The £8bn overshoot: what the borrowing figures actually mean for the Budget

Every month the Office for National Statistics publishes a bulletin on the public finances, and every month in the run-up to a Budget it gets read like tea leaves. The one published on 22 September was worse than usual for the Treasury, and it landed five weeks before John Healey delivers his first Budget as Chancellor on Wednesday 28 October. This explainer sets out what the numbers say, what they do not say, and why a gap of £8bn matters in a budget of more than a trillion pounds.¹

What the numbers say

In August the public sector borrowed £18.3bn, the difference between what it spent and what it took in. That was £2.9bn more than in August last year and £3.5bn more than the OBR's forecast for the month. Only August 2020, in the middle of the pandemic, was higher. Across the financial year so far, from April to August, borrowing stands at £77.3bn. That is £2.2bn less than the same five months last year, which the government will point to, and £8.1bn above the OBR's forecast, which everyone else will.¹

The forecast figures in the chart are derived from the ONS's stated gaps: £3.5bn for August and £8.1bn for the year to date. The full-year forecast published in March had borrowing falling from £132bn in 2025-26, a year that finished almost exactly on target, which is one reason the OBR's numbers carry weight.³⁴

Where the gap comes from

It is tempting to assume a borrowing overshoot means the economy is producing less tax than hoped. That is not what happened. Central government receipts for the year so far are £460.7bn, up 5.4 per cent on last year and slightly ahead of the OBR's profile, with income tax, National Insurance and corporation tax all growing. The problem is the other side of the ledger. Central government spending is £546.3bn, up 4.9 per cent, and about £7.4bn above forecast. Two lines explain most of it: net social benefits, which include the state pension and are uprated with inflation and earnings, and interest on the national debt.¹²

Debt interest deserves its own sentence. The government paid £8.8bn of it in August alone, the highest for any August since 1997, and £50bn in the year so far. A large slice of British government debt is linked to the Retail Prices Index, so when inflation rises, the interest bill rises with it within months. Inflation was 3.1 per cent in August and forecasters expect it to drift up through the autumn. That is why a war in the Middle East that pushed up energy prices and bond yields shows up, a few months later, as a bigger number in a Treasury spreadsheet.¹⁵

One smaller item is worth knowing about because it will be revisited at the Budget. VAT receipts were soft partly because of the Great British Summer Savings scheme, which cut the VAT rate on some family activities and children's meals from July to September. The government has estimated its cost in the hundreds of millions rather than billions, but it is a reminder that a headline tax cut has a line in the borrowing figures.¹

Debt: the number that is going the right way

Public sector net debt was £2,985.5bn at the end of August, just short of three trillion pounds and £78.5bn higher than a year earlier. Measured against the size of the economy, though, it is 93.8 per cent of GDP, 1.3 points lower than a year ago and 0.8 points below what the OBR expected. That is because the cash value of the economy has been growing faster than the debt, helped by inflation. It is the same effect that inflates the interest bill, seen from the other side. The wider measure the Treasury uses for its investment rule, public sector net financial liabilities, stood at 82.3 per cent of GDP.¹

Headroom, in plain English

The Chancellor has two fiscal rules, both inherited from Rachel Reeves and, so far, kept by Mr Healey. The first, the stability rule, says that by 2029-30 day-to-day spending must be paid for by taxes, so that borrowing only funds investment. The second, the investment rule, says that net financial liabilities must be falling as a share of the economy by the same year. "Headroom" is simply the margin by which the OBR's forecast beats the rule. In March that margin against the stability rule was £23.6bn, which sounds like a lot until you remember it is the buffer for the whole of 2029-30 and a forecast can move by that much in a bad quarter.³⁹

Nobody outside the OBR knows the current figure, because the OBR only recalculates it for a fiscal event. Everyone else is estimating, and the estimates vary widely.⁵⁸

The point is not the precise number. It is that every estimate is well below March, and that the cause is mostly outside the Chancellor's control. As KPMG put it, the Chancellor "will have limited scope to provide significant support for growth or the cost of living when the Budget is delivered next month, as higher borrowing costs and weaker growth have reduced the Government's fiscal headroom."⁵

So what can the Chancellor do?

The government has ruled out raising the rates of income tax, VAT, National Insurance or corporation tax. Those four raise the great majority of the £1.1 trillion collected last year, so the pledge pushes any tax rise into the smaller, more visible corners of the system. The options that keep being reported, according to a Reuters round-up of the choices, are these.⁶

Capital gains tax, currently 18 and 24 per cent and raising about £24bn a year, could be moved towards income tax rates. Academic estimates put the yield of a full reform at up to £11bn, but the tax office has warned that a blunt 10-point rise in the higher rate could lose £3.6bn as people change their behaviour. Property taxes, where council tax raises £54bn and stamp duty £17bn, could be reshaped, though the Prime Minister has ruled out scrapping either. A wealth tax of 2 per cent on assets above £10m has been floated with a £24bn estimate attached, and the Institute for Fiscal Studies has cautioned that valuing private businesses and stopping people leaving make that figure highly uncertain. On the other side of the ledger, there is talk of borrowing up to £9bn more for infrastructure, which would count against the investment rule rather than the stability rule.⁶⁸

There is a third route, and it is the one Westminster is talking about most quietly: accept a smaller buffer. Nothing in the rules says headroom has to be £23.6bn. It has to be above zero. Reports that the Treasury is discussing a lower target are unconfirmed, but the arithmetic is simple. Every billion of headroom the Chancellor gives up is a billion he does not have to raise on 28 October, at the price of a bigger chance that the next forecast puts him in breach.⁸

What the numbers do not say

Three cautions. First, monthly borrowing figures are revised, sometimes heavily: this bulletin revised up its own estimate for April to July, largely because tax receipts were lower than first thought. Second, five months is not a year. The OBR's monthly profile assumes a shape to the year, and self-assessment tax in January can swing the outturn by billions in either direction. Third, the Budget itself will come with a new forecast that rolls in everything since March, from gilt yields to immigration to the price of oil. The £8.1bn overshoot is a real signal, but it is one input among many into a number nobody will know until the OBR publishes it on Budget day.¹

The Institute for Government's advice to the Chancellor was to stop the speculation by publishing a tax strategy and making a few large, coherent changes rather than many small ones. Whether he takes it, the next public finances bulletin arrives on 21 October, a week before he stands up. Expect it to be read like tea leaves too.⁷