Is Keynesianism dead? Far from it – it provides the life support for a crisis-ridden capitalism

On September 28, 1976, then British PM gave an historic speech at the Annual Labour Party Conference held in Blackpool. The speech was written by his son-in-law, one Peter Jay, who himself was mired in scandals throughout his career. For example, the nepotism allegations after he was appointed British Ambassador to the US, his wife’s extra-marital affair with Carl Bernstein, his own affair with the family nanny at the embassy and the resulting paternity lawsuit that Jay initially denied but was found to be the father, not to mention his demeaning relationship with Robert Maxwell. A good candidate for a speech write. In that speech, Callaghan more or less claimed that the Keynesian consensus up to that point (from the Great Depression) was dead and that the fiscal tools that had delivered prosperity in the post WW2 period were no longer effective and should fiscal deficits should be abandoned. How exactly when the non-government sector as a whole desired to spend less than they earned overall was not specified because the lie that cutting government spending was a growth tool dominated. This narratives that Callaghan introduced have been repeated many times since that time when conservative commentators and economists seek attention. The latest appeared in the Financial Times (September 5, 2026) in the form of an article by former Bank of England board member Andy Haldane – Is Keynesianism dead?. He says yes, I say no!

I wrote about that Speech and the historical context etc in sequence of blog posts, which includes (which are best read in chronological order):

1. The British Left is usurped and IMF austerity begins 1976 (June 29, 2016).

2. The conspiracy to bring British Labour to heel 1976 (June 15, 2016).

3. The 1976 British austerity shift – a triumph of perception over reality (June 13, 2016).

4. The British Cabinet divides over the IMF negotiations in 1976 (June 8, 2016).

5. British Left reject fiscal strategy – speculation mounts, March 1976 (May 18, 2016).

6. The Bacon-Eltis intervention – Britain 1976 (May 11, 2016).

7. The 1976 currency crisis (April 26, 2016).

Callaghan said (among other things in the Speech) that:

Britain faces its most dangerous crisis since the war … The cosy world we were told would go on for ever, where full employment would be guaran­teed by a stroke of the Chancellor’s pen, cutting taxes, deficit spending, that cosy world is gone …

When we reject unemployment as an economic instrument – as we do – and when we reject also superficial remedies, as socialists must, then we must ask ourselves unflinchingly what is the cause of high unemployment. Quite simply and unequivocally, it is caused by paying ourselves more than the value of what we produce …

We used to think that you could spend your way out of a recession, and increase employ­ment by cutting taxes and boosting Government spending. I tell you in all candour that that option no longer exists, and that in so far as it ever did exist, it only worked on each occasion since the war by injecting a bigger dose of infla­tion into the economy, followed by a higher level of unemployment as the next step. Higher inflation followed by higher unemployment.

Here is a relevant snippet from his Speech for interest:

Callaghan and the Chancellor Denis Healey had become infested with the Monetarism of Milton Friedman and were trying to work out ways to reneg on the Social Contract with the trade unions so that they could begin a period of fiscal austerity.

The IMF ruse (that the UK government had to borrow from the Fund) was designed to depoliticise the austerity – the Labour government could simply blame the IMF for the hardships that the austerity would bring.

All the mainstream ruses were propagated by Callaghan and his gang:

1. Unemployment was the result of excessive wage demands.

2. The provision of unemployment benefits undermined the incentive to search for jobs and subsidised unemployment.

3. Government spending carried a negative multiplier – in other words, increasing government spending reduced overall spending.

These themes are regularly repeated by the orthodoxy despite several periods since the 1970s when Capitalism required fiscal life support assistance, which proved categorically that ‘Keynesian’ remedies remained relevant.

Haldane’s article cites a number of sources none of which have been accepted as definitive in the area.

Each study cited has been subjected to many criticisms relating to data, method etc.

But, of course, Haldane leaves the qualifications of the results he cites out, and just asserts them as authorities to back up his spurious reasoning.

Typical approach.

The use of the term “Keynesianism” is loaded and I discussed that issue in this blog post – Those bad Keynesians are to blame (November 5, 2009).

There are many meanings given to it and most miss the mark.

The blog post just cited provides all the detail necessary to develop and understanding of these variations.

Haldane briefly notes that the ideas of John Maynard Keynes, which were not exclusively his ideas, allowed governments to bring their economies out of the Great Depression through government spending in excess of taxation revenue.

The facts are that it wasn’t until the military expenditure associated with prosecuting WW2 that the downturn ended.

The Post WW2 period also delivered major gains in prosperity and declining prosperity because governments took responsibility to ensure there was sufficient aggregate demand to generate the jobs that were sufficient to provide full employment.

Haldane notes that this approach broke down in the early 1970s as Monetarism emerged as the dominant paradigm in the academy and then infested policy circles and central banks.

The words are mine but that is what he was referring to.

He makes out the end of the Keynesian period and the rise of Monetarism as if it was a natural or necessary development.

The rise in acceptance of Monetarism was not based on an empirical rejection of the Keynesian orthodoxy, but in the words of American economist Alan Blinder:

… was instead a triumph of a priori theorising over empiricism, of intellectual aesthetics over observation and, in some measure, of conservative ideology over liberalism. It was not, in a word, a Kuhnian scientific revolution.

[Reference: Blinder, A. (1988) ‘The fall and rise of Keynesian economics’, Economic Record, 64(187), 278-294.]

Tracing this ideological shift is important because it allows one to reject the proposition that the ideas about the dominance of fiscal policy etc were defunct and the Monetarist ideas were superior.

Haldane demurs on that task.

He acknowledges, though, that the recent history is dominated by the “remarkable comeback” of the Post WW2 ideas of Keynes and others:

… in response to a new set of conjoined crises, the global financial crisis, Covid and the Russia-Ukraine war. Indeed, the new interventions have dwarfed any previous peacetime stimulus.

These interventions are the life support systems that I referred to at the outset and without them the Capitalist system as we know it would have collapsed.

Haldane’s claim then is that:

But while cushioning the effects of crises, such mammoth interventions have left their own lasting scars: public debt across the G7 now exceeds annual GDP, with debt ratios doubling — in the UK and US almost trebling — since the start of the century. This raises questions about the effectiveness of future fiscal stimulus, especially in high-debt countries such as the UK, US and Japan. Is Keynesianism now diminished, perhaps defunct?

His argument is in three parts:

First:

Keynesian policies need to catalyse private spending. But if people anticipate that borrowing today means higher taxes tomorrow, stimulus might prompt saving rather than spending.

This is the familiar Ricardian equivalence theorem being rehearsed.

Of course, the first part is not entirely accurate – government spending on public infrastructure does not work through “private spending” and provides direct stimulus to output and employment.

I provided a detailed response to the Ricardian claim in several blog posts including these ones:

1. Pushing the fantasy barrow (February 25, 2010).

2. Deficits should be cut in a recession. Not! (November 30, 2009).

3. We are sorry (February 14, 2010).

Haldane does not tell his readers that the theoretical models on which his claim is based require several assumptions to be met to establish the result:

(a) Capital markets must be ‘perfect’ – totally equal access to finance for all households – Not a real world reality.

(b) The future time path of government spending is known and fixed – that is, households/individuals know this with perfect foresight – – Not a real world reality.

(c) There is infinite concern for the future generations – Not an accurate description of human behaviour.

The theory was espoused in the late 1970s and the large US tax cuts in August 1981 were seen as the first real world experiment to test the validity of the claim that households would cut spending to offset the government expansion.

The mainstream economists predicted there would be no change in consumption and saving would to pay for the future tax burden as public debt rose.

if you examine the US data you will see categorically that the personal saving rate fell between 1982-84 (from 7.5 per cent in 1981 to an average of 5.7 per cent in 1982-84).

The Ricardian equivalence phenomenon has really never been observed in practice.

Second:

… if extra borrowing is perceived as increasing the likelihood of governments inflating away or defaulting on their debts in the future, this will raise borrowing costs and depress demand today.

He cites the recent increase in bond yields around the world.

It is a separate blog post to analyse this recent behaviour in the bond markets but jumping to the conclusion that the trend is driven by concern for the solvency of governments is far fetched.

Bid-to-Cover ratios across the G7 countries (as being representative) are consistently about 2 and in Japan above 3, which means that the demand for government debt is well above the supply from governments.

Long-term debt issues usually have lower bid-to-cover ratios but even a comparative 30-year performance appraisal sees the ratios well above 2.

Further, the current trends are being swamped by the corporate borrowing boom associated with AI investments, which has push increased supply of debt into the financial markets and shifted reduced demand for government bonds – in the frenzy.

Moreover, as long as Donald Trump pursues his mindless war in the Middle East and energy prices remain high, supply-side inflationary pressures will persist and investors will require higher bond yields to protect real returns.

All of the above have nothing to do with fiscal policy settings nor the amount of outstanding government debt.

And when the next crisis hits, expect to see central banks once again drive bond yields down towards zero (and negative in the case of Japan), when they reengage large-scale bond buying programs.

Third:

… multipliers depend on how monetary policy reacts — for example, whether central banks accommodate fiscal stimulus or instead offset its effects by tightening policy. The latter is more likely when inflation itself is above target.

Of the studies, Haldane cites, here are a few comments:

1. “In a study of 44 countries, Ilzetzki, Mendoza and Végh find evidence of fiscal stimulus depressing growth — a negative fiscal multiplier — when countries’ public debt exceeds 60 per cent of GDP.”

The study conflated nations with their own currency with those that use foreign currencies, which means the bond market dynamics are quite different.

There is no evidence that ‘crowding out’ occurs in currency-issuing nations.

Further, the study failed to separate out the debt dynamics from the economic cycle, which means that the causality imputed was not able to be established and therefore became just an assertion based on their ideological priors.

Also, they assumed symmetry in the way the economy responds to fiscal changes, which biases there statistical method towards the conclusion they sought!

The authors also did not distinguish between consumption-based stimulus and investment-based stimulus – the latter being public infrastructure developments which the evidence shows strongly ‘crowd in’ private investment.

That omission biases the results towards the conclusion they reached.

2. “Surveys in the UK, US, France, Italy and Japan suggest future tax rises are now one of the main factors damping sentiment and curtailing spending. For example, fears of tax rises among UK companies rose sharply ahead of each of the last two Budgets, topping their worry list. This caused savings to rise and growth to stall. Next month’s Budget is shaping up for a hat-trick.”

The last two Autumn Fiscal Statements (aka ‘the budget’) in the UK were released on October 30, 2024 and November 26, 2025, respectively.

The – Autumn budget 2024: Key announcements and analysis – provided for “a significant increase in public spending, financed by a combination of tax rises and higher borrowing.”

It “represented one of the largest fiscal expansions in recent history, adding roughly 1% of GDP per year to borrowing to step up public sector investment.”

The – 2025 Statement – forecast a modest decline in the deficit and public debt continuing to rise as a per cent of GDP.

In other words, neither signalled any major austerity shift.

What happened to savings?

The household saving rate actually fell from the fourth-quarter 2024 (when it was 11.1 per cent) to 8.9 per cent in the March-quarter 2026.

Exactly the opposite to Haldane’s claim.

And what happened to GDP growth, which he said ‘stalled’?

In fact, annual GDP growth grew by 1.1 per cent in 2024 and 1.3 per cent in 2025, with major contributions come from government spending and the external sector.

Further, neither household consumption expenditure nor private business investment fell across 2024 and 2025.

Household consumption expenditure rose by 0.4 per cent over 2025 and business investment by 4.3 per cent.

3. “The effects of fiscal precarity on borrowing costs — the second channel — are also increasingly visible. Long-term yields in the G7 countries have risen 4 percentage points since Covid, due to inflation and fiscal concerns.”

Fiscal deficits have been at elevated levels since the GFC and before that in Japan.

The following graph shows the 10-year bond yields in the UK and Japan since 1990 (and the movements are representative of other nations) over this period.

The recent increase in yields is associated with one major change – the rise of inflationary pressures following the supply constraints arising from COVID and then later the inflationary pressures arising from the two wars (Ukraine and Iran).

The correlation between the bond yields and the fiscal position is poor.

But the correlation with inflationary pressures is high.

Further, Haldane cited the work of British economist David Aikman who he used as an authority for the claim that “Yields are also becoming more sensitive to news about deteriorating deficits”.

The evidence on that question is not clear cut as Haldane suggests.

Rather, the evidence strongly suggests that the absolute size of fiscal deficits is not a driving factor in the bond yield dynamics.

The more influential factors are economic growth expectations, inflationary pressures, and the concept of the ‘safe-haven’.

The so-called ‘safe-haven’ effect refers to the fact that when there is global turmoil and growth forecasts fall, the large global investors quickly move funds into government bonds because they have no credit risk (for currency-issuing governments).

The shift drives yields down because the increased demand drives prices up.

The opposite happens when there is a shift in sentiment.

At present, investment funds are shifting out of government bonds into AI-type debt products, which is a major reason that bond yields are rising.

And think about Japan – with the highest growth debt-to-GDP ratio and significantly elevated fiscal deficits for near three decades – yet bond yields didn’t move until inflation entered the picture.

Similarly after the GFC, most currency-issuing governments ran much larger fiscal deficits than historically were the norm with no substantial uptick in bond yields – the demand for the debt was strong throughout.

Finally, there are distinct financial market dynamics involved that have nothing to do with the fiscal positions.

The role of the ‘dealers’ in bond markets is important – they are the “select group of market-making financial firms” (SUERF Policy Brief No 1173 May 2025 – Dealer constraints and government bond markets: A transatlantic perspective).

The dealers are:

… typically affiliated with large banking organizations. These so-called primary dealers participate directly in the primary market for Treasuries, where the US government auctions new debt issuance, and then resell the Treasuries to investors in the secondary markets.

These market-makers exist in every nation.

They also “act as intermediaries between sellers and buyers in the secondary market” and interact with the central bank in terms of the open market operations (the central bank buying and selling government debt to manage liquidity).

Regulative changes to this intermediation capacity have been closely related to the movement in bond yields.

The discussion is too technical for this audience and I might write a separate blog post about it in an attempt to make the discussion accessible.

Put simply, when new financial regulations (for example, the Basel III Supplementary Leverage Ratio (SLR) rules) impinge on the legal balance sheet space that dealers have to absorb selling pressure in the bond market, the central bank has to step in.

In the interim, bond yields rise because the dealers require higher compensation to justify taking on more debt.

A study by the Federal Reserve Bank of Boston – Evidence That Relaxing Dealers’ Risk Constraints Can Make the Treasury Market More Liquid (published March 4, 2025) – demonstrated this point beyond doubt that changing regulations governing the financial markets alters bond yields without any shift in the fiscal position.

The SLR essentially requires the tier-1 banks to hold a minimum layer of capital against their leverage positions and doesn’t distinguish between quality of the asset.

So the SLR = Tier 1 Capital as a proportion of Total Leverage Exposure.

Tier 1 capital is the bank’s core equity capital and disclosed reserves – that is, the buffers in case of losses.

Total Leverage Exposure is all on-balance-sheet assets (such as loans and cash) plus off-balance-sheet items (such as derivatives and repo-style transactions).

Every dollar of Leverage Exposure is counted equally, regardless of the quality of the asset.

So a risk-free Treasury bond has to have the same amount of capital backing it as does a risky corporate bond held by the bank.

This acts as a disincentive for the banks to hold Treasury bonds and so the SLR acts as a penalty on low-risk, market-making activities by the dealers in the bond market.

During the early COVID-19 years, the Federal Reserve exempted Treasury bonds from the SLR requirement to free up space in the dealers’ balance sheets.

So within 24 hours, US Treasury bonds no longer consumed regulatory capital backing and in the first week, banks increased their Treasury positions by around 10 per cent.

See also this paper from the Bank of International Settlements – Dealer capacity and US Treasury market functionality (October 26, 2023).

New regulative changes in the US, for example, are working to push up bond yields.

The US SEC – Treasury Clearing Implementation – rules, which were announced in December 2023 and come in operation by December 31, 2026 and then further shifts by June 30, 2027, basically make transactions relating to bonds more secure but have had the impact of elevating costs for the primary dealers.

The result has been that the increased costs are being passed on by the dealers demanding high bond yields to clear new auctions.

None of this is related to the fiscal position of the US government.

And on this theme, when Trump announced widespread tariffs in April 2025, the US bond market became very volatile as the ‘dealers’ reached the limits of the capacity to act as intermediation agents.

This was also nothing to do with the fiscal position of the US government.

Conclusion

I understand that all this detail is not suitable for an 1000-word Op Ed, which just means that an author should not make bold claims and drop in a simple reference as if it is an authority for the bold claim, without at least noting the contestability of the argument and providing the nuances.

But the evidence does not support Haldane’s central claim that “Keynesianism, if not dead, is potentially defunct, except when squarely focused on public investment.”

I think the evidence suggests quite the opposite.

That is enough for today!

(c) Copyright 2026 William Mitchell. All Rights Reserved.