Is the Chancellor about to start closing the self-employment tax gap?

Published on Public Finances and the Economy

The Treasury is worried the self-employed aren’t paying enough tax. Conservative backbenchers are worried the Chancellor might do something about it. But what are the facts lying behind this pre-Budget anxiety outbreak?

First things first, the Treasury is right to think there’s a fairly simple issue with the self-employed and tax – they pay a lot less of it. Specifically, on the personal tax side of things, the self-employed pay 9 per cent National Insurance (NICs) on their earnings above £8,060, compared to 12 per cent for employees. There is a straightforward fairness question about whether that difference is justified (summary: it’s not).

However, the real source of anxiety in the Treasury is a fiscal rather than fairness concern; that these tax differences are actually driving the big increase in self-employment we’ve seen in recent years which in turn is undermining the taxman’s ability to get revenues in. To put that in context: 45 per cent of the employment growth since 2008 has been driven by rising self-employment (and no, it’s got very little to do with headlines about the gig economy), with the lower tax take that implies.

But just because there’s a problem doesn’t mean doing something about it is easy or straightforward. Those warning the Chancellor not to do anything to reduce the Treasury’s anxiety have two arguments. First, that self-employment is desirable in of itself as the beating heart of British entrepreneurialism, and second, that many of the self-employed are exactly the ‘just managing families’ that Theresa May wants to help.

On the former the evidence is now pretty clear cut that much modern self-employment has very little in common with the 1990s stereotype of someone starting their own business with a white van, growing it and then employing others. Yes there are lots of plumbers, but there are also many people carrying out IT and accountancy work (not to mention banking) in offices where they work alongside employees doing pretty similar work but paying more tax. Indeed the proportion of the self-employed that are themselves employers has halved since 2002 to just 11 per cent. Workers in the gig economy itself generally don’t set their own prices, a further sign that these aren’t entrepreneurs in the normal sense.

On the question of who would be affected by increasing (Class 4) NICs paid by the self-employed the picture is complicated. Those worried about a tax rise on the low earning self-employed have a point given that individuals would pay 3 per cent more on earnings above £8,060.

But the debate needs to take account of two wider points. Firstly any rise in Class 4 NICs could be timed to coincide with a tax cut for the self-employed that is already in train, the abolition of Class 2 NICs from April 2018. This welcome tax cut saves almost all self-employed workers around £150 a year however much they earn. Secondly those earning self-employment income live in households with other sources of income – complicating the distributional picture of who would lose from any tax rise.

Taking those facts together the charts below gives us the following conclusions about who would be affected at an individual and household level by such a change. The first chart shows that there are indeed losers, as you would expect from a tax rise, with someone earning £15,000 paying an extra £50 a year. The biggest losers though are higher earning self-employed workers who would pay up to £1,000 more. Lower earners would be less affected because they benefit in full from the cut to Class 2 NICs but only pay a small amount or nothing extra by way of increased Class 4 NICs. Crucially, because the self-employed typically have lower earnings than employees (the typical self-employed worker might earn in the region of £13,400 a year in 2018-19) the combined reforms to Class 2 and Class 4 NICs would actually leave the poorest half of the self-employed unaffected or marginally better off.

This next chart turns from individuals to how such a National Insurance change would affect households, with poorest households on the left and richest on the right. As with most increases in direct taxation, the impact is highly progressive with little effect on low earners and better off households paying more. This reflects the fact that better off households simply have much more self-employment income to tax.

So yes increasing taxes for the self-employed – with a likely cost of £1 billion – will, like most tax increases, involve some people paying more. And yes some low earners would be affected (although not the lowest) but the change overall would be highly progressive. The big losers would be self-employed management consultants, accountants and lawyers rather than a low paid plumber or hairdresser. And obviously you can do a lot with £1 billion, including helping lower paid losers from the change. More broadly any such change should sit alongside wider support for self-employed workers, including with maternity pay provision and pension saving.

Stepping back for a second, this week’s row about the tax paid by the self-employed feels a lot like a phoney war. That’s partly because the question of equalising employee and self-employed personal National Insurance feels more like a debate about when rather than if (an increase could come in the Autumn rather than tomorrow). But more importantly it’s because the real debate about tax and the self-employed lies not in the National Insurance individuals directly pay but with the fact that firms pay 13.8 per cent employer National Insurance for everyone they employ, but nothing if they use self-employed labour. How to close that huge gap without causing wider problems is what our limited capacity for anxiety should really prioritise…