A measure of UK annual core CPI inflation excluding direct policy effects fell further to 2.8% in October, the lowest since August 2021 – see chart 1.

Chart 1

The measure adjusts for the imposition of VAT on school fees and bumper one-off rises in water bills and vehicle excise duty. It does not strip out the indirect impact of government actions, including national insurance and minimum wage rises and new packaging waste fees. The NI increase alone may have boosted annual core inflation by 0.25 pp, based on projections in the February Monetary Policy Report.

Policy effects are fading from shorter-term rates of change. The adjusted core measure rose at a 2.5% annualised pace in the three months to October from the previous three months, and by 1.9% between July and October – chart 2.

Chart 2

The core slowdown is consistent with lagged monetary trends, which suggest further deceleration in H1 2026.

Lows in annual broad money growth preceded lows in adjusted core inflation with mean and median lags of 26 and 29 months respectively since WW2. Money growth bottomed in October 2023, suggesting an inflation low between December 2025 and March 2026 – chart 3.

Chart 3

The indirect policy effects cited above may have delayed transmission, raising the possibility of a longer-than-average lag on this occasion.

Money growth, moreover, has remained weak since 2023, suggesting that low core inflation will be sustained into 2027, at least.

Beyond core, food inflation has remained sticky but could break lower in 2026. Annual food inflation of 4.8% in October compares with 1.9% in the Eurozone. Readings were similar at the time of the October 2024 Budget, suggesting that most of the current wedge reflects policy effects. Average UK food inflation was below the Eurozone level over 2015-24.

Based on plausible core / food assumptions, and assuming a neutral Budget impact, annual headline CPI inflation could fall to c.2.25% by Q2 2026 (versus a Bank forecast of 2.9%), with a return to target during H2.

The UK primary fiscal balance is in sizeable deficit, while the effective interest rate on debt is above a trend level of nominal GDP growth consistent with the 2% inflation target. These conditions, if sustained, imply an explosive path for the debt to GDP ratio.

Central government interest payments in the 12 months to August were equivalent to 3.4% of the gross debt stock, according to the latest public finances release.

The Bank of England’s Asset Purchase Facility (APF), however, still owned about a fifth of the debt at the end of the period, so received a significant portion of these payments.

The Bank bought its gilts at much higher prices, so is earning an average interest rate of only 2.0% on the purchase cost of its holdings.

It financed its purchases by creating bank reserves on which it pays Bank rate, currently 4.0%.

Accounting for this carry cost of QE, which the Treasury is obliged to cover, the effective interest rate on government debt over the last 12 months was 4.0%.

This is unexceptional by historical standards but well above an average of 2.7% over 2010-19, when QE was delivering an interest gain – see chart 1.

Chart 1

Achievement of the 2% inflation target over the medium term implies nominal GDP growth of no more than about 3.5% pa, assuming trend economic expansion of about 1.5% pa. An effective interest rate above this level requires the government to run a primary surplus to avoid a trend rise in debt to GDP.

The OBR projected a significant decline in the primary deficit in 2025-26 but the 12-month rolling gap has continued to widen – chart 2. A worse starting position, policy retreats and expected changes to the OBR’s economic assumptions cast strong doubt on the previous forecast of a medium-term return to surplus.

Chart 2

The effective interest rate is subject to conflicting influences and may remain above trend nominal GDP growth. Cuts in Bank rate and QT are reducing the APF net interest loss. On the other hand, the current redemption yield on the stock of gilts, of 4.6%, is well above the interest yield of 3.4% on the stock of debt. Unless the yield curve shifts down, the interest yield will trend higher as existing gilts mature and are refinanced.

The Chancellor’s fiscal rules place emphasis on the current budget but the primary balance is key for stabilising the debt to GDP ratio. Budget measures need to deliver an early return to a primary surplus to calm fears of a fiscal doom loop.

A measure of UK annual core CPI inflation excluding direct policy effects eased to 3.1% in August, 0.5 pp lower than a year earlier.

Published core inflation (i.e. excluding food, energy, alcohol and tobacco) of 3.6% was unchanged from August 2024. The wedge between the published and adjusted measures reflects the imposition of VAT on school fees, a bumper rise in water / sewerage changes and expensive changes to vehicle excise duty – see chart 1.

Chart 1

The adjustment takes account only of direct policy effects, not indirect upward pressure from changes that have loaded additional costs on firms, including the national insurance raid and a double-inflation rise in the minimum wage. The February Monetary Policy Report suggested that the NI changes alone would push up annual core inflation by about 0.25 pp by now.

How should monetary policy respond to above-target inflation driven largely by government-determined prices / costs?

The monetarist recommendation, as always, is that policy-makers should aim for stable money growth at a rate consistent with trend economic growth and the inflation objective. Such an approach avoids accommodating cost-push pressures while minimising any loss of output.

The suggested range for UK broad money growth at present is 4-5% pa. Current annual growth, as measured by non-financial M4, is 3.7%. So the MPC should lower rates / slow QT despite policy-driven inflation.

The Bank of England’s QT programme has been fiscally expensive, is contributing to worrying monetary weakness and wasn’t required on operational grounds.

The Bank estimates that cumulative QT to date has raised 10-year gilt yields by 15-25 bp, up from 10-20 bp a year ago. Gross gilt issuance in 2025-26 is projected by the DMO at £299 bn. Assuming a 20 bp yield impact across the curve, the implied boost to the annual interest cost of the issued gilts is £600 mn.

To emphasise, this is a repeating cost locked in for the life of the securities.

QT started in February 2022. Gross gilt issuance in 2022-23, 2023-24 and 2024-25 combined was £686 bn. Assuming a smaller 15 bp yield impact of QT in those years, the implied extra interest cost on those gilts is £1.0 bn pa.

So the total boost to the interest bill to date could be £1.6 bn pa.

QT could continue through the end of 2026-27. It will have to stop when bank reserves, currently £674 bn, fall into the “preferred minimum range”, previously assessed by the Bank to lie between £345bn and £490bn. Reserves are being reduced by repayments under the term funding scheme as well as by QT. Still, QT could continue at its current pace for another 18 months before reserves reach the middle of the target range.

The yield boost, presumably, will persist at least until the flow of QT is halted. So there could be an additional QT interest bill of £500 mn pa from gilts issued in 2026-27, pushing the total above £2 bn pa.

QT involves the public sector selling additional gilts across the maturity spectrum to repay bank reserves, which earn Bank rate. With the curve disinverting, this currently involves a net interest loss, to be added to the numbers above.

Furthermore, active QT crystallises valuation losses, requiring additional gilt issuance to finance an increased Treasury grant to the Bank.

What were / are the justifications for QT to balance against these fiscal costs?

A “monetarist” argument is that QT was necessary to correct an “excess” stock of money left over from the 2020-21 fiscal / QE splurge.

However, annual broad money growth – as measured by non-financial M4 – had already fallen back to about 5% when QT began in early 2022, subsequently turning negative in 2023.

The previous monetary excess has by now passed fully into prices / activity (mostly the former). The ratio of broad money to nominal GDP has fallen below its end-2019 level and is further beneath its pre-pandemic trend (noted in the May and August Monetary Policy Reports).

Current money trends, moreover, are worryingly weak: non-financial M4 rose at a 3.0% annualised pace in the six months to June, below the 4-5% pa judged here to be consistent with medium-term achievement of the 2% inflation target. (This judgement assumes potential GDP expansion of c.1.5% pa and a 1% pa trend fall in velocity.)

An alternative debt management argument is that QT was / is necessary to reduce the sensitivity of government finances to future changes in Bank rate. According to this view, QE was a reckless policy because it dramatically shortened the maturity of public sector debt (by replacing gilt liabilities with bank reserves), resulting in enormous losses when Bank rate was subsequently raised significantly.

The issue is whether a desirable reduction in the future volatility of interest costs warrants incurring an additional fiscal loss now. It would, obviously, be preferable to undertake a maturity extension when gilts are in a bull market, not a grinding bear.

The Bank’s justification for QT is that a reduction in its balance sheet has been necessary to free up headroom to respond to future economic / financial emergencies. This is unconvincing for several reasons.

First, repayments under the term funding scheme have reduced the balance sheet significantly, with £80 bn of loans still outstanding – see chart 1.

Chart 1

Secondly, the balance sheet would have shrunk considerably relative to nominal GDP and public sector debt even without QT. The asset purchase facility has fallen from 37% of GDP at end-2021 to 20% currently. It would be at 30% if the stock of asset purchases had been maintained at its maximum.

More importantly, the concept of “headroom” as applied to a central bank balance sheet is dubious, and the Bank was far from reaching any form of constraint even when the balance sheet was at its peak.

The maximum Bank share of the stock of gilts was 41%, compared with a 53% peak in the Bank of Japan’s share of outstanding JGBs. Should their holdings of government securities become excessive, central banks have unlimited capacity to lend against private collateral, with appropriate haircuts.

The QE / QT experience raises uncomfortable questions about Bank independence and accountability. Should the MPC attempt to balance monetary policy and operational goals against possible fiscal costs of its actions? If not, who bears responsibility when large losses are incurred?

UK monetary alarm bells are ringing louder.

Six-month growth of the preferred narrow money measure here – non-financial M1, comprising holdings of households and private non-financial corporations (PNFCs) – fell further in June, to just 0.1% annualised. Growth of its broad equivalent, non-financial M4, remained at 3.0%, below a 4.5% average over 2015-19, associated with beneath-target average CPI inflation – see chart 1.

Chart 1

Monetary warning signals are being ignored partly because official / consensus focus is on the Bank of England’s headline M4ex broad aggregate, which grew by 4.4% annualised in the six months to June – exactly in line with its 2015-19 average.

M4ex relative strength, however, reflects rapid expansion – by 14.1% annualised in the six months to June – of money holdings of “non-intermediate other financial corporations (OFCs)”, mainly attributable to increases in balances of securities dealers and fund managers. Such holdings are volatile and – unlike non-financial M1 / M4 – uncorrelated with future activity / prices*.

Six-month growth of M1 / M4 holdings of private non-financial corporations (PNFCs) fell further in June, to 1.4% / 0.4% annualised respectively. Household M4 growth firmed to 3.8% but M1 momentum moved into marginal contraction. The shift from sight deposits into time deposits and ISAs suggests weak spending intentions and a preference for saving – chart 2.

Chart 2

A previous post argued that falls in six-month non-financial M1 / M4 growth in April / May were partly payback for upward distortions related to portfolio adjustments before the October Budget and a front-loading of housing market activity ahead of the end of the stamp duty holiday. With such effects fading, ongoing monetary weakness is stronger evidence of overrestrictive policy.

*Correlations of the two-quarter rate of change of nominal GDP with two-quarter changes in money measures, lagged two quarters, over 1998-2019: M4ex +0.19, non-financial M4 +0.41, M4 of non-intermediate OFCs -0.08, non-financial M1 +0.65.

UK money momentum has weakened alarmingly. The broad non-financial M4 measure – comprising holdings of households and private non-financial corporations (PNFCs) – grew by just 1.9% annualised in the three months to May. Non-financial M1 contracted at a 2.7% pace – see chart 1.

Chart 1

Three-month bank lending growth held up but is likely to fall sharply as a large monthly rise in March – related to the end of the stamp duty holiday – drops out of the calculation. Lending typically follows money trends.

It is unusual for narrow money to lag broader measures when interest rates are falling – lower rates reduce the opportunity cost of holding more liquid forms of money, encouraging a shift out of time deposits and savings accounts. 21% of non-financial M1 is non-interest-bearing. The average interest rate on the stock of household time deposits fell by 31 bp between August and May, according to BoE data.

The demand to hold narrow money is driven mainly by the need to finance future spending, so weakness despite rate cuts is ominous for economic prospects. Put differently, money trends support the view here that MPC policy easing has been too slow, providing insufficient support for activity and increasing the risk of an inflation undershoot.

The monetary relapse could partly reflect payback for temporary factors that boosted growth in late 2024 / early 2025.

A jump in money numbers in October appears to have been related to asset sales in anticipation of changes to capital taxes in the Budget at the end of that month. An asset disposal can boost broad money if financing by the purchaser involves – directly or indirectly – an expansion of banks’ balance sheets*. The effect, however, would be expected to reverse as the seller of the asset deployed the proceeds.

Mortgage lending and broad money were boosted in Q1 by front-loading of housing transactions ahead of the end of the stamp duty holiday. Increased activity may also have resulted in a temporarily higher demand for narrow money.

A reversal of these effects may explain broad money stagnation and a narrow money decline in April / May. Still, annual rates of change should be free of such influences and have slowed to 2.5% for non-financial M1 and 3.6% for M4, from recent peaks of 3.4% and 4.8% respectively. Eurozone annual non-financial M1 growth, by contrast, has risen further to 4.3%.

The sectoral breakdown shows that the recent fall in narrow money reflects a switch by households into time deposits / cash ISAs – their aggregate money holdings have continued to expand, though at a slower pace. By contrast, corporate broad money contracted in April / May, consistent with a negative financial impact from NI and minimum wage hikes – chart 2.

Chart 2

The annual rate of change of corporate broad money is back in negative territory, following small positive readings over December-April, suggesting further weakness in employment and fading capex prospects.

*More precisely, an expansion of banks’ domestic lending or net foreign assets, or a fall in their net non-deposit liabilities.

The directional signal from UK money growth is that annual core inflation – excluding policy distortions – will fall through end-2025. The level suggestion is that core will undershoot 2%. This suggestion is supported by recent exchange rate appreciation.

Turning points in annual broad money growth – as measured by non-financial M4 – have led turning points in core CPI or RPI inflation by a mean 26 months over the last c.70 years. Chart 1 highlights related troughs (gold dashed lines). (See a previous post for an equivalent chart highlighting peaks.)

Chart 1

The May 2023 core inflation peak occurred 27 months after a money growth peak.

Annual broad money momentum troughed at a 67-year low in October 2023. The mean 26-month lead suggests a core inflation low in December 2025. The median lag at troughs, however, was 29 months, so an inflation low may well occur later.

Core inflation fell sharply in H2 2023 and H1 2024 but has stalled since September. The expectation here is that May numbers released next week will show a decline, possibly to below 3%. (The core measure adjusts for the imposition of VAT on school fees and above-normal increases in water / sewerage charges and vehicle excise duty.)

Annual broad money growth averaged 4.2% in the 10 years to end-2019. Core inflation averaged 1.8% in the 10 years to February 2022 (i.e. allowing for a 26-month lag in the relationship).

Annual money growth moved slightly above 4.2% in late 2024 / early 2025 but dropped back to 3.9% in April. So the levels relationship of the 2010s suggests that core inflation will fall below 2%, with no significant rebound before 2027.

Historical variations in the lag between money growth and inflation – and in the levels relationship – often reflected the influence of the exchange rate.

For example, an inflation decline into 2000 occurred earlier than suggested by monetary trends because of a strong disinflationary impact from a prior surge in the exchange rate: the effective rate rose by 26% in the two years to April 1998 – chart 2. This impact was fading by early 2000, contributing to an unusually short interval between lows in money growth and inflation (six months).

Chart 2

Exchange rate considerations are aligned with the monetary message currently, with a 7% rise in the effective rate in the two years to May suggesting that import prices will remain under downward pressure into 2026.

UK April inflation numbers were much less bad than reported.

Annual headline and core CPI inflation rose by 0.9 pp and 0.4 pp respectively from March, to 3.5% and 3.8%. These increases, however, were entirely attributable to hikes in government-controlled prices and vehicle excise duty (VED).

Water and sewerage charges rose by 26% in April versus 8% a year earlier, boosting annual headline and core rates by 0.18 pp and 0.23 pp respectively.

“Other services for personal transport equipment” – a category dominated by VED – rose by 19% versus 4% a year ago, adding 0.22 pp and 0.28 pp to headline and core rates.

The household energy price cap was raised by 4.7% versus a 12.4% fall in April 2024, boosting the headline rate by 0.65 pp.

Summing the above, official actions added 1.05 pp to the headline rate and 0.51 pp to core – more than the actual March-April increases.

Accordingly, the adjusted core rate calculated here fell from 3.2% in March to 3.1%, equalling its recent low (in December and September 2024) – see chart 1.

Chart 1

This measure, moreover, takes no account of Easter timing effects, which may have further inflated the April outturn. For example, air fares rose by 27% last month versus 7% in April 2024, implying a 0.13 bp lift to annual core.

Underlying softening is consistent with lagged money trends and sterling appreciation – the effective rate is currently 3% above its 2024 average level and 7% higher than in 2023.

The MPC is concerned that another inflation pick-up, although unrelated to monetary policy, will generate “second-round” effects. Still-subdued money growth, currency strength and a weakening labour market argue for a relaxed view.

UK monetary trends have been arguing for faster MPC easing. Labour market news is now reinforcing the message.

Employment developments appear notably weaker in the UK than in other major economies. The PAYE payrolled employees series fell by 0.09%, 0.15% and a provisional 0.11% in February, March and April respectively. These declines are the equivalent of falls in US non-farm payrolls of 140k, 240k and 170k.

US payrolls, of course, have risen respectably year-to-date, while Eurozone employment grew by 0.3% in Q1.

The February-April UK payrolls contraction followed a pick-up in the rate of decline of the official single-month vacancies series (seasonally adjusted here), which has fallen to its lowest level since 2016 – see chart 1.

Chart 1

Indeed job postings numbers closely track the official vacancies series and have declined further so far in May – chart 2. The Indeed numbers have fallen by more in the UK than elsewhere, to a lower level relative to the pre-pandemic (February 2020) starting point – chart 3.

Chart 2

Chart 3

UK underperformance may be partly attributable to government-imposed rises in labour costs, in the form of the increases in employer national insurance and the minimum wage, and prospectively via the Employment Rights Bill.

Pessimism here about employment prospects, however, also reflected weak corporate money trends. The six-month rate of change of real M1 holdings of non-financial corporations has remained negative, in contrast to a rise into solid positive territory in the Eurozone – chart 4.

Chart 4

Relative money weakness suggested that UK firms were under greater financial pressure to cut costs than their Eurozone counterparts.

A hopeful sign is that UK six-month corporate real money momentum has recovered recently, narrowing the gap with the Eurozone. A key issue is whether this revival is sustained as the NI and minimum wage hikes take effect.

UK employment weakness appears at odds with Q1 GDP “strength”. The UK quarterly numbers, however, have been volatile and year-on-year growth of 1.3% in Q1 is little different from the Eurozone (1.2%).

Front-running of US tariffs temporarily boosted GDP in the rest of the world last quarter. US real imports rose by 10.8%, equivalent to 1.4% of US GDP, between Q4 and Q1. GDP in the rest of the world is 2.7 times the US level measured at current market prices and 5.8 times based on purchasing power parity. Depending on which divisor is used, the US imports increase implies a 0.25-0.5% lift to GDP elsewhere.