NIC Reeves and the wonder stuff

And it's you, girl, making it spin

I'm going around in circles all the time

After making people cross with my views on Bluesky last time, I’m going to make everyone even more convinced that I’m an Evil Tory, by attacking the Labour budget. Boo! Except, I’m going to attack it from the left. You’ve never debated anyone like me before!

As far as I’m concerned, it’s a story about Rachel Reeves relying on the incuriosity of too many of the media, trying to be too clever, and making life harder for people on low incomes as a result. Let me explain.

You might recall that in advance of last month’s budget, it was widely trailed that there would be a 2 percentage point increase on employer National Insurance contributions. And then, all of a sudden, Reeves pulled a rabbit out of her hat by announcing only a 1.2 percentage point rise. That’s lower, that’s good news right?

Not exactly. The not-particularly-concealed small print was that the threshold for employer National Insurance contributions would go down, from £9,100 to £5,000. Most people can’t do percentages in their head, so not enough people worried about this at the time, and only now are employers’ fears of job losses really starting to gain traction.

In my view, the problem is that nobody properly compared the effect of the trailed rise and the announced one. However the effect is quite striking. For fairness, I’m actually going to propose an alternative plan. I suggest a 2.2 percentage point rise (slightly higher than the advertised one), but I’m going to leave the thresholds where they were.

We can compare the extra amount that employers have to pay for people of different salaries: the actual rise is in red, my proposed rise is in blue. You can see that changing the thresholds has a huge impact: for pretty much everyone there is an extra £4,100 of income that is taxed at 15% - or a flat rate of £615 per employee. While the slope of the blue line is steeper, it’s only 1 percentage point higher in tax, so it takes an extra £61,500 in income above the old threshold to balance out this “tinkering with the threshold” effect.

As a result, employers pay less for every employee earning under £70,000 under my plan than the Reeves one. That’s a really high level for crossover to happen: the median earnings for full-time workers were £37,430 in the last tax year. A salary of £62,000 put you among the top 10% of workers who pay income tax (which is far from everyone) in 2020/1. Labour’s feels like a massively regressive plan compared with mine!

But the effect is even worse than that for some people. Because of the £615 flat rate increase per worker, it’s much less attractive for companies to offer job shares or part time work under the Reeves plan than mine.

We know that lots of people don’t work full time, and there can be good reasons for this that we’d like to support. Some people are managing health problems. Some people return to work only part time while their children are young, to take advantage of Government-funded childcare or for other reasons. A lot of young people work weekend and/or evening jobs to support their studies. All these people just became less desirable employees. Further, the blow won’t fall equally on all sectors of the economy: the hospitality industry will offer part-time contracts geared around busy times (evenings or weekends) for example. That just became a lot more expensive for them, and hence for you when you go out for a drink or meal next year.

It feels like quite a retrograde step for a Government whose manifesto said:

Good work will be the foundation of our approach to tackling poverty and inequality. We will create more good jobs, reform employment support, and make work pay so that many more people benefit from the dignity and purpose of work.

In comparison, my scheme seems much fairer. My hard-earned reputation as a “conservative mathematician who writes in the Spectator” is in the mud!

Girl, you've got control of me

Ok, but there’s a catch here, right? I’ve shown that the burden on employers for the lowest 90% of earners will be lower under my plan than the Reeves one. That must mean that the tax take is lower, mustn’t it? So my plan is no good.

Except, here’s where another simple but important maths idea comes in: skewness of distributions. Here’s a graph of incomes taken from Numbercrunch:

If the distribution of salaries were symmetric (like a bell-shaped normal distribution curve) then the median and mean salary would be exactly the same. However, that’s not the case here: the mean is higher, because there’s a long tail of high earners. And that’s exactly who my proposed tax plan will be asking slightly more of.

We can do a proper comparison with the Reeves scheme: according to the Office for Budget Responsibility, her plan will bring in £23.8 billion. However, she’s also tinkered around thresholds in a different way, by making the employer allowance more generous to try to deaden the effect of her tweaks at the bottom end. This costs about £4 billion, so actually her plan only raises £20 billion.

In comparison, how does my plan perform? Well, in advance of the budget, the BBC told us that:

A two percentage point rise in National Insurance to take the employer rate to 15.8%, for example, would raise about £18bn a year according to published Treasury data.

I’ve gone slightly bigger, 2.2 percentage points not 2, so I would raise 1.1 times as much, or basically exactly the same £20 billion as Reeves. (I’m not going to tinker with thresholds, so no need to subtract anything off).

We can even sanity check the BBC’s calculation: the Institute for Fiscal Studies tells us that 63% of total National Insurance takings in 2023/4 were from employer contributions. HMRC tell us they took £179 billion in National Insurance in that year. Putting this together, you get £113 billion in total employer contributions. I’ve just raised that by 2.2/13.8, or 16% (percent, not percentage points), or £18 billion. Add in a bit of inflation and fiscal drag, and £20 billion sounds about right.

None of this is exactly rocket science. I’ve sketched two curves, and used public data. But how much coverage has there been of the relative effect of the two changes, as opposed to simply focusing on the effect of taxes going up?

It’s hard not to conclude that this is an unforced error on Reeves’ part, relying on a media too busy celebrating the fact that “The grownups are in charge … We can now feel free to get on with our lives without worrying what the halfwits in Westminster might do next” to do the basic sums properly.

In my view, this lack of scrutiny means that not enough people currently understand the real effects of Reeves’ changes on the employment prospects of people of different incomes and shift patterns. Hopefully however, a bit of simple maths has helped shed some light on this for you, and if you found it helpful, do please share it on whatever social media platform you are on this week.