Is NI a ‘tax on jobs’? Evidence suggests otherwise


4:45 pm - April 8th 2010

by Chris Dillow    


      Share on Tumblr

Does David Cameron know anything about the labour market of northern Finland? It seems not, judging from this:

If we put up National Insurance contributions on every job, employers will have to pay more for them. That’s bound to be a tax on jobs…Putting up the cost of employment will lead to fewer jobs being created and it will probably lead to jobs being lost.

But do NI contributions really affect jobs? This is where Finland comes in. In 2003, the Finnish government wanted to boost employment in the northern part of the country. So it abolished some equivalents of our NI for firms in northern regions. This saved the average firm there around 4% of its wage bill.

Did this create jobs? The beauty of this plan was that it gave us a natural experiment. Because the tax cut only applied to firms in part of the country, it‘s possible to compare these to firms elsewhere in the country, to see if the lower “tax on jobs” created employment.

And it didn’t – at least not to any degree. This paper shows that “the payroll tax experiment has not had a significant effect on employment in the target region.“  It estimates that the average firm that got the tax break created less than one-tenth of a job as a result.

This experiment has been replicated elsewhere. Sweden has also tried regional variations in the equivalent of NI contributions, and has also found no great employment effects.

There are two reasons for this. First, wages rose in response to the lower payroll taxes, so the overall cost to many firms of employing labour did not fall as much as the tax fell. Secondly, the price-elasticity of demand for labour is low. Multiplying a low elasticity by a small cost change gives us a tiny number indeed.

My strong suspicion is that all this would also be true for the UK. We know from the effects of minimum wage laws that the price elasticity of demand for labour generally is low.

And with capital powerful and labour weak, higher employer NICs are highly likely to be passed onto workers in the form of lower wages than they would otherwise get. The combined effect is that very few jobs would be destroyed.

And herein, I fear, lies the farce within the way this issue is debated by the two main parties. Cameron’s comments display an ignorance of both experimental evidence and the principles of tax incidence.

And yet Labour cannot point this out, because to do so would be to draw attention to the fact that a rise in employers’ NICs would amount to, in all probability, a wage cut.

And some people wonder why we’re repelled by the election campaign.

    Share on Tumblr   submit to reddit  


About the author
Chris Dillow is a regular contributor and former City economist, now an economics writer. He is also the author of The End of Politics: New Labour and the Folly of Managerialism. Also at: Stumbling and Mumbling
· Other posts by


Story Filed Under: Blog ,Economy ,Westminster

Sorry, the comment form is closed at this time.


Reader comments


It’s an experiment but not a conclusive one in this context.

The Tories aren’t claiming that a cut in NI will create jobs but that an increase in NI may reduce them.

If a company has thousands of employees and considerable spare capacity but is trying to avoid redundancies, increasing the cost of each employee makes it harder to avoid those redundancies. And we know that employment has not fallen as far as output so we may infer that many companies have spare capacity. So there is a chance that increased NICs will indeed reduce employment. We also know that shoving people onto the dole makes a big hole in the deficit as tax revenues become benefit payments. The increased likelihood of redundancies (although difficult to calculate) should therefore be set against the revenue raised from the increase in NIC.

It would appear that the Finnish experiment was intended to increase employment in the north of the country. This would presumably have happened in two ways – either by allowing existing companies to increase their workforce or by attracting new companies in. The paper doesn’t seem to have considered the latter effect and concentrated on the former. I can’t say I’m enormously surprised that didn’t work.

You’re a moderately successful paper mill – just because your payroll taxes go down doesn’t increase your order book. You can meet your current demand with the current staffing levels so you have no great incentive to hire. It looks as though bosses therefore split the gain between themselves and their staff (who then had more money to spend on goods and services of course and may thus have stimulated the local economy a bit).

The Swedish paper seems to have found the same in respect of the effect on hiring that a reduction in payroll taxes has but it does note the fact that the reduction led to a net inflow of employers to the region. If reduced payroll taxes result in a net inflow of companies then we have to at least consider the possibility that an increase in NIC will result in a net outflow.

You counter this by saying that if NICs go up, British workers will simply shrug and accept lower pay – they might. Or they might, for example, strike. There’s plenty of industrial action going on at the moment because of people not getting a raise. They do tend to get angrier still when the pay actually goes into reverse.

2. Charlieman

Chris Dillow: “First, wages rose in response to the lower payroll taxes, so the overall cost to many firms of employing labour did not fall as much as the tax fell.”

A clarification, please. Are you saying that workers demanded a pay rise *because* they had observed a reduction in the wage cost of employers? A pay rise to get their share of increased company revenue?

“Secondly, the price-elasticity of demand for labour is low. Multiplying a low elasticity by a small cost change gives us a tiny number indeed.”

Would the experiment have performed differently had the charge been reduced for particular sectors of the labour market than for a geographical region? If the charge had been reduced for many service sector workers (assume minimum wage or thereabouts) or for construction workers (more labourers = quicker construction = release of capital), the effects would have been more pronounced.

Why should David Cameron study the labour market of northern Finland? Does he believe that Santa Claus and his reindeer are based in Lappland?
I hope not!
Is Chris Dillow economically illiterate in arguing that wages are equally elastic when rising or falling?
I hope so, because otherwise he is a deliberate liar.
Employment contracts do not permit unilateral reductions in wages.
There are over 140,000 businesses in “Critical or Significant Financial Difficulties” If the NI contribution rate is increased many of them will ber pushed into insolvency
Chris Dillow seems to assume that all their employees will volunteer to take a pay cut?
Why? Many employees will just not believe that it is necessary. Others will not care – if you are 64 and entitled to 80-odd weeks pay on being made redundant why should you opt for a pay cut to keep your job for a few months? If you are on maternity leave with no intention of returning to a boring job, why accept a pay cut?
There are some *honest* arguments against Osborne and Cameron so why not use them instead of making false analogies about a reindeer-infested wilderness? Unless you actually *want* to generate a re-run of the General Strike.

So… NI is just a tax on my wages.

Employment levels don’t look like they’ll change too much, which is good I suppose, but I’ll be relatively poorer. Grand.

I’m really really glad the Government have transfered so much of my money (in the form of delayed tax rises) and wealth (in the form of the gutting of the welfare state, of which I own a part) to the banking sector and the wealthy!

[/Rant ]

Did the reductions create / save more jobs than would otherwise have been the case?

The paper does not answer that question.
(I don’t know whether it is answerable…)

@John77 – your point that higher NICs might tip some firms into bankruptcy is reasonable, but the force of it is weakened by two things. First, the rise won’t take place for another 12 months. Byt this time, many troubled firms will either have closed anyway, or – if the economy recovers – be in a better state. Second, those that remain in trouble will be in a strong position to demand pay cuts. Contrary to George V’s assertion, many firms in this recession have agreed pay cuts or freezes without strike action. And the examples of workers you give who might resist a pay cut are exceptions, not typical cases.
@ Charlieman – yes. It is the wage bargaining process that generates these effects. Note it needn’t be as explicit as outright wage cuts. Instead, firms faced with a 1% higher NIC rate might offer, say, a 2% rise whereas they’d have offered 3% in the absence of a rise.
It’s surely highly plausible that, when capital is internationally mobile and unemployment high, firms will be able to use higher NICs as a reason (pretext?) to hold wages down.
@cjcjc – the paper comes as near as we’ll get to demonstrating what would have happened without the payroll tax cut. It’s plausible that, without the cut, northern Finland would have fared similarly to southern Finland. So the fate of the latter is a reasonable counterfactual. And that’s the comparison the paper uses.

@6 Chris
(i) There are always a few firms on the edge of going bust and the number peaks sometime after the recession bottoms out. I could have said “there are 140,000+ firms now but there will be more in 2011” but that would have required me to explain some elementary economics.
(ii) Anti-discrimination legislation prevents selecting those who should be subject to pay-cuts on the grounds of age, sex or pregnancy. That still applies if those groups are selected to get a 0% pay-cut. So …

“It’s surely highly plausible that, when capital is internationally mobile and unemployment high, firms will be able to use higher NICs as a reason (pretext?) to hold wages down.”
This is based on the assumption that inflation is always sufficiently high for any necessary adjustments to real wages to be made by changing the rate at which they rise in monetary terms. The Brown bust is the worst since the 1930s and, as you say, *some* wages have actually been cut in monetary terms but only a relatively small percentage of the total (unlike in Greece). It is not enough to hold wages down – if Employer’s NI Contributions are increased by more than the total profit earned by the firm then either you have to get the *unanimous* agreement of the workforce in the now loss-making divisions to take a pay-cut or people lose jobs.

Surely this whole ‘tax on jobs’ concept is pretty mute anyway? Even if it eventually goes through, which seems unlikely, how can we measure how many jobs will be lost compared whatever other actions may have taken place. As far as I have read, one alternative is a rise in VAT. Surely a further rise in VAT could kill jobs just as easily? Doesn’t VAT @ 20% put people buying goods or services too?

If you are a SME on the brink of closing down could trying to absorb a 2,5% increase in VAT kill you off, just as easy a NIC hike?

What about ‘Government waste’? Isn’t one man’s waste another industries lucrative enterprise. Let us say, for the sake of the argument that the ‘waste’ identified is a couple security guards at a hospital. Let us say that we stop using that security firm. Doesn’t that weaken that firm too? What if that firm go under taking forty or fifty people with it?

The problem with all these things is that there are apparently less high profile people arguing against them.

Is anyone surprised that some of the richest men in the Country are against a rise in progressive taxation whilst demanding that services and jobs of the poorest members of society are cut? Nobody sees that the CEO in M&S sees a 2.5% rise in domestic fuel or job cuts at the local hospital preferable to his NIC going up as a vested interest?

@8 Jim
Fair question BUT the answer is No
If you are an SME registered for VAT supplying other VAT-registered businesses the direct impact of a change in the rate of VAT is a big fat zero. If you are an exporter or supplying to exporters, the impact on exports is another big fat zero. There will be an indirect impact if the end-customers in the UK reduce purchases as a result of a rise in prices to the end-consumer, but although you may have noticed the impact of the increase in the rate of VAT at the turn of the year I didn’t and today NIESR is suggesting that GDP that fell while VAT was at a reduced rate of 15% grew after it went back up to 17.5%. I agree that there must be some impact on consumer spending but only some of that affects UK manufacturers as some of the reduction will be borne by imports whereas ALL of the NI conts are borne by UK firms. Raising NI Contributions will make UK manufactures less competitive in export markets – raising VAT will not.
Incidentally the tax take from a 2.5% rise in VAT is a large multiple of the tax foregone from blocking a 1% rise in NI contributions for those on middle but not high incomes. Think about it – 1% of a slice of labour costs against 2.5% of total selling price! A rise in VAT of 0.25% would be more than enough: how much of an impact do you think that would have?

“Isn’t one man’s waste another industries lucrative enterprise” Are you joking? Government departments and local governments have a standard practice of sending out letters saying “we have received your letter and intend to reply within 15 days” – which they do not do because they are too busy sending out these letters. I more than once had one of the unfortunate souls working in the complaints section of HMRC ‘phone me up to ask about one of my letters pointing out a whole series of errors in a form they had sent me because their wonderful computer system had overwritten that form with another form with a different set of errors in the two months between my writing it and her being asked to reply. I should be deemed a criminal if I failed to point out their errors but they cannot be bothered to get it right first time, or second time, or third time or…
Some waste is just waste.


Reactions: Twitter, blogs
  1. CHRIS PURSER

    RT @lucyopenshaw: RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  2. Mark

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  3. superbrutal

    RT @HouseofTwits: RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  4. Robin Cox

    RT @pickledpolitics: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  5. Chris Wood

    RT @pickledpolitics: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  6. Job Search Pages

    Liberal Conspiracy » Is NI a 'tax on jobs'? Evidence suggests …: But do NI contributions really affect jobs? Thi… http://bit.ly/ad4fE3

  7. tenpn

    @N1tch why increased NI (by 1%, in next year's stronger economy) doesn't mean job cuts: http://j.mp/aC0nOc via @libcon

  8. Tom Major

    RT @pickledpolitics: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  9. Paul Glossop

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/aBHhxj and this is the major election issue? Give me strength.

  10. Fabian Neuner

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/aBHhxj

  11. Taxation

    RT (@libcon) Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  12. Jim B

    Liberal Conspiracy » Is NI a 'tax on jobs'? Evidence suggests …: If we put up National Insurance contributions o… http://bit.ly/9FgMoy

  13. Jenni Jackson

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  14. Daniel Rees

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/aBHhxj

  15. Liberal Conspiracy

    Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  16. House Of Twits

    RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  17. Lucy Openshaw

    RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  18. Claire French

    RT @HouseofTwits: RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  19. James Cowley

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  20. Jimmy Chen

    RT @HouseofTwitsLab: RT @libcon Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  21. Small Green Bear

    RT @jamescowley: RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  22. uberVU - social comments

    Social comments and analytics for this post…

    This post was mentioned on Twitter by houseoftwits: RT @libcon Is NI a ‘tax on jobs’? Evidence suggests otherwise http://bit.ly/cMHJ8G

  23. Jamie Cater

    RT @libcon: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  24. sunny hundal

    Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  25. Al Wagner

    Liberal Conspiracy » Is NI a 'tax on jobs'? Evidence suggests … http://bit.ly/d4BNtr

  26. disarticulate

    Proof that raising NI has no effect on jobs from Finland of all places – http://bit.ly/cikjnD #UKElection #NI

  27. Kirsty Styles

    RT @pickledpolitics: Is NI a 'tax on jobs'? Evidence suggests otherwise http://bit.ly/cMHJ8G

  28. disarticulate

    Proof that raising NI has no effect on jobs from Finland of all places – http://bit.ly/cikjnD #UKElection ~NI

  29. The Early Skirmishes « Paperback Rioter

    […] Dillow’s excellent post here scuppers any Tory attacks that NI rises are a tax on jobs – seeing as in Finland a lower NI […]





Sorry, the comment form is closed at this time.