John Healey should cut National Insurance, not raise the personal allowance

We have a new Chancellor – and it looks like a tax cut may be surprisingly high on his agenda. Andy Burnham has said the freezing of income tax thresholds was “the thing I heard the most on the doorsteps”, and that it was “lodged in my mind”.

The personal allowance – the amount you can earn before paying income tax – has been frozen at £12,570 since April 2021.¹ Had it risen with inflation it would be £16,070 today. That £3,500 of missing allowance costs a basic-rate taxpayer £700 a year.

It is not hard to see the appeal of delivering a tax cut by increasing the personal allowance. There is an obvious case: it’s simple, it is progressive, and it keeps low incomes out of the income tax system altogether.

The detail, however, is less convincing. The same thing that makes a personal allowance increase tempting politics – that it applies to everyone – means the tax cut it delivers is very small. £6bn used to increase the personal allowance delivers a benefit of £140 a year to almost all basic-rate taxpayers. More targeted measures deliver more cash. The same £6bn used to fund an employee national insurance cut gives someone on the median wage £265. And raising the allowance creates a new and thoroughly irrational tax trap at £126,000.

So it is not crazy to increase the personal allowance. But it is not what I would do.

But if the government is looking at income tax and national insurance, it should look at fixing some of the serious problems that are plausibly holding back productivity and growth. The distortions around £100,000 – where the marginal rate leaps to 62%, and where a cliff edge in childcare support can leave people worse off for earning more – are the worst feature of the income tax system. The Centre for British Progress has set out how they could be fixed, and I wrote about it in the FT. The Centre for British Progress says this can be done on a cost-neutral basis – the Government should be seriously considering whether that’s correct.

Do people really think tax is too high?

There is a widespread belief that tax is too high.

I don’t think that’s correct by historic UK standards. The tax an average worker pays – income tax and National Insurance together – is lower now than in most years since 1975, and far below the 1982 peak. The striking thing is the other line: total tax as a share of GDP is simultaneously at a record high. We are a more heavily taxed country, but not because of the tax on an average wage:²

Nor by international standards. Across the earnings range, the UK sits below almost every comparable European country. There is a particularly striking difference for low incomes, where the UK’s high personal allowance means that we tax much less than any other developed country. The detail, and the reasons, are in that same article:

These charts are, however, politically irrelevant because 63% of GB adults say the tax ordinary people pay is high and just 4% say low. Majorities of every party’s voters, including Labour’s own, says high:³

Perceptions seemed to shift markedly immediately after the 2024 Labour general election victory. For most of the previous five years more people wanted the state to tax and spend more; now 44% say too much against 21% too little:⁴

I don’t hide my personal political views; I’m generally in favour of a more expansive state. However, right now I worry that we have a crisis of tax legitimacy: people think they pay too much tax and that they’re not getting enough in return. The underlying reasons for this are complex and contested. But, if I were the Government, I’d prioritise both trying to noticeably deliver better public services (within fiscal constraints) and delivering a (noticeable) tax cut.

Five years of tax increases

Whilst tax on the median worker is lower than it’s been for most of modern British history, it has gone up in the last five years – and that’s because of fiscal drag.

Income tax thresholds were set at their current levels in April 2021 and have been frozen ever since. The November 2025 Budget froze them for another three years, to April 2031. Wages keep rising, pushing people through the thresholds and into higher rates of tax, or into paying tax for the first time.

Fiscal drag used to be a nice little trick that the public largely didn’t notice. That may have changed.⁵

Here’s what has happened to the three main income tax thresholds, with everything converted into 2026 money:⁶

In the 1990s and early 2000s, people accused Gordon Brown of “stealth tax” by fiscal drag. But this was a gentle drag: he increased thresholds by inflation rather than by wages. What has happened since 2021 is far more aggressive – thresholds frozen outright, through the sharpest burst of inflation in thirty years.

The result has been dramatic. There were 1.7 million higher rate taxpayers in 1990. There are 7.6 million now, and on current policy there will be nearly 9 million by 2030-31:

And the overall tax increase that we as a country are experiencing is mostly down to fiscal drag.⁷

If all thresholds had been uprated by inflation since 2021:⁸

• The personal allowance this year would be £16,070, instead of £12,570. For a basic-rate taxpayer that is a tax rise of £700 a year.
• The higher rate threshold this year would be £64,470 rather than £50,270. Someone on £60,000 would not be a higher rate taxpayer; they pay £2,646 more than they would have. At £80,000 the figure is £3,540.⁹

Undoing this and the parallel National Insurance threshold freezes would cost about £44bn a year.¹⁰

No Government is going to do this.

Is a £500 rise in the allowance the answer?

Let’s assume, for the rest of this article, that £6bn is found – one way or another; I return below to where it might come from. Let’s also assume that the Government wants to use it to finance a tax cut.

£6bn buys about £500 on the income tax personal allowance, and about £500 on the National Insurance primary threshold to go with it.¹¹ There is no reason other than neatness and simplicity why the two thresholds have to move together, but neatness and simplicity are good aims for a tax system.

I don’t think this is the best use of £6bn, but it certainly isn’t a crazy idea:

• Everyone understands it. “You can earn another £500 tax-free” is simple. Almost nothing else in tax policy is as easily understood.
• It helps part-time and lower-paid workers. Someone on £13,000 would have almost all of their income taken out of income tax and National Insurance altogether.
• It reduces the tax penalty on starting work. Moving the point at which tax begins improves incentives around the threshold. Someone on £12,500 may have held back (rationally or irrationally) from earning another £500. Now they won’t. Though the evidence that this matters is surprisingly weak. ¹² More importantly, the decision whether to work at all is exactly the margin where tax matters most for the low-paid.¹³
• It reverses some of the fiscal drag above. I don’t like fiscal drag because it’s not being honest with the public. Reversing fiscal drag therefore is in principle a good thing.
• It reaches beyond working people. Landlords and people living off investment income all benefit and – far more importantly (in political terms – so do pensioners.

These are real advantages.

The big disadvantage is that this is a broad but extremely shallow tax cut. The package means a £140 a year tax cut for almost every basic-rate taxpayer – about £2.70 a week. People on higher incomes benefit a bit more, at £240 a year.¹⁴

That same brilliant advantage of delivering a tax cut to every taxpayer turns out to spread the tax cut very thinly.

There is another disadvantage which is more subtle and technical, but I don’t think it can be overlooked. Increasing the personal allowance creates a new, and very irrational, £126,000 tax trap.

The personal allowance is clawed back once your income exceeds £100,000, at £1 for every £2. Today the clawback finishes, and the allowance is gone, at exactly £125,140 – which is (by design) where the 45% additional rate starts. A larger allowance takes longer to claw back, so the taper now runs past that point and overlaps the additional rate band.

In that overlap each extra £1 of income is taxed at 45%, and costs 50p of allowance which is itself taxed at 45%. That is a marginal income tax rate of 67.5% – or 69.5% once you add employee National Insurance – between £125,473 and £126,140.¹⁵

The fix is obvious: move the additional rate threshold too. It’s peculiar that a Labour chancellor could increase the additional rate threshold, but in this case, doing anything else seems like a bad mistake.

Where would the £6bn come from?

There are only five possibilities:

1. Print money. Reasonably clear this is not going to happen.
2. Increase borrowing. Highly unattractive at current gilt rates.
3. Higher-than-expected growth boosting tax revenues. That worked during the Blair/Brown years – there’s not much sign of it now.
4. Cut spending. Many would like to see the Government do that, but it’s much easier for a think tank to propose than for a government – of any political stripe – to deliver. Andy Burnham’s political background, and the circumstances in which he became Prime Minister, make that seem an unlikely option.
5. Increase other taxes. An equalisation of the capital gains rate with the income tax rate, with an allowance for the normal return on investment and relief for entrepreneurs, could raise a significant sum. Or several smaller tax increases could raise a similar sum.

Is there a better way to spend £6bn?

An obvious alternative would be to cut employee National Insurance. The same £6bn finances a cut of 1p in the main rate of employee (Class 1) National Insurance, and a matching cut to Class 4 for the self-employed.¹⁶

Here’s what that looks like compared with the threshold package. The threshold line is flat – almost everyone gets the same £140 – while the National Insurance line rises with earnings, overtaking it at £26,570 and reaching £377:

So the clear advantage is that it rewards work, and gives the median worker a much larger tax cut than the threshold package (£265 against £140). The clear disadvantage is that the low-paid – those earning under £26,570 – receive a smaller tax cut.

Beyond the politics, there is a powerful technocratic reason to prefer the rate cut: its effect on marginal rates. It’s the marginal rate – the tax on the next pound you earn – that changes behaviour, and a rate cut reduces it for everyone paying it. Costas Meghir and David Phillips’s survey of labour supply and taxes, written for the Mirrlees Review, sets out the UK evidence: responses are concentrated among lower earners and second earners, and they respond to rates, not to lump sums. We want “responses” because they boost growth.

This is not a Laffer-curve argument from the libertarian Right. It’s the mainstream position across the profession – Peter Diamond and Emmanuel Saez’s account of optimal progressive taxation, hardly a right-wing tract, rests on exactly the same logic about marginal rates.

Another advantage: people not working don’t benefit – for example landlords and people receiving investment income.

But the corollary, and the big political problem: pensioners don’t benefit.

To state the obvious: pensioners vote, and a government that hands a tax cut to workers and nothing to pensioners will be heavily criticised. But it’s fundamentally wrong that two people with the same income pay different amounts of tax simply because one is older than the other (or one is working and the other is receiving passive income). National Insurance is the main reason that happens. Cutting national insurance narrows the gap rather than widening it – and I would do this in a heartbeat.

Jeremy Hunt had the right idea when he delivered tax cuts by cutting employee National Insurance. Andy Burnham’s Government should follow it.

Is a high allowance good in principle?

There was a time fairly recently when people somewhat on the left and somewhat on the right both believed there should be a higher personal allowance.

I think it’s a mistake for several reasons.

• It causes unfairness – single-earner households are taxed significantly more heavily than dual earner households (who get two personal allowances!). For this reason, it’s questionable whether a high personal allowance actually benefits the poor (compared to a counterfactual with a lower personal allowance and lower rate).¹⁷
• It causes a muddle at the other end of the pay distribution, where the desire to stop high earners receiving the benefit of the personal allowance generates a marginal tax rate of 62% on people earning £100,000 to £125,140.¹⁸
• And a fundamental point: the best tax cut is a tax cut that reduces the marginal rate of tax for the greatest number of people – all those people receive an increased incentive to boost their taxable income (whether by working more, or planning/avoiding/evading less).¹⁹ A cut in employee National Insurance reduces the marginal rate for every employee paying it – about 20 million people.²⁰ A £500 rise in the thresholds reduces the marginal rate only for the few hundred thousand people sitting in the £500 band just above them (and for the similarly small number between £50,270 and £50,770, whose higher rate threshold moves with the allowance).²¹ For everyone else it is essentially a cash transfer, not a marginal-rate cut – with no incentive to earn more.

There’s no uniquely correct level. But I think we’ve attached far too much symbolic importance to having a very high headline tax-free allowance.

So what should Burnham do?

The Government shouldn’t ignore the mess around £100,000, where the withdrawal of the allowance and of childcare support combine to produce marginal rates above 100% for some parents. That can plausibly be fixed on a cost-neutral basis, which makes it the rarest thing in tax policy: a large improvement that doesn’t need to be paid for. If Andy Burnham is looking at income tax, he shouldn’t ignore this.

But if there’s £6bn for a tax cut, spend it on the rate and not the threshold. A 1p cut in employee National Insurance gives most workers more money than a £500 rise in the thresholds, rewards work, and narrows the gap between how we tax earned and unearned income. If the allowance goes up anyway, the additional rate threshold must go up with it – otherwise we acquire a 67.5% tax rate at £126,000 for no reason at all.

The personal allowance may be the more attractive headline, but a National Insurance rate cut is the better tax reform.

“Press conference by John Healey, then UK Secretary of State for Defence” by NATO North Atlantic Treaty Organization, CC BY-NC-ND 2.0

Further reading

37 ways Andy Burnham could raise £4.7bn

Keir Starmer announced £15 billion of additional defence spending – £4.7bn of that is currently unfunded. That leaves Andy Burnham, expected to become Prime…

The £2bn lawyer tax – should Rachel Reeves tax LLPs?

Doctors, lawyers, accountants, fund managers, and other high earning professionals are often members of partnerships and LLPs. They’re not employees – and so there’s…

The Budget – what it says

Here’s our summary of the Budget and a quick take on what the various measures are likely to mean. (The first draft of this…

The Budget small print: six key reforms to watch

Most of the attention during and after the Budget will be on the big tax-raising measures. But there are an unusual number of important…

The Budget 2025 tax calculator

This online calculator calculates your tax on employment, self-employed or partnership income, and shows how it changes under a variety of Budget proposals. It…

Council tax on ‘expensive homes’ – but most of the money comes from the not-so-rich

The FT has reported that Rachel Reeves is planning a “Budget tax raid on the owners of expensive homes”, expected to raise around £4bn….