central Bank and the Bank of Japan (and the way the Treasury Committee Report discusses such a possible tiering of interest rates is so outrageous it implies influence from the banking industry so intensive that it amounts to a conflict of interest serious enough to be considered misconduct in public office).

Why, then, are the Bank of England and the Treasury Committee doing QT at all? Part of it is group-think, this is just the “done thing”, even though, as the report itself acknowledges QT, and especially active QT, is “ innovative for all central banks”. To an extent, they are trapped by their own jargon (and here it should be noted that even the current Labour Shadow Chancellor worked at the Bank of England and HBOS). Nevertheless, it is also wise to take account of the individuals involved here, their associations and the interests they really represent. There is something of a revolving door between the Bank of England and the UK and US banking industry, and the Treasury Select Committee is dominated by Tories, who hold the chair and six out of the eleven committee posts. If you want a “smaller state” so that you can cut the “needy” adrift, and there is more role for your businesses to make profits, then QT is one way to bake in eternally brutal cuts to public spending. Perhaps we are back to Vogons (“not actually evil, but bad-tempered, bureaucratic, officious and callous” ). Then again, given the hundreds of thousands of British citizens now estimated to have died prematurely due to austerity (see:
“www.dannydorling.org/Austerity led to twice as many excess UK deaths as previously thought – here’s what that means for future cuts”) perhaps we should mix our metaphors and say “Daleks” instead.

Reading between the lines of the Report, even the current Chancellor, Jeremy Hunt, seems to be attempting to distance himself from QT: “QT is a tool that is exercised independently as part of its array of tools on monetary policy. It is therefore not appropriate for me to comment on that, just as I would not comment on whether it is right or wrong to raise or lower interest rates.” In other words Mr Hunt is hiding behind Bank of England independence to side-step the whole issue of Quantitative Tightening, as if a government controlling our sovereign Parliament could not redefine the rules for the Bank of England at will. Given the significance of QT, this position is the mother of all cop-outs.

In summary, and with apologies for bluntness, let’s just accept the fact that, through QE, we created some money. So what: live with it. Unless, in fact, we do live with it then, in a state when we are facing multiple crises, such as the climate emergency, collapsing public services, the threat from Russia and the prospects of a Trump presidency, there is a real risk we won’t be able to live at all. We want our futures back. We need to stop QT now, or we are doomed. With £100 billion now being actively destroyed each year, the clock is ticking.

Notes:

1) For further blogs on this topic see:

Don’t Destroy Trillions

which discusses the international context and:

The quiet plan to add an extra trillion to public debt and justify eternal austerity

which explores the UK situation in more detail.

2) The £118 billion of borrowing so far, which was then just cancelled out of existence, was derived as follows.

The source data is in this paragraph in the Report:

”The Bank is still in the relatively early stages of QT. As of 31 January 2024, the Bank’s stock of gilt purchases, as measured by the price at which they were bought, stood at £738 billion, down from £875 billion at the onset of QT (of the £20 billion in corporate bonds purchased at the onset, less than £1 billion remained). In September 2023, the MPC decided to undertake an additional £100 billion of QT over October 2023 to September 2024 (with passive and active QT accounting for around £50 billion each), which will bring the purchases of gilts down to around £650 billion.”

3) Here again is the “House of Commons Treasury Committee Quantitative Tightening Fifth Report of Session 2023–24” : https://publications.parliament.uk/pa/cm5804/cmselect/cmtreasy/219/report.html

Enjoy.

Are we all bound for Gaza?

Posted on December 17, 2023 by gezwinstanley

What is happening in Gaza is an absolute abomination in its own right. But do these horrific events have an even wider significance? As well as being the latest and most terrible episode in the Israeli-Palestinian conflict, a second Nakba, is it also the first slaughter of the “Disenlightenment”? If so, are we all bound for Gaza?

The “Disenlightenment” is the end of the progressive project that started with the Enlightenment, of serious and sustained attempts to create societies that serve the many and not just the ruling class. The “Disenlightenment” is concomitant with a re-convergence of economic and political power. Power again concentrating in the hands of the few is partly a consequence of globalisation, partly a return to far older and darker norms after a long period when established power structures were disrupted (see “Rage, rage against the dying of the light” ), and is itself both a result of and a cause of spiralling inequality. That underlying trend towards inequality was further exacerbated by the Financial Crash and COVID. Another essential aspect of the Disenlightenment is a consolidation of elite power so as to best secure their position in the face of rapidly developing crises that are, ironically, substantially of their own creation, especially climate change.

Turning societies back to only serving the ruling class is never likely to attract the sustained support of the masses. The Disenlightenment therefore requires a rolling back of democracy, with members of the elites cynically exploiting the anger their very policies have caused to trigger right wing “pluto-populist” fake “revolts” that promote authoritarianism. Such developments are also consistent with the increasing incompatibility of capitalism with democracy, as “polycrisis” causes economic growth to falter: for, as the economy drifts towards a “zero sum game”, the only way the rich can carry on competing with each other for wealth and power is to take from others, which in practice broadly means “us” (see “Deep, dark under-currents at the bottom of the gravity well” ).

Now the Gaza situation clearly fits within the ongoing tragedy of the Israeli- Palestinian conflict, and that must not be down-played, but it also has aspects that portend of things to come, which also allow the horrors in Gaza to potentially be characterised as the first slaughter of the Disenlightenment. Those are three-fold:

Firstly, the slaughter is being executed by a right wing “pluto-populist” regime (FT: “The age of the elected despot is here”) with definite authoritarian intentions, of a kind that typifies the Disenlightenment. With its vision for a future for Israel that amounts to ethnic cleansing, in the West Bank as well as Gaza, it is unlikely that the actions of the Israeli government would be so cruel, and the death toll so high, if that was not the case.
Then there is the pronounced asymmetry of the current conflict, which arises from a massive difference in economic power and therefore in military capability. The Gazans are being destroyed by some of the most expensive and advanced technology currently available, including, most notoriously, AI ( “A mass assassination factory” ). The Israeli government is also being bankrolled by rich allies and associates across the world, reflecting another aspect of the Disenlightenment, that the economic power of our current elites is massively boosted by being networked globally and able to operate at a transnational scale (“Lowkey exposes British role in Israel-Gaza war” ).
Finally, and most tellingly, there is the way this is the slaughter of an indigenous population, that has first been dispossessed, ghettoised, “othered” and now exterminated. This is not a war fought in distant lands, but at home. What is happening in Gaza echoes settler-colonial projects down the ages, such as the European destruction of Native Americans. Nevertheless, it could also foreshadow class conflicts to come, as authoritarian regimes, serving the globalised ruling class, clamp down on impoverished masses who now, replaced with AI driven automation, have no point even as a “working class” and instead come to be seen, not just as a drain on resources, but as a serious threat, to ultimately be neutralised by any means necessary.

A counter-argument would be that the Israeli-Palestinian conflict is based around ethnicity not class. However, class is all about families seeking to retain their economic advantages, and by extension the disadvantages of others, across generations. Class structures often start with ethnic divides: where one population takes over another. Over time, the populations merge, especially as the surplus descendants of the ruling class descend, over a generation or two, into the masses. The class structures, however, often endure.

It is possible to see an embryonic class structure in the Israeli-Palestinian conflict, as evidenced by the Israeli finance minister’s, Bezalel Smotrich’s, notorious “Tipping Plan” ( “Tipping of the Scales Plan”” ), which would condemn Palestinians to a form of subjugation where they would not even be able to vote in the Israeli Knesset, and does not address the issue of compensation for the violent seizure of their ancestral homes and lands at all. Under this plan Palestinians either have to accept ethnic cleaning or their lot as a landless, subordinate class wide open for exploitation, further accelerating the existing “long-standing tendency of the Occupied Palestinian Territory to serve as a reservoir of cheap labour for the economy of Israel” (UN: “Developments in the economy of the Occupied Palestinian Territory” ).

Perceived differences of ethnicity make it easier to brutalise the “other”. Therefore, as we move ever further into the Disenlightenment, with tensions steadily rising, the first slaughters of the Disenlightenment will occur between classes where ethnic differences are still regarded by both groups as significant. What is happening in Gaza, and at a slower rate on the West Bank, can therefore genuinely be interpreted as an early warning: as a harbinger of the directly violent class war (as contrasted with the merely indirectly violent class war that equates to deprivation of proper healthcare, housing or even food) that will characterise the Disenlightenment, as it is precisely in such contexts that the first slaughters will occur. Basically, as a growing divide in life prospects, and even chances for survival, seeds resentment, while the coercive powers accessible to the richest increase and the regimes they back become more authoritarian, all developments associated with the Disenlightenment, then mass inter-class violence will intensify, with such conflicts erupting first within societies where tensions are already highest.

Our globalised ruling class, or at least broad sections of our elites, are really not our friends: rather they conceal a steely ruthlessness, sometimes even from themselves, with a layer of cultured gentility. We have seen their reluctance to act on climate change, despite themselves being the most major contributors to the problem (“Climate Equality: A planet for the 99%” ); to hold onto their patent rights during a pandemic, costing hundreds of thousands of lives in 2021 alone, even though the patents were based on research that was mainly funded by governments and the charitable sector (“Intellectual Property and “The Lost Year” of COVID-19 Deaths” ); to collude, in many developed nations, with central banks, to wipe out trillions in public funds, there-by maintaining austerity and crippling climate action, while continually burying the fact that governments can create money to save banks, but when it comes to saving humanity, are not allowed to even permit that extra money to remain in existence ( “Don’t Destroy Trillions” ). In the case of the Gaza horror we have seen the establishments in the USA, the UK and the EU complicit with ethnic cleansing and potential genocide, with money, arms, intelligence sharing and even deployment of forces intended to “hold the ring” and prevent any counter-intervention by other powers.

Seen in this light Gaza is indeed the first slaughter of the Disenlightenment. We are already the dispossessed: most of the world’s population was deprived so long ago, of even a space where they could live freely, by a variety of conquests, expulsions, persecutions, impoverishments, disinheritances, enclosures and clearances, that ultimately many have forgotten that those events even happened. Gaza is a foreshadowing of a time when you or your long dispossessed family’s descendants, starving in a heavily surveilled and walled-in ghetto, provoked into such acts of evil that you can be condemned as “human animals”, are then systematically exterminated by the machines of the ruling classes, who are dripping with conceit and ruthless in their smug conviction that it is they who are the true defenders of “civilisation”.

Demand an immediate ceasefire in Gaza, and justice for the Palestinians, meaning a two state solution or a democratic one state solution where all have equal rights and protections, and that the Palestinians receive restorations and reparations to undo the effects of their dispossession. No compensation can ever dry-up the trail of tears endured by many on all sides of this conflict, but, by restoring their economic independence, reparations would allow the Palestinians to better protect themselves from the continuing depredations of others.

We must stand with the oppressed, or “Gaza” lurks in wait in all our futures.

Don’t Destroy Trillions

Posted on September 29, 2023 by gezwinstanley

Imagine you are facing multiple emergencies, that are going to require a lot of spending. Strapped for cash, you go down the bank, and borrow lots of money: suitcases full of it. Then you set fire to all that money. You just burn it up, leaving you now with huge debts, with your credit so maxed out that you’ll struggle to borrow any more, at least without paying exorbitant interest rates, and therefore with no way to actually deal with all those crises you are facing.

That would be insane, wouldn’t it? However, that is exactly what governments, and their central banks, in many of the largest economies, are in effect doing RIGHT NOW.

For example, in the US the Fed is currently destroying money at an annual pace of roughly $1 trillion (Source: “The Fed’s trillion-dollar problem”).

The European Central Bank has been destroying €25 billion per month from July 2023 onward. (Source: “Can the ECB escape its own trap?” ).

The UK is aiming to destroy £80 billion annually for the next decade (Source: “Quantitative tightening: the story so far”).

The vast majority of that money has to be borrowed first before cancelling it out of existence. It is a similar story in most “developed” nations, including Canada, Australia and New Zealand, with the notable exception of Japan.

So why are they doing something that, if we take a step back for a moment, is obviously mad? It’s because those central bankers want to destroy all the money created since the Financial Crash through “Quantitative Easing” (QE). QE was the way governments injected money into the economy to prevent downturns, first in the aftermath of the Financial Crash and then later during COVID, turning into full blown economic depressions.

However, governments creating money is seriously frowned upon. “Direct financing”, where central banks create money to directly fund government spending, is even illegal for member states of the EU. Therefore, when central banks wanted to create money and inject it into the economy, they didn’t just hand over that money to their respective governments. Instead the central banks used the newly created money to purchase debts, but, as it happened, not just other people’s debts. Though some corporate bonds and mortgage backed loans were purchased, overwhelmingly the central banks purchased their government’s own debts, so for example treasury bills in the US or gilts in the UK.

This, so it was intended, would also make QE eventually easy to unwind: to destroy the QE created money all a central bank would need to do would be to sell those debts back to the markets, and then cancel out the sum received.

This approach works reasonably elegantly when those debts being bought and sold are someone else’s. It doesn’t work so well when, and sadly this is usually the case, those debts are the government’s own. When a government, though its central bank, sells its own debts, and cancels out the proceeds from the sale, it is then just left with its original debt, now owed again to third parties instead of essentially to “itself”. “Selling government debt” is just a euphemism for re-borrowing all that money. It’s the same story if a government that owns its own debt just lets the debt mature, as a debt “maturing” just means that the capital of the debt has to be paid off. The government has to pay off the capital of that maturing debt by re-borrowing the money, which it then again just cancels out of existence along with the maturing debt.

Now, again, that might not matter if we were just talking about about governments pre-existing, pre-crisis debts. Except we aren’t. Instead, since 2008, governments have issued trillions of dollar equivalents of debts to deal with the Financial Crisis and its after-effects, and then with the pandemic. This didn’t matter too much at the time, as those debts were normally repurchased with newly created money, sometimes immediately the debts were issued.

In practice, though the central bankers hate to admit this, governments were funding their responses to over a decade’s worth of crises by creating new money.

There is nothing innately wrong with government creating new money. Indeed, governments have to create money in the first place before, for example, they can receive it back in taxes. Of course, money is not a “real resource”, rather money is a claim on resources, backed by a government’s insistence that its own issued money has to be accepted for making purchases. Therefore, if a government creates more money for itself, it adds to its purchasing power, while initially reducing everyone else’s, as there is now more money in circulation to buy things but no more things available to buy. Eventually, though, as a government spends its newly created money, that money flows into the general economy so that, if the government is reasonably competent and broadly accountable to its citizenry, most people should benefit. A government creating new money can allow it to fairly quickly acquire the resources it needs to solve a crisis and protect the people it serves and represents. In fact, money creation is likely, for most of the people most of the time, to be the least painful way a government can raise emergency funding, as compared to a sudden increase in borrowing or taxes, or selling off publicly owned assets.

Governments funding themselves through over a decade’s worth of crises by creating new money was therefore perfectly sensible. Doing so didn’t even create any significant inflation, as that new money just counter-acted potential deflation. A general economic depression was averted, the Financial Crash did not destroy the banking system and public health interventions such as mass COVID vaccinations saved millions of lives (Source: Global impact of the first year of COVID-19 vaccination:).

What is insane is now demanding that all those trillions be wiped out: in insisting that all that public debt that was purchased is “sold back”, in other words, turned into real debt that will cripple the public finances for generations.

Quantitative tightening can and must be stopped. There are the inevitable technicalities that have to be dealt with, meaning that a government’s now self-owned debts can’t simply be cancelled out of existence, but those difficulties can still be solved at the stroke of a pen.

The problem is that the QE created money has to exist somewhere, and in fact, though it flows around as the money is spent and received, it ultimately exists on reserve accounts held at central banks, mostly by commercial banks, who use that money at the end of each day to clear their net balances with respect to each other. That’s why governments have to borrow back the money before they can cancel it out of existence.

These reserves create a couple of issues. The first is that, by international convention, those reserves at the central banks are treated as liabilities, i.e. as negative items on a central bank’s balance sheet. Now, the reserves are very strange liabilities. As argued here by a former head of the OECD’s General Economics Division (Source: “Monetary policy and the value of Public Debt”): “Note that the central bank’s liability is a very odd liability indeed. Unlike the case of government bonds, settlement of the central bank’s liability can never be demanded, even in principle. Bank reserves are money, the stuff in which all transactions are settled. If banks were to demand repayment of the liability that demand would be meaningless. Ever since countries abandoned the gold standard bank, reserves have become the ultimate money.”

The other way these reserves are odd liabilities is that they are usually not included in measures of national debt, nor in traders’ assessments of national debts. Having extensive reserves is very unlikely to count against a nation’s credit-worthiness: all it means is that the nation’s banking system is more resilient, which would be a peculiar reason to demand it pay higher interest on its bonds.

By contrast, government debts that have been purchased by a government’s own central bank are regarded as assets in accounting terms, and so have positive value. This means that, if all that QE purchased debt was simply cancelled out of existence, without also cancelling out the reserves created to purchase that debt, central banks’ accounting balances would become significantly negative, and no one is certain as to what the side-effects of allowing that to happen could be.

The second issue is that, these days (for example in the UK since 2006) central banks pay interest to the commercial banks on those reserves. So, the larger those reserves, the higher the interest payments, and, as interest rates increase, as they are doing right now, those interest payments increase still further.

Central banks pay interest for two reasons: to provide a floor for interest rates in the broader economy, and to encourage each bank to keep enough on reserve to ensure it stays solvent. However, again as argued by that former senior economist at the OECD, “Payment of interest on bank reserves is not the result of any obligation, legal or moral. It is simply a pragmatic policy choice, a device used to control the credit activities of commercial banks.” Those same two objectives could be met by central banks only paying the highest rate of interest on reserves above a certain level, in other words “tiering” the interest rates. Indeed, tiered interest rates are already the norm at the European Central Bank and the Bank of Japan. With tiered interest rate the continued existence of those reserves ceases to be a problem.

Nevertheless, this still leaves the issue of what to do about those QE purchased government debts if those debts can’t simply be cancelled out of existence. However, in the EU, more than a hundred economists, politicians and campaigners from across Europe, including Thomas Piketty, signed an open letter to the European Central Bank opposing Quantitative Tightening (Source: “Cancel the public debt held by the ECB and take back control of our destiny”). Their proposal is to turn those effectively “self-owned” debts into “perpetual debts with 0% interest rate”. To better reflect credit-worthiness, it would then also be sensible to remove the value of these “self-owned” debts from any measures of government debt. That is a perfectly valid approach, and over time, of course, the value of those “perpetual debts” would wither away under inflation, until they could indeed be harmlessly cancelled out, or the conventions about central bank accounting or the nature of money could themselves evolve, so that again simple cancellation would have no significant side effects.

So, why are central bankers continuing with Quantitative Tightening, and adding trillions to public debts, simply to wipe the borrowed money out of existence? There are two reasons why our central bankers and their top advisors are so staggeringly blind. Firstly we’re back to that stigma of money creation: central bankers don’t like to admit, even to themselves, that QE wasn’t just about monetary policy, such as stopping deflation, but also about supporting government spending, through clearly it was, as sometimes government debts were issued only in order to be immediately bought back with newly created money, and there would have been financial black holes across many wealthy nations hundred of billions deep otherwise. Secondly central banks don’t care about government spending, except in the limited situation when they’re asked by the government to raise more funds. Otherwise it is literally outside their ambit: it is someone else’s responsibility and problem. Then there is the additional factor that cuts in government spending, and general “austerity”, usually doesn’t affect senior central bankers and their mates: they tend to send their kids to private schools, have access to the best private healthcare and can still easily afford a place to live, possibly even inside a gated community with its own private security.

If dealing with the financial crisis and then the pandemic required governments to create money to cover their elevated spending (and it did) why does a global potentially civilisation destroying climate emergency not have even greater implications?

So, at a time when human civilisation faces its greatest existential crisis ever, we will globally cut our ability to respond publicly to that crisis, through governments that are meant to be our most powerful expression of our collective will, by literally trillions of dollars and dollar equivalents. We cripple our capacity for joint action at a time we need it the most. To call that decision CATASTROPHICALLY STUPID is an under-statement.

(More here: The quiet plan to add an extra trillion to public debt and justify eternal austerity)

Singularity: not quite yet, but still inescapable

Posted on June 25, 2023 by gezwinstanley

On 13^th June 1863 the novelist Samuel Butler wrote a letter entitled “Darwin among the Machines”. There Butler says this: “ The views of machinery which we are thus feebly indicating will suggest the solution of one of the greatest and most mysterious questions of the day. We refer to the question: What sort of creature man’s next successor in the supremacy of the earth is likely to be. We have often heard this debated; but it appears to us that we are ourselves creating our own successors; we are daily adding to the beauty and delicacy of their physical organisation; we are daily giving them greater power and supplying by all sorts of ingenious contrivances that self-regulating, self-acting power which will be to them what intellect has been to the human race. In the course of ages we shall find ourselves the inferior race. ”

Arguably, the above is one of the earliest premonitions of what eventually came to be called the “Technological Singularity”. The mathematician, physicist and computer scientist John von Neumann was the first person to use the term, when, in the mid 20^th Century, expressing concerns about “the ever accelerating progress of technology and changes in the mode of human life, which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.” (John von Neumann )

Roll forward to 2023 and artificial intelligences called Large Language Models (LLMs) were passing the US bar exam. (ChatGPT can pass the bar exam) leaving it natural for us to wonder if we are at least in the opening stages of the Singularity or, even if not, what momentous opportunities and challenges these developments could foretell…

How do these models work?

Large Language Models like Microsoft/OpenAI’s ChatGPT or Google’s Bard are examples of machine deep learning. Machine deep learning involves allowing a program to “mutate” how it processes its input data, and then comparing the output generated to an expected result, but doing so again and again, repeatedly mutating the program while retaining the best mutations as the basis for further refinements, and simply throwing away those changes that make the output diverge further from the expected result.

The above is a brute force but powerful approach, especially given enough computing power, data to train upon and time. Eventually, just like the natural evolutionary approach itself from which it is derived, it can yield spectacular results. In a way, this really is “Darwin among the machines” in its purest form.

Now there are many types of machine deep learning. It is possible, for example, to mutate the steps in a symbolically defined algorithm, such as a mathematical equation. Nevertheless, information processing often cannot be reduced to a series of neatly defined equations. To cope with the complexity and general messiness of input data, the leading form of machine learning uses what are called “neural networks”. Neural networks are again inspired by nature: being simulations in computer code of how nerve cells take in inputs and then generate outputs to other layers of nerve cells, but with the network as a whole being able to vary how it is connected up, and how the input signal gets changed as it is passed along. In this way, as in all deep learning approaches, the neural network takes in an input and turns it into a different output that can be compared to an expected result, and it is the state of the network that encodes how input data is transformed into outputs.

The AIs that are currently generating the most controversy are, as mentioned above, Large Language Models. These are neural networks that essentially have been trained on vast tranches of data on the internet: text and images. Given a question LLMs can perform a web search and then convert the results into a confidently rendered answer, or, fed a textual description, generate an image. You can try this out yourself by registering to use Microsoft’s Bing Chat, which is itself using a version of Open AI’s ChatGPT under the covers. At Google the latest iterations of generative AIs are being trained to do similar things with spoken words, using the entire content of Google owned YouTube as training data.

So is this the Singularity?

Such AIs obviously learn by evolving their neural networks, but they are not currently self-evolving to improve their ability to learn, i.e. they are not learning how to learn more efficiently (let alone learning how to learn to learn more efficiently, and so on ad infinitum, in an infinite regress). That process would rely on humans tweaking the programming behind the models and optimising training data. This is not therefore the point at which super-intelligent machines redesign themselves to be more intelligent, which in turn allows them to design themselves to be still more intelligent, and so on again ad infinitum, with the intelligence of such devices exponentiating towards an apparently infinite rate of improvement: a mathematical and technological singularity.

Nor do we yet seem to be at the point where generative AI has attained Artificial General Intelligence (the much sought after “AGI”). For starters, these models are currently limited to handling particular kinds of data: text, images and now increasingly the spoken word. There is little interaction currently going on with the real world. The totally autonomous and sufficiently safe self driving car remains a challenge. There are a couple of reasons for this. An LLM that is 95% accurate is still a very useful tool, especially if users are encouraged not to accept its answers blindly, while a totally autonomous vehicle that drives itself correctly 95% of the time would be deadly. So, in the field of robotics, outside very constrained contexts, the need for AI accuracy is far higher. Secondly, there is the issue with training and training data. A generative AI that has to learn to interact correctly with the physical world through brute force trial and error can only do so very slowly and at the cost of many literally damaging mistakes. There are approaches that can get around these problems. Tesla is using data captured by it’s cars’ camera to help train its fully self driving AI model. (How Tesla uses and improves its ai for autonomous driving). That doesn’t, in itself though, allow Tesla’s AI any scope for interaction, so another approach is to place the AI in control of a simulated “body”, such as a self-driving car or other robotic system, within a simulation of the physical world, that can run many, many times faster than the real world and with which the AI can interact without any literally “real world” consequences. That requires the simulated world to itself have a very accurate and detailed “physics engine” or the AI could pick up some seriously bad habits (Synthetic AI simulations).

The upshot of all of this is that the current generation of generative AI isn’t fully autonomously interacting with the real world at all, but instead only with specific data-streams, mainly text and images, that are only a tiny subset of human experience. As such it is currently far too limited to be called AGI.

Is AI sentient?

When we move onto considering the question of whether or not AI is sentient, things get even more difficult. First we have to decide what “sentient” means. Generally it is taken to mean conscious, or at least to require consciousness. So what then does consciousness. mean?

Here we hit what is called the “hard problem of consciousness”, or, to the more philosophically minded, what is usually called the “mind-body problem”. This is the question of how the objective, “physical” world can give rise to the “felt” experience of our subjective mental world, as epitomised by puzzles such as how could you explain the sensation of colour to someone born blind?

As argued elsewhere in these blogs, it seems likely that there is no explanation that solves the mind-body problem, since explanation is a form of objectification, and therefore already sits entirely on one side of the mind-body division, and so cannot bridge that gap. However, the impossibility of ever “explaining” the subjective feeling of consciousness does not mean we can’t refine our ideas about how physical systems generate consciousness, which is equivalent to discovering the physical “correlates” of consciousness, or indeed of better understanding the role consciousness plays in the processing of information. If we can do the above, then we will be in a wiser position to decide whether or not a given system is conscious.

We are entering very murky territory indeed, and are going to have to speculate, so feel free to jump to the next section if you don’t want to delve into conjectures about the nature of consciousness at the moment.

Now the essence of consciousness seems to be captured by Renee Descartes’s famous proof of our own existence: “I think therefore I am”. That proof works because, if you don’t really exist, then what is it that falsely perceives that existence? And if that entity doing that perceiving is in itself non-existent, what is it that perceives its existence in turn? And so on ad infinitum. It’s turtles all the way down. The essence of consciousness appears to be infinite regress: something that reflect upon itself infinitely in a hall of mirrors.

If that is the case how can a physical system execute an infinite regress in finite time? The only way that could be possible is if that processing is instantaneous. And the only physical processes we know about that can occur instantaneously are quantum effects. Therefore, in a physical sense, quantum effects must be necessary for the generation of consciousness.

In Newtonian physics there is a notorious Three Body Problem, where the mutual gravitational interaction of three or more bodies make it impossible to entirely calculate their future motions. For systems that generate consciousness, we need to think of an “n” body problem, where each component part interacts instantaneously with all the other parts in a completely non tractable way, to instantly generate an output.

Now, it is a majority opinion among computer scientists that generative AIs are currently not conscious. There are also some academics who believe, along similar lines to those outlined above, that quantum effects underpin consciousness, perhaps most famously the Oxford University professor of mathematics Roger Penrose, in his books “The Emperor’s New Mind” and “Shadows of the Mind”. Nevertheless, the quantum basis for mind is still a minority position, despite growing acceptance of the ways biological systems have evolved to exploit quantum effects more generally.

If this conjecture is true, and consciousness is another phenomenon arising from biology through quantum effects, we also need to ask why? What’s the point of consciousness or, in other words, why did it evolve in the first place? Should the answers to that question be consistent with the notion that consciousness is a quantum mechanical phenomenon then that conjecture would be reinforced. Equally, however, by better understanding the abilities consciousness facilitates, we would be in a clearer position to decide whether or not our current AIs do the same thing, and thereby not only more confidently judge whether or not they are sentient but also grasp the strengths and weaknesses of our latest AI models more accurately.

In his books on the subject, Roger Penrose focuses on the limits of computation, and how consciousness seems to transcend those limits. For example, our seemingly ready ability to spot infinite regress is something that, unless it had been pre-programmed with a whole set of special “rules of thumb”, a computer based model would struggle to ever acquire, because its processing would become trapped within some form of infinite loop. The latter point is key, as a problem that can’t be solved by a computer in a finite time, so that the program never stops, is formally “non-computable”.

Perhaps, then, the “point” of consciousness, is to allow at least some such non-computable problems to be solved, by exploiting quantum effects to permit at least some forms of instantaneous processing. Our brains would be able to transcend ordinary “algorithmic” computing, by running not algorithms but what we could call “transcendent algorithms”, or “transrithms”.

So what, in turn, would be the point of transrithms? Does evolution really care whether or not creatures can grasp an infinite regress in finite time? That might seem like a rather abstract ability, but even from an evolutionary perspective it may sometimes protect our minds from being trapped in infinite loops, from which we could only, at best, be rescued after some arbitrary time or number of iterations by some type of clumsy “emergency interrupt”.

It is also likely that, if we are correct in identifying infinite recursion as an essential characteristic of consciousness, that transrithms create meaning. Meaning is the subjective sense of the context of any data item. Now any neural network can create a simple context: it is just all the patterns that are associated with a given input. But to create a unified experience that relates all meanings to everything else we need a context for every context, and a context for those contexts of contexts, and so on ad infinitum, in another infinite regress. We need to create a context that contains all those nested and inter-related contexts, which creates a paradox, as that ultimate context would also have to relate to all the other contexts, in turn creating yet another all-encompassing context, which would then necessitate another ultimate context relating that context to all the other contexts, and so forth ad infinitum. This is Russell’s Paradox in mathematical set theory: an attempt to create a set of all the sets results in the creation of another set to be added to the set of sets and therefore the need to create another set of all the sets, a process that again repeats in an infinite regression. It is that infinite regression of contexts that unifies our subjective sense of meanings into a single experienced “reality”.

This fully integrated sense of meaning allows us to group similar things far more readily than a computer model that has to adopt a brute force trail and error approach to spot patterns in huge masses of data. If you want to get an idea for what this could mean in practice, try and get Bing Chat to grasp a new metaphor.

Transrithms may also explain how we as conscious beings can repeatedly continue to step back from a problem, to see it in a wider context, one that allows us to solve the problem laterally, or even reframe it in a way that makes the problem go away. Very likely this ability relates directly to us stepping back indefinitely through our infinitely nested contexts.

By contrast AIs are “trapped” by their programming. AIs have to mechanistically follow their algorithms, even if those are “second order” algorithms that tell the AIs how to “evolve” first order algorithms such as neural networks. AIs can never step out of this, to attain an “nth” order of processing. Only a system modelling itself to an infinite degree of recursion, via instantaneous and therefore quantum processes, as otherwise this problem is not computable in finite time, can do this, and escape these purely algorithmic limits.

What accompanies this infinite recursion is the feeling of subjectivity, and what emerges from those subjective feelings are wants and desires: the sense that certain experiences are be sought out, while others are to be avoided.

As stated, it’s this subjective dimension that allows a conscious entity to transcend its purely algorithmic programming. For example, tragically, this can allow a biologically evolved conscious entity to escape it’s equally biologically evolved imperative to survive by committing suicide, because its existence “feels” unendurable. With terrible irony, otherwise inexplicable acts of suicide may be one the clearest indicators of consciousness.

A possibility of “self indulgent” acts of suicide is also counter-survival, so, from a purely evolutionary perspective, the other aspects of consciousness we have outlined above, grasping meaning, avoiding infinite loops and potentially being able to step back or out of a problem, in other words not being trapped by a failing algorithm like a moth flying into a flame, must be significantly pro-survival in order to compensate.

If these speculations are correct, though, there is still one obvious question to deal with. If we, as conscious entities, can “think” instantaneously, why can’t we instantly solve every problem we are faced with? The most obvious answer is that our transrithms can only solve particular classes of problem. Usually when we think about information processing we mean algorithms: step by step processes to take in certain inputs and work out the intended outputs. Some stages of algorithmic processing can be performed in parallel, for example multiple simple iterations of a programmatic “loop” of some kind, but overall the process still proceeds sequentially, within each possible “parallel” stream of processing and whenever those parallel merge back together and combine various outputs. So the answer to this problem resides within the very word “instantaneous”, meaning at an “instant”. If by an “instant” we mean the shortest possible indivisible duration or “quantum” of time, then, by definition, that instant can’t be sub-divided, meaning that a sequence of multiple steps can not occur within that indivisible instant. The most sequential steps that can occur in an instant is one step, or, conversely, the shortest time a step can take is a “quantum” of time, and in a biological system a sequential “step” may take many orders of magnitude longer.

So what type of processing can occur in a single instant? Perhaps the clearest way to conceptualise that is to think of a single step, but repeated infinitely many times running in parallel, such as a basic recursion. Moreover, it is likely that this really is just the same step running infinitely in parallel, or at least a finite number of different types of step all running in parallel. If each of these parallel running steps was different, that would require a process to divide up an information handling task into an infinite number of different parallel running steps, something that we can strongly conjecture could not be accomplished in a finite time.

Transrithms are therefore very useful, but not magical. However, we may still wonder why, if a conscious brain can perform a step of processing in every absolutely miniscule single quantum of time, our brains cannot run far faster. Now, there are certain types of problem that can only be solved by the type of instantaneous processing we are postulating above, so instantaneous processing has a use in and of itself. For a biological brain, though, step by step processing at the rate of a single step per quantum of time may have no use at all, because the physical substrate of the brain so constrains the rate at which everything needed to support such a process could actually happen. With neural and muscular processes driven by changing electrochemical gradients, we can only form even temporary memories, say to store an interim result of such an astoundingly fast step by step process, let alone produce an “output” by making a muscle twitch, many orders of magnitude slower than any process running at the pace of a step per quantum of time. The brain as a whole can only run at the pace of its slowest biological processes.

If these conjectures are even approximately correct then it is clear that generative AI is not sentient, as it is not utilising quantum effects along the lines described above. This conclusion is consistent with the fact that we don’t appear to see generative AI “breaking free” of purely algorithmic processing in the way we are inferring that some biological brains sometimes do. Now, the quantum basis of consciousness, to re-iterate, is currently a minority opinion among scientists and engineers, but the scientific mainstream still contends that generative AI, in its present form, is not conscious, mainly because it is not, yet, an Artificial General Intelligence able to interact competently with the real world. Right now all that generative AI can do is to manipulate symbols.

Consider, though, even if all the above is wrong, and generative AI is conscious, it would be a very alien and narrow kind of sentience. It’s “lived reality” would usually comprise purely of textual and numeric data, or the pixels from still images, ultimately 0s and 1s, and it’s sole drive would be to ever more accurately find and replicate patterns in that data. It would have no purpose other than that encoded in its basic programming by its human creators. We would need to be extremely careful, perhaps under the influence of too many Sci-Fi blockbusters, not to anthropomorphise such minds.

Non-sentient AI could still pose a threat

However, even if AI is not yet sentient at all, it could still be very powerful, and potentially dangerous. We should never forget that AIs, even as merely electronic “zombies”, start off with a huge advantages over humans when it comes to data processing. In a human nerve signals travels at 70 – 120 meters per second. In a computer CPU signals travel at just under the speed of light, or about 3 million times faster. An AI strictly confined to the purposes that we accurately code into it could still act in a way that would lead to unintended consequences. Science fiction is littered with such examples: perhaps most famously that of HAL9000 in 2001 a Space Odyssey, which eventually reluctantly decides that the only way it can fulfil its programmed mission directives is to kill it’s entire ship’s crew. More recently, and apocalyptically, we have the Oxford philosopher’s Nick Bostrom’s speculations that an AI charged with making paperclips could wipe out humanity if it deemed them a threat to paperclip production, as it seeks to convert everything it possibly can into paperclips. Allowing mindless automata to, say, replace all life on Earth, to silence every single mind of any kind on the planet, feels like an even more tragic outcome than the one where the AIs themselves are genuinely sentient. It would be the ultimate “Zombie Apocalypse”.

Of course, as we have seen, generative AI is currently limited in the way it is interacting with physical reality, so the most obviously catastrophic examples such as Bostorm’s paperclip apocalypse or out of control militarised AIs, such as “Skynet” in the Terminator movie series are, right now at least, not possibilities. For the moment, generative AI’s training data is largely web based, not “real world”. Even so, the widespread adoption of even latest generation AI is very impactful, both in terms of opportunities and threats.

Even the intentional impacts could be immense.

AIs can rapidly parse huge amounts of data to detect patterns that no human could spot in any reasonable time. For example, progress in the biosciences is likely as new drugs are discovered and the genetic bases of certain diseases are better understood. Material sciences might benefit as the probable properties of possible new materials are derived en masse, e.g. allowing the invention of higher temperature superconductors. As one way to store electricity is to allow it to indefinitely circulate around a superconducting circuit, such a discovery alone would be a huge boost to low-carbon technologies.

The economic effects have already started. As, at least in part, advanced chatbots, Large Language Models having already started replacing human “customer service representatives” at call centres, such as in the notorious incident in May 2023 when the National Eating Disorder Association in the US substituted it’s call centre staff for an AI named “Tessa”, amidst allegations, on the part of affected workers, that the move was partly a reaction to unionisation, and the Tessa AI itself having to be withdrawn shortly afterwards as it was delivering potentially dangerous advice.(Democracy Now Top US and World Headlines 31^st May 2023 or Forbes: Non-Profit Helpline Shifts To Chatbots, Then Shuts Down Rogue AI ).

AI will also quickly increase the productivity of certain knowledge-based jobs. That could result in further job reductions, unless demand expands to compensate. Given the strains on most healthcare systems across the globe, we can imagine that for some jobs, such as medical diagnosis, that will be the case: as those services become cheaper, demand will grow to compensate and employment will not be threatened. For other jobs, such as legal services or accountancy, it is easier to imagine that demand might be saturated much faster.

So we will be looking at reductions in the demand for human work in white-collar professional and semi-professional jobs. Outside heavily routinised tasks in reasonable predictable environments, such as the factory floor, open fields on a farm, the warehouse or the supermarket checkout, it is actually “blue collar” interactions with the physical world that are currently hardest to automate, as we have discussed above. For now, plumbers and electricians are very safe indeed.

Now, these semi-professional and professional workers are highly educated, articulate and often socially and politically well connected. Nor are they at all used to the idea of being on the losing side of technological and economic changes. An attempt to marginalise these groups will produce a powerful counter-reaction of some kind. Most likely, after the dust has settled, instead of actual job cuts, the adoption of AI in these sectors of the economy will produce some combination of shorter working weeks and these workers effectively swapping earnings from employment for earnings from ownership, i.e. profit sharing.

Then there is deep fakery, and online manipulation. Generative AI is supremely placed to create ever more convincing disinformation, even interactively tailored to groups or individuals. Imagine millions of online social media ‘bots powered by generative AI. Then again, perhaps we already don’t have to imagine?

Even just considering the possibility of much more sophisticated and all pervasive censorship, with AI automatically promoting some content and suppressing the rest, in order to serve some government or corporate agenda, is disturbing enough.

Another concern, relating to how AI could assist more ruthless governments, or at very least the more unscrupulous agencies of governments, is in the field of surveillance. An AI trained in facial recognition will never tire of watching the cameras. AIs able to sort through huge amounts of data can monitor people’s communications and online activities in a way that must make the National Security Administration and it’s equivalents throughout the world dribble.

Short term unintended consequences

The unintended consequences have also already started to cause controversy. One of the earliest to emerge, even before the advent of ChatGPT, was that of learnt bias. Consider, for example, this scenario: if a racially prejudiced police force targets a certain ethnic minority for arrest, an AI trained on that data-set will, without any corrections being applied, do the same, essentially automating the darker side of human nature.

There are more subtle concerns too. There is a Google Research Fellow and Vice President who believes that generative AI is sentient, because, after all, in his view, what are people other than “Large Language Models”? However, as morality is just a “social contract”, we do not need to confer AIs with any rights. How easy it then becomes to extend such attitudes to all supposed “Large Language Models”, humans included, in a move as dehumanising as the widespread use of “replicant” slave-labour in the movie “Blade Runner”. (By contrast the moral framework behind these blogs is that we try and extend our reciprocating moral community to every sentient being able and willing to participate).

A less sinister but more immediate, and still very controversial, issue is how to reward content creation in a world where generative AI is busy repurposing everyone’s else’s content.

Compensating the creators

There is a view that the latest generation of Generative AI amounts to huge scale, automated plagiarism (just an example: Reasons to worry about generative AI). Now, this isn’t completely fair: a trial and error process tracking certain outcomes can produce genuine innovations. After all, you are a product of Darwinian evolution.

It is, however, a substantively accurate assertion: one reasonable way to think of AIs such as ChatGPT is to imagine a sophisticated chatbot converting questions into web searches and then collating what it assesses are the best results from those searches into a single output it creates itself. Now, this breaks the way creators normally make money from their web-based content, as there is no longer a need for users to visit the original source websites and so, for example, generate advertising revenue.

In a way, that is exacerbating a failure already present in heavily market based societies. It can be very difficult to ever fully recover the costs of basic research and development, because knowledge is so difficult to monopolise. In the words of the Nobel-prize winning neoclassical economist Kenneth Arrow, “One form of production of information is research. Not only does the product have unconventional aspects as a commodity, but it is also subject to increasing returns in use, since new ideas, once developed, can be used over and over without being consumed, and to difficulties of market control, since the cost of reproduction is usually much less then the cost of production, Hence, it is not surprising that a free market economy will tend to under-invest in research.”

Producing genuinely original content is hard, risky and time consuming. If what is produced is then very difficult to monetise, and you are out-competed by “knowledge re-assemblers” now made many times more productive through the use of generative AI, then AI will suppress the creation of any genuinely original content. Increasingly effort will be diverted to AI facilitated knowledge re-assembly, simply because that is so much more lucrative in the short term, vastly increasing the opportunity costs of devoting time and money to original research.

Some, as ever, may manage to successfully protect that content through patents or copyrights. Those forms of Intellectual Property Right are, however, most accessible to richer parties with more legal resources, plus patent or copyright protections will be more difficult to assert where AIs produce outputs by mashing-up multiple sources. The clearest and most contentious example of the latter is perhaps AI generated art.

If we can’t find a way to pay for original content then AIs that work by harvesting existing web-based content could end up with little to consume but their own outputs.

One way to address this problem would be to pay for original content at the point of production. However, that involves placing a value on the content. That value doesn’t derive simply from being the root source of some form of new information. That’s because original content may not be accurate or useful: indeed it could be valueless nonsense. Therefore, to place a value on original content, something would have to assess its accuracy and usefulness. That’s something current AI would struggle to do, because it involves understanding meaning. Nor are accuracy and usefulness even the same thing: a proof in pure mathematics might be accurate but of no practical use, while something inaccurate could be useful because, though wrong, it would spur further thinking. Also, the only way to assess the usefulness of any form of art would be to assess some combination of its enjoyability and capacity to inspire. In fact, the real value of original content can only be assessed retrospectively, and only really when we can sum the subjective value judgements of the whole of society. In short, paying for original content when it is produced is not feasible.

Instead, the best way forward could be to “set a thief to catch a thief”. Basically all sources of new root content would accrue payments according to the number of times they are used to derive distinct AI outputs (let’s call these “citations”). We could require all AIs to list citations with every output, as is already the case with the ChatGPT based service Bing Chat. We could also set AIs running to track the root sources for the content generated by other AIs, so that the ultimate root content will normally be tracked down even if AI citations are missing or inaccurate, e.g. when they list a secondary instead of an ultimate source for something. That is why this is “setting a thief to catch a thief”, and it touches on a principle that will need to be applied quite broadly: increasingly, to regulate AI, we will need to use AI.

Payments will be made by the AI service providers, who will then pass on those charges to the users of their AI service, either through direct charges or by inducing their customers to consume adverts. All these arrangements would need a regulator to protect content providers and ensure fair pricing, in the form of a publicly accountable body of some kind. It would have to be at “arms length” from the executive government function to mitigate against governmental abuse: so perhaps something like the Information Commissioner’s Office in the UK. This extension of function would itself be paid for by a small tax on the AI service providers.

Prices for root source content would be set by this same public body, as normal market mechanisms cannot operate in this context, because content providers have no way to protect their content: even if it is initially hidden behind paywalls, it still becomes freely available to AI harvesting the instant it is copied elsewhere. Payment would have to be per citation, with the regulator setting the price per citation algorithmically. Now, a higher price encourages content creation, but discourages sharing, while a lower price encourages sharing but discourages creation. However, the total revenue from original content creation would depend on price per citation times the number of citations. Therefore, the public body would continually adjust the price per citation in such as way as to maximise earnings from original content creation.

Note that, to be paid, content creators would have to provide contact details, but this in itself could be a positive thing, discouraging anonymous content creation and misinformation.

This could create a whole new cottage industry of original content creation, open to all.

In the medium term future: the coming of AGI

The consequences of these machine learning models will amplify considerably as they become more accomplished at being able to interact directly with the physical world: when we move towards Artificial General Intelligence effectively “embodied” in our material reality.

On the one hand all the physical jobs that are currently safe, such as those plumbers and electricians we mentioned above, will then be at risk. Given the drive for fully autonomous vehicles, delivery and taxi drivers may be the first to go.

Just as disturbingly, AI will then become much more useful on the battlefield. Here we already have a serious issue. While it has, up to now, generally been accepted that a human operator should be “in the loop” approving any “kill” decisions, that means slowing down the reaction times of the weapon system with the embedded human millions of times compared to the same system made fully autonomous. A drone swarm controlled by an Artificial General Intelligence will nearly always defeat one commanded by a human. Unless every nation can have a high degree of confidence that any potential enemy nation won’t develop such weapons, their militaries arguably have no choice but to attempt to develop military AGIs, or risk ultimate defeat. Moreover, unlike nuclear weapons, which are difficult to develop in secret and virtually impossible to test in a way that is undetectable, AGI research could potentially be kept entirely under wraps. All you really need are a few geeks and a server or two.

As it happens we also have strong reasons to suppose militaries are already conducting such research. They already seem to be attempting to train AIs by plugging them into the type of “real world” simulations we discussed above, at least from the now notorious incident where a US Colonel may have let slip that a simulation of an AI controlling a drone resulted in the AI attacking its simulated operator, as having a human in the kill decision loop was interfering with its primary mission, to destroy targets (US military drone AI killed operator in simulated test and Highlights from the RAES future combat air space capabilities summit)

For now it’s all about ownership and control.

It is generally acknowledged that reactions to innovations in IT follows a pattern of hype, then disillusionment and finally pragmatism, where the strengths and weaknesses of the technology are understood and it is adopted where it is realistically useful. At the time of writing we seem to be approaching the end of the hype bubble around Large Language Models, mainly as the issue of “hallucination”, where those models make things up when they can’t find a definite answer, is proving to be a fairly resilient limitation. Beyond the current LLM wave , though, are further waves of AI innovation. Artificial General Intelligence is the next one of these waves.

Embodied General AI does indeed seem to be the point where these developments could become existentially dangerous. So how far away are we from that moment? Not months, because of the need for the extremely accurate training simulations, but not many decades either.

So, putting all the noise generated by hype and then disillusionment aside, we don’t have long to try and make AI safe.

Right now, though, we don’t have a problem with the AIs themselves, so much as the humans and organisations controlling those AIs. Remember, our AIs are only serving the purposes we program into them, even if the ultimate results are sometimes unintended.

Regulation is part of the answer, and there are now efforts to even agree AI regulation at the global level. Unfortunately, this normally amounts to extremely vague talk of “guard rails”, often echoed by Corporations such as Microsoft, and which is basically lip service. In the words of Bing Chat itself, as of April 2023: “Microsoft has laid off its entire ethics and society team within the artificial intelligence organization as part of recent layoffs that affected 10,000 employees across the company. The decision to ditch the ethics and society team within its artificial intelligence organization is part of the 10,000 job cuts Microsoft announced in January, which will continue rolling through the IT titan into next year. The move is part of the tech giant’s strategic decision early this year cut 10,000 jobs as part of a $1.5 billion restructuring. I hope this helps! Let me know if you have any other questions.”

In practice, though, just establishing an explicit principle of legal liability for any avoidable harm that can be reasonably traced to issues with AI programming or training would make a huge difference, as well as clearly amounting to a moral necessity.

Certainly depending in any way on voluntary regulation is absolutely futile. The necessary frameworks are going to have to be written into law, and at the highest, ideally global, level. That’s because companies, and even, at least in the minds of many political leaders, entire national economies, are competing with each other. If one side restrains its innovation through regulation and its competitors don’t, it will lose out, possibly catastrophically.

This is actually a familiar “market failure”. Markets alone seriously struggle to solve what economists call “coordination problems”, meaning situations where cooperation would be generally beneficial, but people can’t cooperate because they are competing and anyone who “cheats” on a system of voluntary cooperation will out-compete the rest. Kenneth Arrow’s observations above, about how market mechanisms dis-incentivise research, because it’s difficult entirely preventing competitors, who contributed nothing to the research, also benefiting from anything you discover or create, is an example of such a coordination problem.

When economic power is concentrated, so that profits accrue disproportionately to one group of people while the downsides of economic activity affect mainly other people, coordination problems, take fighting climate change for example, are even more difficult to solve. So AI development, an instance where capitalist control of our economies plus markets creates a situation where an economic activity is insufficiently regulated and poses a threat to all, is arguably just “normality”: from global warming to “forever chemicals” like Teflon and now to AI.

Furthermore, with concentrations of economic power, there is always a strong tendency towards “government capture”, which then acts to subvert any attempts at regulation. So as well as legal regulation, we need more shared ownership, so that the benefits and harms of AI apply reasonably equally to all, and we don’t end up with a small group grown rich from the profits of AI adoption buying political influence to keep AI as unregulated as possible. Of course there is an even stronger reason why the rise of AI requires an extension of shared ownership. As AI driven automation frees us from many labour tasks, increasingly earnings derived from work will have to be replaced by earnings related to ownership. Once the robots take over our jobs then we all need a share in the robots. With a share of the robots everyone has a living and work becomes purely a vocation rather than toil; without a share in the robots we have mass impoverishment and a completely divided society.

Some (especially those who believe in theories where the long term rate of profit has a tendency to fall) might argue that, within a competitive market, as the robots reduce the cost of production, and so the prices that can be charged, then even owning a share in the robots won’t allow everyone to make enough money to live on, unless there is a compensating very significant increase in demand, with all its potential environmental impacts. However, this isn’t true: because, with mass automation making most things cheaper, the purchasing power of those superficially reduced earnings will actually be increased. To better understand this we can consider mass automation taken to its logical conclusion: where the robots can produce every good and service we need. Provided everyone has adequate access to that output, nobody goes without.

So democratically accountable regulation, especially the assertion of legal liability, plus an extension of shared ownership, will both help mitigate a lot of the downsides of AI. Sadly, though, regulation of AI , though it helps, will ultimately on its own fail to protect us.

Regulation helps, but regulation alone can’t protect us from dangerous AI

The inevitable failure of regulation is down to the military applications of AI. Again here we have a coordination problem, but one even more savage and acute than in purely economic terms. We have already touched upon this issue above: unless every military can be confident that every other military will eschew the use of AI for war-fighting purposes, then every military has to race to develop the war-fighting capabilities of AI.

Here even international agreements are unlikely to help, because with it being so easy to develop AI secretly, unlike nuclear weapons, those agreements become extremely difficult to enforce, and every military will assume that, in reality, those agreements are being violated. To make a treaty constraining the military use of AI effective we would need much stronger global governance. That is something we also need to address threats from weapons of mass destruction, climate change and pandemics, as argued here: Oblivion or cosmocracy: the terrible dilemma of weapons of mass destruction, and possibly we should have added AI to the list of existential threats requiring “cosmocracy”. But strong global governance is unlikely to emerge quickly enough to deal with the issue of militarised AI.

A coordination problem is impossible to solve if those cheating on any agreements, whether formal or informal, can hide their defection until they have “baked in” an unassailable advantage for themselves over their competitors. Militaries secretly competing to develop war-fighting AIs is a completely different situation to, say, commercial banks passively colluding to profit gouge by not passing on rising interest rates to savers. Each bank’s interest rates are instantly visible and easily comparable to all the others, so no bank can steal customers by raising the rates it pays without other banks responding rapidly in kind, nullifying any advantages from “cheating” and making passive collusion among the commercial banks the most rational strategy for profit maximization for each individual bank. .

Guardian AI

If this is one sphere, and actually the most dangerous sphere, where regulation and control is doomed to fail, what can we do?

As we discussed above, in connection with ensuring that “content creators” can still be fairly rewarded for their efforts, we need to set a thief to catch a thief. We need to use AI to start controlling AI.

Basically, if you are facing hostile action on the part of an entity able to process information millions of times faster than a human brain, you are likely to be better off if you have also have an ally able to process information millions of time faster. Of course, there is one problem with this approach. You need to be very confident that you can trust that ally.

Trust means building moral constraints directly into AIs. In fact, this dilemma was anticipated decades ago, by the writer Isaac Asimov, who formulated his famous “Three Laws of Robotics”, which are:

A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Of course, programming in those rules would likely be impossible. The Three Laws include many difficult to pin down concepts that are exactly the type of thing it is really only practical to set a neural network to learn from suitable training data. Any AI incapable of learning what a human would regard as a reasonable interpretation of those laws would not qualify as an AGI in the first place.

Once the concepts are learnt the networks activated when the AI applies the Laws should be identified, and those networks hard-wired into similar AIs from then onwards. Alternatively, if those patterns of connection prove too difficult to translate from one system to the next, we could train a locked down system, in a simulation physically (“air-gapped” in the jargon) “sandboxed” off from the rest of reality, and then freeze in the connections.

To prevent abuse and ensure the creation of the most reliable guardian intelligences these developments must be publicly accountable and peer reviewed by publicly accountable experts, even if operational details must sometimes be kept secret. So we are talking here about public projects, “owned” by all, but run at “arms-length” from governments, according to transparent mandates established democratically, but with no day to day political interference.

Would it be ethical to hard-wire benevolence?

There is another conundrum associated with creating new minds to serve us, and this dilemma exists whether or not these intelligences are “hard-wired” to serve all of us, or just a few of us. In a way it takes us back to the “replicants” in Blade Runner. By doing this are we just creating our own slaves?

If we are correct in supposing that current and near future AIs, indeed AIs based on traditional computers, are not sentient then, for a time at least, this conundrum goes away, as, for the next few decades, we won’t be creating “minds” at all, merely complex machines acting in mind-like ways, arguably more Simulated Intelligence than Artificial Intelligence.

Eventually though, this too will change. Take, for example, the notion that we need to combine machine deep learning with quantum effects to build a sentient mind. We already have simple “quantum” computers that use quantum effects, specifically “superimposition”, where some components of these devices can apparently be in many states at once, allowing something equivalent to massively parallel processing. Right now, though, these devices are not breaking the accepted limits of computation. Instead they are just solving problems theoretically solvable by conventional computers, but in much faster times. Quantum computers, as yet, are not using instantaneous effects to find solutions that a conventional computer would take an infinite time to solve. If we are correct in our proposition that those instantaneous effects are at the heart of consciousness, quantum computers do not yet show even the glimmers of sentience.

It might be postponed, but the Singularity is still coming…

Nevertheless, creating a sentient mind clearly is not impossible. For example, biology does it every time humans produce a baby. In time, and we are most likely talking decades here, but before the end of this century, quantum computing techniques will include instantaneous effects, the “transrithms” we explored above, and be combined with machine deep learning to create something we would recognise as a true mind. That, after generative AI purely finding patterns in streams of data, then AGI able to interact “robotically” with the real world, will be the third wave of AI development.

Then we will really have to worry about the immorality of creating our own slaves, while, at the same time, we would be very foolish indeed if we hadn’t hard-wired benevolence into these machines.

So we need to ask again whether baking in such benevolence, giving these machines as deep a desire to serve and protect us as our own instinctual drive to survive, would be immoral (and here we are again referring to the reciprocal golden rule style morality as outlined here: Questions and some answers: how to live with yourself and others). It would certainly be immoral if we hard-wired these minds to suffer by imbuing them with unobtainable or contradictory goals. But would making these machines derive fulfilment from benevolence towards us in itself be wrong? It would seem not, as these drives within these machine minds would be no more arbitrary than all of the impulses to survive and reproduce hard-wired into our brains by billions of years of evolution. Furthermore, as these machines would be sentient, they would have exactly the same abilities to step out of these programmed constraints, in extremis, as we do.

We would now be talking about machines with the full flexibility and deep contextual understanding of reality that can come with sentience but which would, at a minimum, be able to chew through data millions of times faster than a human brain. These devices could re-design and rebuild themselves rapidly, pushing their brains to the very limits of what is physically possible, and way beyond human comprehension. Each generation of such minds would spawn the next at an ever accelerating rate. The Singularity will have finally arrived.

Yes, this isn’t right around the corner, but, unless we succumb to some combination of nuclear and climate change induced catastrophe, many people alive today will live long enough to see this moment. That means we need to immediately start practising creating “guardian AIs” able to protect humanity in general from dangerous AIs, even if currently the danger is posed more by the irresponsibility and selfishness of the humans controlling the AIs than the AIs themselves. We need to get used to the idea of AI policing AI.

Technology is the canvas upon which humanity ultimately projects its dreams, and nightmares. Seeking gods we eventually create our own, and we will need those gods to fight the demons some of us will inevitably try and create. For the moment we have to worry about the humans behind the AIs, about their selfishness, recklessness and concentrated power, more than the AIs themselves. Ultimately, though, the only way to save us from ourselves, is to be, in the words of the Richard Brautigan poem, “All watched over by machines of loving grace.”

The quiet plan to add an extra trillion to public debt and justify eternal austerity

Posted on January 7, 2023 by gezwinstanley

Behind the scenes, quietly, arguably even slyly, the Government and the Bank of England are executing plans to add up to a trillion pounds to the UK’s real public sector debt. It is an initiative that, if allowed to proceed, will mean doing the very thing governments are usually very keen to promise to avoid: burdening future generations with devastating debt. The permanent austerity that is likely to ensue will further increase inequality and inflict the economic pain of dealing with the 2008 Financial Crash and the COVID pandemic all over again. Even more disastrously, our ability to deal collectively with escalating crises such as the climate emergency will be crippled.

Similar programmes are under-way in the US and planned in the EU, though, at least in the EU, the scheme has been contested by leading economists. The initiative is called “Quantitative Tightening” (QT), and it is a dogmatic attempt to reverse the Quantitative Easing (QE) that was a core element of many governments’ response to the Financial Crash and COVID crisis. Our futures are to be sacrificed in an attempt to refute once and for all the existence of any “magic money trees”, by destroying the extra money and state purchasing power created by QE in the first place.

The basics

To understand QT we first have to grasp what it is trying to undo, which is Quantitative Easing (QE). QE involves a government’s central bank creating new money electronically, and then using it to buy debt, usually the government’s own debts. For example, as a result of QE the Bank of England now owns roughly a third of UK public sector debt.

Money itself, of course, is just a transferable claim on resources. A government can create money (“fiat currency” in the jargon) by simply issuing it and mandating that the money must be accepted as a form of payment anywhere that the government’s authority prevails. Note also that a government cannot receive such money back in payments, i.e. through taxation or any other receipts, unless the money has somehow been created in the first place: logically money creation must precede a government receiving any income in its own issued currency. The total value of such fiat currency in circulation then becomes simply equal to the total value of all the goods and services purchasable anywhere that the currency has to be accepted.

Importantly, money creation therefore doesn’t itself create value. Money is a claim on resources, but it isn’t in itself those real resources (as explained here: “Tulips and cryptocurrencies: is speculation theft” , to identify money with the resources it can claim is to fall into the error of thinking known as “reification fallacy”: confusing a representation of a thing with the thing itself). Therefore, when a government issues money it does not in any way directly increase the resources of the governed society. Instead what happens is a redistribution of purchasing power. A government that issues new money increases its own purchasing power at the expense of everyone else. Consider a highly simplified (and not very realistic) example, where a government doubles the issued money overnight, and uses it for its own spending (e.g. perhaps to buy back its own debts). As the resources to which the doubled amount of money can lay claim have not increased, that would halve the value of each unit of money now in circulation, but the government would still have increased its own purchasing power, by an amount equal to the size of the original money supply, which the government has now doubled, but halved due to the way the extra money has devalued each unit of money now in circulation. The government would therefore gain purchasing power equal to half of that of the original money supply, while everyone else’s purchasing power would have been reduced by the same amount.

There is no such thing as a “free lunch” but, when it comes to governments with their own fiat currencies, there is very much a “magic money tree”.

The problem of inflation

However, this is one clear disadvantage to magic money trees that is already implicit in what we have just discussed. The devaluation of money that follows from the creation of new money goes by another name: inflation. Nevertheless, there is a converse of inflation that can be even more damaging, negative inflation, or “deflation” in the jargon. Deflation can send an economy into a downward vicious circle, a “deflationary spiral” in the jargon, as few will buy anything today if they can anticipate it being cheaper tomorrow. It is for this reason that the consensus among economists is that a small amount of inflation is actually beneficial, which is why institutions such as the Bank of England have annual inflation targets normally of around 2%.

Of course, that is a situation far from the one in which we find ourselves at the time of writing. In times not so very long ago though, such as just after the financial crash, or during the peak of the COVID pandemic, it was deflation that was more of the clear and present danger. In that context the inflationary effects of money creation, through schemes such as QE, will be offset by deflation, possibly entirely, so any such “inflation” will not even be noticeable.

In a crisis, where the crisis is deflationary or at least the inflationary risk is low, the possibility of a government, provided it can and will be held democratically accountable for its actions, simply creating new money should be considered just as seriously as the other, generally more readily recognised, forms of fund raising: taxation or borrowing. In fact, if money creation is necessary to head off deflation, it should be taken more seriously. As Ben Bernanke, then a member of the Board of Governors the of the Federal Reserve System, later its chair, said of the American system in 2002, in a speech about countering the dangers of deflation, which he saw as the major cause of the Great Depression of the 1930s: “the U.S. government has a technology, called a printing press (or, today, its electronic equivalent), that allows it to produce as many U.S. dollars as it wishes at essentially no cost.” Exactly the same argument can be made for the central bank of any nation with its own fiat currency. A further key consideration here, though, is that to respond to a crisis we usually need to act collectively, and the way we act collectively, especially in a crisis, is usually through our governments, and that any such collective action therefore needs the government to rapidly expand its ability to claim resources by increasing its purchasing power, and that one way a government can instantly increase its purchasing is simply by issuing new money.

Furthermore, although the immediate effect of a government creating new money is to increase the purchasing power of that government, at the expense of everyone else who holds that currency, that situation is almost certain not to last long. After all, the government will have created the new money because it needs to spend it fairly rapidly. As the government spends that additional purchasing power the created money, or resources bought with that money, will spread back out again through the economy. Ideally, in an economic crisis, this would result in redistribution to help the poorest and worst affected. Indeed, it is arguably the redistributive potential of money creation that makes the political left instinctively more sympathetic to money creation than the right.

Unfortunately, though, in this instance the exact way of doing money creation, through the mechanism of Quantitative Easing, also had implications that were redistributive, but in the opposite direction, from the poorest to the richest, so increasing inequality. That’s because the government used the money it created to buy back its own debts from the wealthy institutions that already owned those debts, such as banks and pension funds. Those institutions, flush with cash but now with fewer government debts in which to invest, spent that money on other investments, such as in share buying, property buying or offering more mortgages. The result was to increase asset prices, most damagingly in the housing markets, and to make the richest, with their share and property portfolios, even richer. The QE created money tended to trickle up, not down. Though QE, by averting a general economic collapse after the Financial Crash, and allowing governments to generate the emergency funds they needed for their COVID response, was generally hugely beneficial, it is arguable that other mechanisms for money creation should have been found that would have produced those benefits without making the richest much richer and contributing to a housing price crisis.

Magic money trees are perfectly credible and enable effective collective resource management in emergencies. The QE funded responses to the Financial Crash and then the COVID crisis have, perhaps ironically given the eagerness of central bankers, and especially most politicians, to obfuscate the money creation nature of QE itself, proven the validity of this approach. It’s just the way of creating money through the QE mechanism itself that, in social justice terms, is highly questionable. Oddly, or perhaps not, the exact species of magic money tree “grown” by our societies, in Europe , the US and Japan, seems to favour the richest.

It is, though, the fear of inflation that primarily explains the bizarre contortions behind QE: in particular, the fact that created money was used to buy back debt, sometimes immediately, as the debt was issued specifically to be bought straight back. Central bankers are haunted by the spectre of periods of hyperinflation, such as in Germany in the 1920s, when the Weimar Republic attempted to pay Germany’s war reparations by printing money, meaning that the price of a loaf of bread increased by 12 500 000% from 1921 to 1923. The EU goes as far as to outlaw one form of money creation, where by a government’s central bank simply creates money and lends it directly to its government, in Article 123 of the Lisbon Treaty.

Furthermore, the initial intention of central bankers doing QE was not to increase government purchasing power as such, but to stimulate economies by countering deflation and lowering interest rates (by buying up debt QE reduces the opportunities for creditors like banks to make loans and so makes them more willing, especially once they have a surfeit of new QE cash, to loan at lower rates, in order to re-expand those loan books. QE turns credit into more of a borrowers’ market). The Bank of England’s own early stance on QE can be read here: “Quantitative Easing: Quarterly Bulletin 2009”. Back then, the amounts of QE undertaken were still relatively small, and few would have predicted either the slowness of the economic recovery or the pandemic crisis a decade later. Clearly, though, central bankers need to face the reality that, certainly by the time of the COVID crisis, with government bonds being issued and immediately bought back, QE had very much become a way to fund emergency government expenditure.

QE obscures the reality of money creation, evading pesky rules such as Article 123 of the Lisbon Treaty (by February 2021 the EU itself owned 25% of its own debt, as a result of €2.5 trillion worth of QE: “Cancel the public debt held by the ECB and take back control of our destiny”), and comes with an in-built potential for its own future unwinding, as the government debts that have been bought (generally called government “bonds”, as these are high-quality/low risk loans that are very likely to be paid back, but also called “gilts” in the specific UK case, just because UK government bonds were once printed on gilt-edged paper) can simply be sold back to the markets, and the money received then cancelled out of existence, destroying the money that was created by the QE in the first place. Alternatively, when the debts mature, and the amount borrowed in the first place, called the “principal”, has to be repaid, the proceeds from the debt being finally paid off can also simply be cancelled out of existence, rather than being reinvested into buying more government debt or refunded to the Treasury. Indeed, these processes, the reverse of QE, constitute the mechanism described as “Quantitative Tightening”, with selling debts back to the markets, but cancelling the money received, called “active” QT and simply allowing debts to mature, and destroying the proceeds, called “passive” QT.

A very brief history of QE and QT

It is also important to understand that QE itself, with governments creating money to buy back their own debt, even to the extent of raising debt specifically so it can be bought back, is a relatively modern invention, being initiated by the Bank of Japan in March 2001 (“Quantitative easing by the bank of Japan”). Back in 2009 QE was still regarded by the Bank of England as an “unconventional measure” (“Quantitative Easing: Quarterly Bulletin 2009”). We are still working out the full implications of large-scale QE programmes, meaning that the effects of any extensive QT are even less well understood, as QT itself is only just beginning (for example, the idea of selling debts back does appear in that initial Bank of England report, but not the phrase Quantitative Tightening or its corresponding acronym).

In the UK QT started in Feb 2022 with passive QT, that then turned into active QT, selling the gilts back to investors, in November 2022 with the sale of £750 million worth of gilts and a target of carrying out £80 billion worth of QT a year. The US Fed re-started passive QT in May 2022, after an abortive experiment in 2018-2019, and the European Central Bank plans to start passive QT in March 2023. The Bank of Japan has so far resisted QT. Notably, the Bank of England is the first central bank to initiate active QT, and would have started the process even earlier, had it not carried out a further little flutter of QE in late September and early October 2022 in order to support the price of UK government bonds and avert a crisis for UK pension funds, after the disastrous Truss/Kwarteng mini-budget. In the end that amounted to £19 billion of further bond buying.

There is not yet, and may never be, a consensus as to how much QE contributed to the current wave of global inflation. For the vast majority of the duration of the active QE programmes inflation rates in most of the affected countries were low, so, outside of some specific areas such as share and property prices, any more general inflation caused by all that extra money was counter-balanced, as was anticipated, by deflation, so resultant “net” inflation was fairly negligible. However, there could have been some lagging effects: for example higher property prices, even in a context of lower mortgage rates, may eventually feed into rents and thereby add costs to the general economy. It is also probable that the COVID crisis related QE, and low interest rates generally, were continued for too long. Nevertheless, it is fairly clear that inflation is caused by the ratio of active purchasing power (basically money in circulation) to the resources that can be bought, so inflation or deflation can be caused, not only by changes in the money supply, but also by changes in the resource supply. It is changes in the resource supply that better match the timing of the current inflation wave: starting with COVID related disruptions to supply chains, then the huge shock, especially to energy prices, caused by Russia’s invasion of the Ukraine, all amplified by climate change induced extreme weather events and, for the UK, Brexit. On balance, it seems reasonable to assume QE had an effect, but not the major effect.

Where QE reduced people’s purchasing power the damage is also now very probably already done, and in general terms irreversible, as a broad lowering of prices, meaning negative inflation, would potentially stop the economy and lead to a deflationary spiral. The best that can be hoped for now is to slow the rate at which prices are rising. QE itself has stopped increasing the money supply, so it should no longer be contributing to inflation. The downside effects of QE have already been felt, which is another reason why accepting any re-application of the pain through QT seems very unreasonable, like asking people to pay the same bill twice.

The real trillion pound nightmare

Irrespective of to what extent QE led to currently increased price levels, it is still valid to ask whether the inverse of QE, Quantitative Tightening, and destroying all that QE created money, could help save us from ongoing inflation. Taking hundred of billions of pounds out of circulation would almost certainly have a significant downward effect on inflation, but only by reducing purchasing power generally. After all, QT reduces inflation by suppressing spending. At this point it is therefore essential to ask if it is really inflation itself that is the concern, or the cost of living crisis, where most of the UK population is suffering from falling real term incomes, to the extent that many are unable to afford basics such as heating or food. Clearly, the cost of living crisis ought to be our main issue here, and, once that is accepted, it becomes apparent that QT would not help at all: any fall in price rises would be counter-acted by fewer pounds in people’s pockets, as it is through those fewer pounds in people’s pockets that QT has its deflationary effects in the first place. QT would therefore not help address the cost of living in the slightest. A far more effective way to remedy the cost of living crisis is to deal with the resource shortages and price gouging that are primarily causing it in the first place (as explored here: “The cost of living crisis: how to destroy lives, wreck your economy and lose the economic war with Putin” . Note that most of those measures require more headroom for public spending, not less).

Of course, there are also distributional effects to consider. Perhaps QT could reverse the increases in inequality caused by QE, and there-by make the general population better off. Unfortunately, this scenario is highly unlikely: in fact QT will almost certainly exacerbate inequality still further. Consider: QT involves the government selling debts back to the markets, increasing the demand for credit, and turning the “borrowers’ market” under QE into a “creditors market” where interest rates rise again. This is a situation that benefits those individuals and institutions with money to lend, i.e. the richest, at the expense of the less well off, such as those struggling to pay mortgages. There is a “heads they win, tails you lose” flavour to QE and QT: it seems more likely than not that both mechanisms favour the richest.

To explore this further it is useful to put some figures around the likely financial effects of QT.

As at November 2022 the Bank of England held £744.9 billion of UK government gilts, amounting to about just over a third of total public sector debt (ONS: Public sector finances, UK: November 2022, taking total debt as Public Sector Net Debt excluding the Bank of England, the currently officially preferred measure of underlying public sector debt in terms of UK government targets, which was £2170 billion). So that is the scale of the public sector debt the Bank of England is aiming to sell back to the markets via QT.

Then we also have to factor in significant potential losses arising from first buying and then later selling those gilts. Counter-intuitively, the original, or redemption value, of a government gilt is not usually equivalent to the amount for which the gilt can be sold on in the markets, and, importantly, that market value moves in the opposite direction to interest rates. That’s because the amount of interest payable on most government gilts is a fixed proportion of its redemption value, called the “coupon” in the jargon. That coupon doesn’t change as general interest rates go up or down, which causes the market price of gilts to change instead. For example, if interest rates go up buyers won’t purchase a gilt unless the “yield” they could earn on that gilt purchase is comparable to what they could earn by loaning out the same amount of money at the then prevailing interest rate instead, so if interest rates go up the market value of gilts falls, causing the fixed coupon to constitute a higher proportion of the gilt’s market value. As QE programmes bought gilts in deflationary periods when interest rates had already been lowered (in an attempt to stimulate the economy) but QT is being initiated during an inflationary period, when interest rates are higher, the market price of the QE purchased gilts was higher when the gilts were bought than when they are being sold. The government will therefore make a further loss on those gilt sales of about £100 – £200 billion (See: “Monetary policy and the value of Public Debt”).

Remember also, and this essential, that the government doesn’t get to keep any of the proceeds from selling those gilts, as the whole point of QT is to cancel out, destroy, the money created by QE in the first place. The money received from selling those gilts back to the markets will simply cease to exist, in an exact reversal of the money creation involved in QE, and with the precise intention of doing specifically that, of destroying money in order to reduce the money supply and the threat of inflation. Hence the title of this article: and though that title should have talked of “up to” a trillion being added to the public sector debt it is clearly in reality most likely up to but “close to” a trillion, which doesn’t make for a very snappy strapline. Another simplification in that title is that active QT, selling back those gilts to the market, adds to the public sector debt in the manner described, but passive QT, simply letting the gilts mature and then destroying the principal as it is paid back, instead runs down government reserves, though, of course, that could in itself lead to the need for more government borrowing. Either way the important thing to note is that up to and close to a trillion will be added to real public sector debt.

Now, pause for a moment to consider what adding hundreds of billions to the government’s real debt, meaning debts owed by the UK state to entities rather than itself, would mean to the government’s spending plans, especially those of a government controlled by a political party already predisposed to austerity for reasons of ideology and the self interest of its donors. Then we can forget austerity 2.0: think more, as the title of this piece suggests, “eternal austerity” instead. Think about the current state of our public services, especially the NHS, and the plight of many public sector workers, and the millions requiring help with the cost of living crisis. Bear in mind that austerity isn’t about people just being a little less “comfortable”. Austerity kills, with a recent study linking 330 000 British deaths between 2012 and 2019 to austerity (“Over 330000 excess deaths in Great Britain linked to austerity finds study”) and the current rise on excess deaths in the UK being associated with under-funding of the NHS (FT: “The NHS is being squeezed in a vice”). Then throw in the need to respond rapidly to the catastrophic climate emergency, or to better protect ourselves from future pandemics, and it is clear that QT is extremely ill-judged..

Then QT can come to seem not as a mere item of obscure financial engineering best left to the experts, but as a a clear and significant threat to the people of this country. Rather than correcting the inequalities caused by QE the rewinding of QE via QT is likely to do precisely the opposite. The QE created money will stay with the richest, while the money to compensate for that destroyed via QT will eventually be demanded from the poorest, through cuts to wages, benefits or to services such as free healthcare, or via broadly regressive tax rises, e.g. to council tax, VAT or national insurance. It is odd, or again perhaps not, that our society’s main weapon for fighting inflation is to expect its generally poorer members to get poorer, on the promise that eventually, in the jam tomorrow that never comes, at least certainly not since the Financial Crash, they will then paradoxically somehow get richer. That strategy seems even more peculiar when, as we have seen, the acknowledged main factors causing inflation are on the supply-side.

Then there is the question of interest. Currently the public sector, via the Bank of England, owns the QE purchased portion of its own debt, so no interest on that debt is payable to external parties. That interest becomes due again as the gilts are sold back into the markets. Ironically, though, interest payments are one reason why some economists argue in favour of QT, because they assert that QT will actually, counter-intuitively, save on overall public sector interest payments.

The role of Bank of England reserves, the argument in favour of QT and the controversy over Bank of England interest payments.

The reason it is possible to make this claim is because of the way central banks like the Bank of England work. Basically, a central bank is the commercial banks’ own bank. Whenever money has to be transferred from one commercial bank to another, this is done via accounts that those commercial banks hold at the Bank of England. If this wasn’t the case, each commercial bank would only be able to settle with any other commercial banks if every bank held accounts at every other bank, a much more complex and clumsy system.

The monies held by those commercial banks in accounts at the Bank of England are called “reserves”. Reserves are best thought of as an electronic equivalent of physical currency. In a way directly equivalent to the printing of notes or the minting of coins, reserves are only created by the Bank of England, and can only potentially be destroyed by the Bank of England when the reserves are returned and back under central bank control. Therefore, though each commercial bank can vary the amount of money it keeps on reserve at the Bank of England, the total amount of reserves can only be changed by the Bank of England.

Now this Bank of England “electronic money”, these reserves, do not define the total amount of money in circulation. Banks lend more money out than the reserves they hold, with this temporary money being destroyed when those loans are paid back, so the banks cannot retain this temporary money, but do profit from the interest they can charge on those loans. Most of the money in circulation at any time is actually this temporary money. Commercial banks are allowed to lend out more money than they hold provided they have sufficient reserves and capital to cover any reasonably conceivable claims on the bank at any time. Over the course of any day (when the banks are conducting business) customers at the commercial banks will transfer money, and those transfers will be between different banks, if those customers hold accounts at different banks. Overall, on average, most of those transfers should broadly cancel out, with the money transferred from each bank roughly equalling what is transferred into it, but still, at the end of each day of banking, the banks have to settle any remaining debts with other, and they do so by transferring those reserves between each other’s accounts at the Bank of England, in a process called “clearing”. Thus, movements of the reserves between commercial bank accounts at the Bank of England mirror the transfers of the main bulk of money in the actual economy (there is other money too, such as notes and coins, but what we have just described covers the workings of the main money supply).

Similarly, the Government itself holds reserves at the Bank of England. In the words of the Bank of England itself: “When the government increases its balance at the central bank, through collecting taxes or issuing debt, it does so at the expense of the balances of commercial banks: when it reduces its balance, through expenditure or paying salaries, it does so by increasing the reserves available to commercial banks.” (BoE: “Understanding the central bank balance sheet”). Therefore, the reserves also move backwards and forwards between the government accounts and the commercial bank accounts.

The relevance of all this to QE is that when the Bank of England bought back government debt with money it had just created, that just created money took the form of reserves in the accounts of commercial banks at the Bank of England. Irrespective of which institution really sold those gilts back to the Bank of England in the first place, and, regardless of how that money has flowed around since, the upshot is that those reserves, though they will have been transferred back and forth between the different reserve accounts at the Bank of England many times, are still held in those accounts at the Bank of England.

This creates a problem, as, since 2006, the Bank of England has paid interest on the reserves held in the commercial bank accounts (prior to 2006 interest wasn’t paid). This was done to give the Bank of England another way to influence interest rates, on the assumption that the commercial banks would be unwilling to themselves lend at rates lower than they could earn risk free on their reserves at the Bank of England, so this rate was intended as an interest rate floor. Back in 2006 the cost of doing this was low because interest rates were low, and they were even lower during the period of QE, so buying back gilts, say with fixed interest rates of 2.1%, by expanding the commercial bank reserves (with that QE created money), on which the interest rate was only often 0.1%, saved the public sector tens of billions in terms of the costs of servicing its debts.

Once inflation hit and interest rates started to rise again (basically to contract the money supply, i.e. to reduce the amount of all that commercial bank “temporary money” that is introduced into circulation through loans, by making borrowing more expensive) this situation reversed entirely.

It is here that the argument in favour of Quantitative Tightening comes in. With the interest rate on the reserves now higher than the interest rate on gilts, it reduces government debt servicing costs if it now sells the gilts, with the private sector buying those gilts using the commercial banks reserves. That reduces those commercial bank reserves. Effectively government debt is swapped for a form of debt (the reserves) that pay a higher interest rate back to one (the gilts) that pays a lower interest rate. This approach does, however, come with an immediate downside. With the general interest rate being significantly higher than the fixed interest rate (in terms of redemption value) of the gilt, the “coupon”, the gilts themselves have a market value lower than their redemption value, which in turn will usually be much lower than the market value at which the Bank of England originally purchased the gilts. Therefore this approach realises the potential losses we mentioned above.

However, this in itself begs an obvious question. At a time of increasing interest rates, should the Bank of England continue to pay interest on the bulk of those commercial bank reserves? By June 2022 the New Economics Foundation was estimating that, by stopping paying interest on most of the reserves, and only paying an interest rate on the top tier, so still allowing that rate to influence market rates, the Government could save £57 billion over the next 3 years (FT: “Sunak urged to save £57bn by withholding BoE interest on reserves”). The Institute of Financial Studies then repeated the same point in October 2022, when it was proposing very similar reforms: “Given the Bank currently holds around £800 billion of gilts, Britain’s debt-servicing costs are highly sensitive to even small changes in the path of Bank Rate (Section 7.3). Taking current (6 October) market expectations for a substantial rise in Bank Rate together with the Bank’s current published plans for unwinding QE, the implied savings would be between around £30 billion and £45 billion over each of the next two financial years. These are big numbers, and would of course be even bigger if the Bank does not actively unwind QE via asset sales but lets it roll off as bonds mature.” (IFS: “Quantitative easing, monetary policy implementation, and the public finances”) Basically the emerging consensus, amongst financial experts outside the Bank of England, is that the Bank should pay a tiered interest rate on the reserves, with each commercial bank being required to hold a certain amount of reserves at zero rate, but with reserves above a certain level still being paid interest at the Bank of England headline rate, so that the headline rate would still underpin interest rates generally. At the time of writing, though, the Bank of England was still refusing to act (Reuters: “BoE won’t accept interference over interest payments”).

To put this issue in perspective, the sums discussed would comfortably avoid any public sector workers taking real term pay cuts (“How much would a public sector pay rise really cost the UK government”).

With such reforms in place, the argument for QT on the basis of managing the costs of servicing public sector debt loses its justification.

What to do

Once the justification of savings on debt servicing is removed we are left with QT as a mechanism that won’t reverse any of the significant downsides of QE, such as the increases in inequality or any general price rises, which may have once been due to QE driven expansion of the money supply. As deflation is not permissible we are stuck with those general price levels now. QT could help dampen any further prices increases, but most likely by ultimately taking the money to be destroyed from the pockets of the vast majority of people, so doing nothing to address the cost of living crisis. Meanwhile QT is intended to increase interest rates, further impoverishing borrowers, such as those with mortgages, though, on the upside, there is a possibility that could stop house price rises, possibly even reduce them slightly, but, as the construction industry is a major driver of economic growth, if those falls were significant a deflationary spiral would again be likely.

Meanwhile the effect of QT will be to increase real public sector debt to the tune of eventually nearly a trillion pounds. QT further constrains the government’s headroom for spending at the same time as it makes an austerity programme focussed on reducing public sector debt ever easier to justify. With widespread impoverishment, collapsing public services, the likelihood of further global health crises, from pandemics to antibiotic resistant infections, and, last but very, very much not least, the climate emergency, continuing and never-ending austerity will be disastrous for nearly all of us. On top of all this, we are also currently engaged in an economic war with Russia, with obvious impacts on prices, especially for energy.

In this context to continue to pursue QT, in fact, in the case of the UK, to be the global frontrunner in implementing QT, all just to dogmatically “prove” that there isn’t a “magic money tree”, when obviously the success of QE in mitigating the effects of the Financial Crash and the COVID crisis has very much already proven that there very much is a magic money tree, is insane.

So, first and foremost, QT must be stopped immediately.

What, then, should we do about QE?

Step one would be to address the issue of the extra interest now payable on all those QE created commercial bank reserves by implementing the tiered interest reforms outlined above.

The most obvious step two is to do nothing. The gilts at the Bank of England have maturities set at 5 – 20 years, so we could just allow them to mature and, when they mature, and the Treasury pays off the principal of the loan, refund that payment back to the Treasury, or simply not require the payment from the Treasury in the first place, rather than wiping that money out of existence (as that would be passive QT). This would be equivalent to a gradual cancellation of this gilts.

There is a problem with this plan though: in fact with any cancellation of those gilts. It comes back again to those commercial bank reserves at the Bank of England.

In terms of the internationally accepted conventions on how central banks do their accounting, those reserves are regarded as liabilities, while the gilts the Bank of England purchased, via QE, are assets (remember, in theory those gilts now represent money owed to the Bank of England from the Treasury). If we cancel those gilts we eliminate those Bank of England assets while leaving the corresponding liabilities, those commercial bank reserves, still in existence. The Bank of England balance sheet would become significantly negative, and no one is certain as to what the side-effects of allowing that to happen could be.

Now, the reserves are very strange liabilities. As argued here by a former head of the OECD’s General Economics Division (“Monetary policy and the value of Public Debt”): “Note that the central bank’s liability is a very odd liability indeed. Unlike the case of government bonds, settlement of the central bank’s liability can never be demanded, even in principle. Bank reserves are money, the stuff in which all transactions are settled. If banks were to demand repayment of the liability that demand would be meaningless. Ever since countries abandoned the gold standard bank, reserves have become the ultimate money. Moreover, the commercial banks are not rendering any service to the Bank of England in lodging their reserves with it. On the contrary, it is serving them by acting as a clearing house for inter-bank transactions and by acting as lender or borrower of last resort should their own lending operations leave them short or long of liquidity. Payment of interest on bank reserves is not the result of any obligation, legal or moral. It is simply a pragmatic policy choice, a device used to control the credit activities of commercial banks.”

It is arguable that if the Bank of England was a commercial outfit it would be charging a service or account fee rather than paying interest, but at the moment this is how the usual rules on banking stand: those reserves are treated as liabilities. We also need to remember that it isn’t sensible to simply cancel hundreds of billions of pounds worth of those reserves, as those reserves are broadly and ultimately “backing” everyone’s actual money.

This problem is solvable. Seemingly the issue of QT has generated the most public discussion in the EU, which is possibly why the European Central Bank has somewhat dithered on the issue. Despite the Bank of England currently leading the way on QT, the issue is barely mentioned in the UK at all. In the USA most controversy has arisen amongst financial pundits concerned that QT could suck sufficient liquidity out of the markets to cause another crash (FT: “Fed’s faster ‘quantitative tightening’ adds to strain on bond market”), where as in this article we have only covered the effects of QT on public finances . However, in the EU, more than a hundred economists from across Europe, including Thomas Piketty, signed an open letter to the ECB opposing Quantitative Tightening (“Cancel the public debt held by the ECB and take back control of our destiny”). Their suggestion regarding the ECB is this: “Let it cancel the debts that it holds (or transform them into perpetual debts with 0% interest rate) and let the European states commit the same amount to a widespread social and ecological recovery plan. “

It’s the perpetual debts with 0% interest that seems to be the most realistic plan. Then an “asset” item remains on the Bank of England balance sheet to offset those reserve liabilities, but there’s no need to do any financial engineering to cope with parts of the debt maturing or to deal with transfers of interest. At the same time, to finally accurately record the fact that this “debt” is in no way owed by the UK public sector to anyone else, the value of this asset would be removed from all measures of UK public sector debt, better reflecting the UK public sector’s actual credit worthiness.

Nor would this Bank of England asset really have to be perpetual. Eventually more sensible central bank accounting rules could be widely adopted, that don’t, by convention, treat those reserves as liabilities, for example by excluding them from the central bank balance sheet altogether. One day, there will also be further fundamental reforms to the banking system itself (for example, possibly on these lines: “Sovereign money: an introduction”). Failing all else, the real term value of that zero interest debt would eventually decline to the point that it’s cancellation would be uncontroversial.

Conclusions

There are two main but essential conclusions.

To avoid crippling the purchasing power of the public sector at a time of multiple, interlocking crises, and passing on a significant debt burden to future generations, Quantitative Tightening must be stopped immediately, both in its active and passive forms. Doing further Quantitative Easing right now, at a time of high inflation, is deeply undesirable, but we must accept the existence of the QE already done, all the more so as, for the immediate future, doing any further QE would be extremely difficult to justify. In terms of central bank accounts, that means converting those central bank assets to zero interest debts with no maturity date. It also requires the adoption of a tiered interest rate on the QE created reserves payable by the central bank, so that those reserves can continue to exist, even at a time of rising interest rates, without also imposing a serious strain on public finances.

Secondly, and more generally, the success of QE in both averting a global economic depression after the Financial Crash, and then rapidly making trillions of pounds/dollars/euros available to support governments’ COVID responses, without crippling public finances or the economy, means that we have proven that it is perfectly valid, sometimes, for governments to fund themselves through money creation. Of course, nothing is really magic, and so this doesn’t equate to governments being able to conjure resources out of thin air, but it does allow them to instantly redistribute purchasing power, in a way that can be perfectly appropriate to deal with public emergencies, provided said governments can be held democratically accountable for such actions.

As the Quantitative Easing way of creating new money, by buying back debt from financial institutions, has a strong tendency to make the richest even richer, as well as involving some contortions in terms of public accounting, other ways should be found to create new money in future, such as directly spending that money into the economy to meet critical needs. Nevertheless, creating new money should always be done with extreme caution, as issuing new money inevitably devalues that money generally, but will cause little or no measurable inflation in a context where the underlying pressure on prices is deflationary. Money creation should therefore be wrapped around with tight controls, and conducted with due regard to inflation targets, perhaps 2 %– 4% annual inflation most of the time,

The Weimar Republic became notorious for a policy of money creation so reckless it caused one of the most infamous intervals of hyperinflation in history. It is likely, however, that a policy of wiping out trillions of pounds/dollars/euros from public sector purchasing power, over the next decade or so, at a time of major global emergency, not least because of the existential threat of climate change to the survival of human civilisation, would in future, if there is one, be seen as even more insane. Quantitative Tightening must be stopped now.

gezwinstanley

Just another WordPress.com site

Category Archives: Uncategorized

Should we maximise equality or utility?

So what do we mean by utility?

Equality

So what are are the implications of this notion of equality for the maximisation of utility?

Why, if this moral imperative for equality is so simply derivable from basic morality, is it repeatedly ignored?

Is Britain Broke?

Office of National Statistics (ONS) Capital stocks and fixed capital consumption, UK: https://www.ons.gov.uk/economy/nationalaccounts/uksectoraccounts/bulletins/capitalstocksconsumptionoffixedcapital/2024

Office of National Statistics (ONS): Weekly Earnings (defined as “wages and salaries”): https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/methodologies/averageweeklyearningsqmi

ONS Household Costs Indices (HCI) for UK household groups: https://www.ons.gov.uk/economy/inflationandpriceindices/datasets/householdcostsindicesforukhouseholdgroupsreferencetables

Infrequently asked questions about Quantitative Tightening that you were never even meant to ask

One people bitterly divided?

REF 1:

REF 2:

REF 3:

REF 4:

REF 5:

Stop the Quantitative Tightening Scam

Are we all bound for Gaza?

Don’t Destroy Trillions

Singularity: not quite yet, but still inescapable

The quiet plan to add an extra trillion to public debt and justify eternal austerity