Another week, another attack on Lib Dem tax plans. This one comes from Kayte Lawton, Senior Research Fellow at the IPPR, whose publication on Coalition Tax Policy concludes that raising income tax thresholds is regressive in its impact.
What do terms like ‘regressive’ or ‘progressive’ mean? Quantitatively, it’s a little difficult to define: we’re not talking about a single number, but about a distribution – the shape of a graph. If we consider net gains to an individual and order our data by income percentile, then a progressive distribution of gains would slope down from left to right; if we’re talking about net losses then a progressive distribution would slope up from left to right.
There are a number of ways to fudge whether or not a policy appears progressive or regressive. For example, a report may be selective about reporting data in cold hard cash or as a proportion of income: this choice could make the difference in the policy appearing to be progressive or regressive.
Secondly, we need to be careful about whether we are labelling a policy itself (eg, a taxation system) as being progressive, or whether we are saying that the change from the previous policy is progressive. Even if the change in the policy has a regressive distribution, that’s not to say that the resulting policy is regressive.
Now, Lawton’s analysis doesn’t claim that income tax distribution after increasing the income tax allowance would be regressive. It would take a fairly serious shakeup of our income tax setup to get anywhere near that. She does, however, argue that the change to a £10,000 or £12,500 personal allowance would represent a regressive change to the system. But is this true and, if so, is it important?
The crux of the argument rests on the fact that we already have an tax allowance of £7,475. People in the lowest income percentiles are already earning less than this threshold so any increase in the threshold will represent zero gain for those people. While this is true, increasing the tax allowance (assuming no change in the higher rate tax threshold) would give the greatest differential gain to individuals around the 25th income percentile and would then steeply decline for those in higher income brackets. This is still a broadly progressive change.
But what if we were starting from a point of having no income tax threshold at all and instead paid 20% on all earnings up to the higher rate threshold? Would we choose a £7,475, £10,000 or £12,500 threshold? Or would we have no tax allowance at all?
The graph below shows that implementing a personal allowance is an irrefutably progressive step – the benefits are high and flat for the lower income brackets (up to the value of the threshold) and fall away with increasing income.
One could argue that a lower tax allowance is more ‘progressive’ than a higher one since it targets the lowest percentiles more extremely, but this amounts to arguing that it is fair for a tax system to punish those earning the minimum wage (which is around the 25th percentile) – hopefully enough of a reality check to discount that argument.
So how has Lawton concluded that implementing these changes to income tax allowance would be a regressive step?
The key is in grouping by household rather than by individual. Since households often have a mix of high earners and low earners the tax gains by the lower earners are shared by the same households as the higher earners which pushes the peak of the graph further towards the right. Also, low earning households are more likely to include those at the very bottom of the scale who would not gain anything from the threshold change.
Another point worth mentioning is that, when the allowance was last increased, it was accompanied by a reduction in the higher rate threshold. This reduces any gains made by higher earners. Lawton acknowledges that she has assumed that no changes to the higher rate thresholds would be made in implementing the Lib Dem policy, which would tend to exaggerate any regressive tendency.
In summary then, implementing the Lib Dem policies on income tax would result in a progressive tax distribution. The distribution of the changes across all individuals would be fairly strongly progressive but the distribution of the changes across households could be weakly regressive (based on key modelling assumptions).
Is this important? It highlights that the increase in allowance does nothing for those earning less than £7,475 and thus the need to tackle unemployment if we want to maximise the effectiveness of the policy, but I don’t think this is a good argument against implementing the changes. Any regressive effect is an artefact of the underlying income distribution and allowing the lowest quartile of earners to save all of their income and invest in their own future offers a tool to change that income distribution.
* Ed Long campaigned for Simon Hughes and Columba Blango in the 2010 election and ran the online Geek the Vote campaign. He is currently working for the Blavatnik School of Government at Oxford University.
Source for percentile income data: Income in the UK.
Follow @LibDemVoice on Twitter
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
The next Senedd elections will present a different challenge to the Lib Dems
Michael's weekly ward surgeries tonight! #dundeewestend
Not all Tories like Rwanda
Conference photos
Our party celebrates Britain's diversity as a great strength - Ed Davey
Eclipse
Political choices in choosing growth
David Blake
Simon R
Chris Moore
Duncan Greenland
19 Comments
This is a very useful article thanks. When Lawton and others claim that raising the threshold is not progressive, what they mean is that the lowest income households don’t gain, ignoring the fact that it is broadly progressive across the piece.
And as you point out, she has also ignored likely balancing changes at the top. I can’t see any way in which the Lib Dems could raise the threshold without a compensating change in the higher rate threshold, or some other mechanism for taxing the rich.
Her analysis also ignores the fact that the Lib Dems would also make other changes which would help those on the lowest incomes, such as increasing the basic pension, thus targeting one of the largest groups of low income households.
Judging from your graph it’s regressive in relation to the poorest quarter, and progressive in relation to the rest. (Of course, how it’s funded might change that, but obviously that’s a separate issue.)
The people it would benefit most are the middle-class target voters, as usual.
This would be the same Kayte Lawton who is a regular contributor to Labour List, secretary of the Barnsbury ward Labour Party and the organiser of Islington South and Finsbury Labour Party’s policy forum?
Labour list: http://www.labourlist.org/kayte_lawton
Barnsbury Labour: http://www.islingtonsouthlabour.org.uk/#/our-officers/4546934245
Islington South and Finsbury Labour: http://www.islingtonsouthlabour.org.uk/#/whats-happening/4546907623
One thought a respectable organisation like the IPPR might have mentioned this as part of their research…
another researcher for the ippr is an islington labour councillor (andy hull). ]
The ippr: the hidden voice of Islington Labour.
I must say, I don’t think it matters one iota that people individually are tied up with one political party or another, if they are arguing a technical case.
It seems to me, Ed, that you are arguing more to have tax assessment returned to family income, rather than individual. That is clearly a separate argument, and one that many of us would sympathise with on a purely distribution of income point of view (there are, of course, other strong arguments against). I have to say, though, I am at a loss to understand why we don’t argue for a much more graduated sliding scale of rates, as wasthe case 40 – 50 years ago. That would then tend to smooth out the progressivity of the whole scale. My problem with exempting more and more lower end income from income tax is that it is likely to decrease amount available for public services.
That IPPR paper referred to those earning full time on minimum wage as “wealthy”.
That’s almost as daft as Eric Pickle”s claim that the 50p rate on earnings over £150 is “an attack on the middle class”.
I usually like what the IPPR have to say but this was clearly a weak attempt at attacking a political opponent.
The gist of the IPPR argument that raising the tax threshold is regressive (not progressive) is that the very poor (earning under the current threshold) will not benefit, but “higher earners” will.
It’s a pretty weak argument. If I may point to my opinion blog here on Lib Dem Voice from 27th Sept which says: No-one should underestimate the fundamental shift in attitude and approach that will flow from [a tax threshold around the full time annual minimum wage]. It encourages aspiration, it clarifies the role of benefits as a safety net and it stops the bad practice of the state taking with one hand while giving with the other.
Now that, I would say, is progressive if we can achieve it!
It is no more relevant that the author of the IPPR publication is a Labour activist any more than the author of this article is a Lib Dem activist.
The only thing that matters is whether the arguments stack up or not.
I would like to see a debate between Kayte and Ed on this issue, that would be the best way to find out who is right.
What I think is fair to say is that at best the policy is only mildly progressive. The people who need the most help do not pay any tax to begin with. Increasing their benefits would be a lot more progressive.
Another one for my A2 sociology students and A2 Politics students to argue over. Very interesting article.
I’m going to disagree with Geoff and argue that its reform of the benefits system that needs to occur – with tapering of benefits rather than a poverty trap created by a financial cliff that you fall off when you go over a certain threshold. The graph would look something like the White Cliffs of Dover 😉
Um… I’d like to retract some of the things I said above about the IPPR.
What I said was based more on a skim read than a careful reading and looking back, where I’d remembered them saying:
“That IPPR paper referred to those earning full time on minimum wage as “wealthy”.”
It had actually said something more along the lines of:
“The wealthiest get a higher proportion of the money spent”
and “Two thirds of the money go to households in the richest half of the country”
I think there’s still criticisms to be made, but a bit more subtle and much less careless than the ones I made.
I hereby promise to be more careful in future. 🙂
@Tim13 “My problem with exempting more and more lower end income from income tax is that it is likely to decrease amount available for public services.”
But perhaps people would rather have more money to spend as they wish than the state deciding for them how it should be spent?
@Neil – thanks very much!
@Chris – depends what you think of as being the link between income and class. The 25th income percentile is £11,800 per year, which doesn’t sound like a middle class income for a full time worker to me. £27,000 (the salary of a newly qualified teacher in inner London) puts you almost at the 75th percentile. The reality is that economic class includes savings and investments as well as income and the lower percentiles in income are inflated by people mainly supported by parents or spouses, students, pensioners or people temporarily out of work.
@Dan, LondonLib, Tim13 – I’m not so much trying to argue whether the change is progressive (hurrah!) or regressive (hisssss) (NB. LDV changed the title of the article to sound like I am), rather to explore how these things get calculated, how subjective they are to the way you organise your data and how thinking of ‘progressiveness’ as a value that you want to maximise doesn’t make any sense.
@Geoff – I’m not disagreeing with Kayte on the figures in her report so there isn’t really any “finding out who is right” to be done. I’m stating my opinion that the policy is a goodun and presenting the same data in a different way to back up my opinion. The debate would basically be this:
Kayte: “Redistribution good, tax breaks bad”
Ed: “Financial independence and incentive to work good, poverty trap bad”
It’s also worth noting that if this policy was being proposed by Labour (or by a Lib Dem party more sympathtic to Labour), the fact that a major group of beneficiaries are low-earning women with non-poor husbands would be being justified on feminist grounds.
Another point, of course, is that a two-earner family is in fact poorer than a one-earner family on the same household income – they have to deal with all the costs (particularly childcare) and aggro from two jobs rather than one. So the “rich” families benefitting from the secondary earner being taken out of income tax aren’t as rich as IPPR’s statistics makes them look.
@Chris – that depends partly on how you measure the ‘benefit’ and on how you define ‘middle class’. I would argue that an extra £1,000 in the pocket for someone earning £12K will make far more difference to their quality of life than it will to someone earning £25K. In percentage terms it is larger. The majority of people who benefit will be earning less than £30K.
@Tim13 – Many of us in the party have argued strongly for income tax and employees NI to be combined as well as raising the threshold. If this happens it would be an ideal opportunity to look at the bands at the same time. We could go for combined lower, middle and higher bands that could be more reflective of income distribution. Raising the threshold only decreases the amount available for services if you don’t balance it by changes at the top. So far the Government have done this. Of course there is a substantial saving in running the system by taking millions of people out of the income tax and NI system altogether which can also be spent on services.
@Geoffrey – to an extent you are confusing the progressiveness of raising the threshold with the progressiveness of the system as a whole. Those in the bottom decile pay 0% of their income in income tax, so the income tax system as a whole remains just as progressive for them. Raising the threshold doesn’t make them any worse off, it simply makes deciles 2-10 more progressive. Raising the threshold does not stop you introducing other policies to help those in the lowest decile. However you also appear to be making an assumption that the lowest income decile equates to benefits claimants, which it doesn’t. Many people in the lowest income decile are groups like students, self-employed people in a low income year, and many benefits claimants are in higher deciles.
I don’t have a problem with the argument that increasing the threshold is progressive. But it will cost a lot and I wonder whether that is the best use of that resource given that it will have no benefit for the very lowest paid.
Great article, Ed.
Two comments. Firstly, the IPPR report falls into the trap of thinking of economics as static. The person earning £7,500 may gain nothing immediately from an increase in the personal tax threshold, but that is not to say that they will not earn £10,000 or £12,500 in the future. Perhaps more crucially, raising the thresholds might provide the incentive for them to do so. As those earning less than £10,000 are part-time workers, they may wish to up their hours and will be more inclined to do so by a 25% increase in take-home pay (ignoring NICs for a moment, which muddy the waters).
Secondly, the Lib Dems may wish to consider the next step not being increasing the personal allowance but reducing the marginal rate: i.e. dropping the starting rate to 19% instead of raising the allowance to £12,500. This would not be as “progressive”, but it would improve the incentives for employees. To quote an economic axiom, “All economic activity takes place at the margain”: when deciding whether to do that extra hour, or to apply for that jobs that pays an extra 25p/hour, the worker does not consider their average tax but how much of the additional sum they will take home. Of course we want to help those on/just above the minimum wage, but we also want to incentivise those earning in the high teens and low twenties.
@ Tom Papworth
It is really important to note this point about incomes not being static. The people who are most likely to fall into the lowest income categories are those at the beginning and end of their working lives i.e. students/those starting in the jobs market and pensioners. The IPPR analysis seems to view people as being trapped in one single category rather than moving between them over time.
I think one of the most important points about the £10,000 (or £12,500) personal allowance is not the comparative statics, but the effect on the dynamics of the jobs market. If you are unemployed, it gives you much more of an incentive to find work if you can earn a living wage without paying tax. If you are earning a low wage, it gives you an incentive to find a job with a higher wage.
As for the IPPR’s analysis being “technical”, you can dress up arguments all you like in “technicalities” but this is only a disguise. In the end, you have to adopt a viewpoint by interpreting the evidence. Therefore it is entirely relevant to note Kayte Lawton’s party political background and potential bias.
I’d also like to add that this has an important regional dimension in that in the most depressed regions there are far more lower paying jobs closer to the £10,000 figure than there are in London and the South East. Proportionately the benefit will therefore be greater in the poorer regions than the richer ones.
I would have thought that the simplest way to achieve a rational tax-benefit structure, free of poverty traps, would be to tax all income and make all benefits universal. Just draw the graph.