Basic Income Part 2- How Much Would It Cost?

Basic Income Part 2- How Much Would It Cost?

In the first part of this two part post, I discussed what basic income is, why it might be a good idea, and how some of the many objections to it can, I think, be overcome.

In this post, I want to say something on the costs, and what level of Basic Income (BI) might be possible, and some of the implications of that. I also want to pick up a few more points that I should have made first time around.

 

Firstly, A Confession

I probably should have realised earlier that my plan to comment on and model the costs of a BI in a simple way were not realistic. That’s because  to set realistic BI goals, not only do you need evidence of market and individual responses to the BI level but you need to:

– understand the current benefits and social protection system in some detail
– understand the current tax system in some detail
– understand household/ family patterns of income and expenditure in some detail
– understand work on living wage and minimum income standards
– understand the broader labour market

So in what follows, I have done my best but lack the information (and ability!) to do more than make some very basic points. My own pretty basic spreadsheet is here:

Basic Income Calculations Feb 2017

And that draws heavily from particularly a single report from the IFS that provides an overview of the benefits system. I also get my numbers on the tax side of the discussion from the RSA BI report that I quoted from last time (1).

So, feel free to challenge the numbers but I drew some comfort from arriving in my own simple way at roughly the same numbers as the RSA.

 

How Much Does the Benefits System Currently Cost and Which Parts Are Relevant to BI?

I draw my numbers for the benefits system costs from the afore mentioned IFS report (2, Table 3.1). It’s worth saying that these numbers are before the full introduction of Universal Credit. I think it makes sense to use these numbers for two reasons:

– the new UC system is still being implemented and numbers for it are hard to come by

– the older system is more familiar to most, and handily splits the types of support needed by type of recipient.

The IFSr report  (2) summarises the current system  here:

(for those whose eyes are struggling, its Table 3.1!)(note that this is far from being all types of social protection expenditure).

Very broadly, including all types of recognised social  tax and benefit spending (but excluding some forms of social protection such as social care), then the costs of the UK benefit system in 2015-16 were roughly £211 billion. With about 64m citizens in the UK, that gives a starting point on average of £3291 if we just divide one number by the other.We could call this a ‘smearing across all’ option.

But as in Part 1, I think it’s clear that we can’t just sweep away all benefits and provide a general sum- people who are sick, disabled and older, and who experience varying housing costs, need to be treated differently.

So, in my basic spreadsheet calcs, I considered two options. Option 1 excludes pensions and housing benefits, removing respectively £98bn and £24bn of spending from the £211 billion sum that could be used for a BI scheme, but also reducing the number of people receiving it from 64m to roughly 51.8m (i.e. excluding pensioners only). Option 2 does the same but also excludes special payments for maternity/paternity, disability, winter fuel payments and free TV licences. Most of those latter items are quite small but the exclusion for disability and sickness is large.

Under these two options, Option 1 allows about £92bn to be used to fund a BI, whereas Option 2 allows only £45bn. Not surprisingly, both options reduce the BI that could be provided from the ‘smear across all’ figure of £3291, to respectively £1772, and £875 per annum.

Keep reading because this is only half the calculation but I think it does highlight a few points:

-there is a big difference between just assuming the entire benefit system spend can go to a BI, and what happens if you then assume some people and recipients will have differentiated needs.

– if we exclude pensions, then a large spend is removed, with quite a large recipient base removed as well (but with spend removal proportionately larger than recipient removal)

– if we start to remove politically popular smaller benefits such as free TV licences, the effect is modest. However accepting that pregnant women and new fathers have different needs seems a no brainer and changes the picture very little, and is also a modest impact so seems to make sense

– However under Option 2 the exclusion of a further £41bn for sick and disabled people makes quite a difference, as I make the assumption that they receive both current payments and the BI. That seems a fair assumption to me, but I lack the data to confirm it.

Clearly, if we stopped here, the BI figure is low and the game ‘looks a bogey’ as they say.

 

We Need To Include Tax Changes as Well

I confess from here on, and with time constraints, I’ve had to fall back on the numbers quoted in the RSA report ((1), page 25 Table 3).

Table 3: Savings from benefits, tax reliefs and allowances34

Policy

Cost (£bn)

Child Benefit and Child Tax Credits

34

Working age benefits (Income Support, JSA, etc)

27

Working Tax Credits

7

Administrative savings and Tax Credits written off

10

Student grants and loans written off

3

Personal allowances (income tax)

68

Primary threshold and self-employed reliefs (NI)

23

State Retirement Pension, SERPS, S2P, Pension Credit, and MIG

90

Higher rate tax relief on pension contributions

10

Total

272

Note(s): Savings based on 2012-13 rates.

I have simply taken (some of) the tax changes proposed there and assumed they are correct and incorporated them into my ‘model’. These changes are to remove personal allowances entirely, to remove NI reliefs and thresholds, and to remove higher rate tax relief on pension contributions. These provide a handy £107 billion to go towards funding a BI. When that figure is added to my two options (i.e. £107+ either £92bn or £45bn), then the BI rate becomes £3836 under option 1 and £2940 under option 2. In passing I should say I assume a 75% administrative savings at DWP rather than the quite conservative 50% that the RSA use.

Again, there may well be basic errors and distorting assumptions in my simple arithmetic – do let me know if you spot them!

For what it’s worth, UK GDP is around £1900 billion, and current benefits and tax spend is around 11-12% of GDP, or around 30% of total government expenditure. Therefore each 1% of GDP diverted via tax to governance spending raises  around £19 billion.

Current UK expenditure by type is here: (taken from Budget 2016 report (3)

The point of showing this is that I couldn’t see an easy way to raise the BI level substantially- to have a BI of £7000 means there is a ‘gap’ to fill of c£210 billion or 11% of GDP compared to my two options. To create a BI of £12,000 appears to need a further £469 billion or 24% of GDP. I don’t see either of those propositions being realistic, ever. What I mean is that regardless of the need to phase in or generate citizen support, that those levels of GDP allocated to a BI are simply unrealistic, distorting and would surely distract from other needs.

But perhaps I am missing something?

I also lack the data to work out what further changes can be applied to the tax system to create scope for further increases to the BI. No doubt, some will argue that taxes need to go up further and that seems sensible. See for example this from Malcolm Torry on alternative means of creating a BI.

But each 1p income tax rise raises somewhere between £5-10 billion  p.a only, and even to raise the BI by a further £1000 a year would seem to require £50-65 billion-  a very large tax increase indeed. It seems to me, that if such a tax rise were feasible, I’d want to consider a much broader range of things as candidates for the extra cash- including new housing, new low carbon infrastructure, additional health and particularly mental health spending, and education including perhaps free tuition again.

So, I leave my basic calculations on Basic Income here!

 

The Link To A Decent Standard of Living? 

I realised in the course of these simple sums that I was approaching the problem by considering what could be funded, rather than what was needed. What I also needed to consider is the extent to which a BI at these levels made any difference to living standards or in providing a better safety net. After all, this is one of the claimed benefits for a BI.

In writing this, I drew upon a few reports on the Living Wage, the calculation of the Living Wage and its relationship to the Minimum Income Standard that underpins the Living Wage calculation.

The key table I draw on for reaching conclusions is this one:

 

In a nutshell, it seems to me that the BI levels I discussed above are very similar to current benefit rates (with the possible exception of sickness and disability benefits being paid on top). Because I exclude pensions, and because I lack the dynamic model to test things, it is hard to conclude other than that the BI at the level above does not significantly impact on a citizens contribution to having a decent standard of living. There are other benefits, but assuming that a  BI allows for a minimum income standard to be met, at a level of £2900-3800 a year, won’t match up with the MIS calculations.
I may be wrong on the contribution to living standards,and others may want to show me why. Nonetheless, it seems to me that a BI of c £3000 would be an excellent thing to have, for the reasons outlined in Part I, but that as I also said in Part I, we need to look to a much wider range of interventions to get us to the basic level of security and dignity that a society based on liberal principles should look for.
Briefly on why exclude pensions? Partly because older people are likely to have higher needs so it makes sense to separate that out. Partly because, intuitively, it makes sense to me to go more slowly, and consider folding in the pensions system in due course. I also think that this element of ‘pay as you go’ contribution works reasonably well, and is accepted, so why change it? With a fair wind, the new Single Tier pension should address a lot of previous pensions anomalies.

 

 

Other Issues

 

As ever, I’ve gone on too long and won’t win any prizes for brevity. But before I sum up, I did want to mention a few other points that perhaps I should have covered in Part I.

One the importance of the pilots, I want to re-emphaise that we really need to understand in some detail, and in a real world way, how people will react to the scheme. To take just one example, the Working Tax Credit (WTC) created a strange incentive whereby marginal benefit losses were much lower if you worked 16 hours rather than 15 hours. Result? Lots of people working 16 hours and not 14 hours a week, regardless of whether that suited them or was socially sensible.

Secondly, and perhaps controversially, is the issue of whether a BI should be limited in some way to only applying to a certain number of children? On the one hand, purists might say ‘absolutely not’ since the whole purpose of a BI is to be unconditional and non-withdrawable. Additionally, if we don’t provide a BI to families with large numbers of children, we risk the future of those children. On the other hand, it seems there would be nothing to stop, under a BI scheme, a family to have 8 children, with no thought to how to manage, and scoop up £25,000 from the state, with no requirement to work for doing so. This is a tricky one and I don’t have the answer. Perhaps, perhaps, the way forward is to limit payments of BI to no more than 3 children?

Thirdly, the BI allows the removal of the current absurd anomaly in the Child Benefit payment system, where a supposedly universal benefit is now means tested. But because of the daftness of our system, those who have a single higher earner in the family but a lower overall income, lose child benefit whereas those two earning split, do not. It is this sort of ridiculous unfairness that a BI can solve.

Fourthly, the more I dig into the UK benefits system, the more detail I find. Things such as passported benefits, Council Tax relief and so on, are also very important but I’ve not seen BI reports considering them, or how they should be treated. Any takers?

Finally, the current benefits system relies on a passporting system- that is, if you are eligible for certain benefits, you are automatically eligible for things such as free dental checks, free eye car, free school means, disabled parking schemes  and so on. More here. These are really important benefits and any BI would have to consider how to minimise administration if they were to be retained separately. Or perhaps they are folded into the BI scheme? I don’t have numbers to hand on the costs of such benefits.

 

Conclusions 

 

I’ve gone on far too long over two posts so time to summarise. To repeat from Part 1, I think there are clear benefits to a BI

– it gives greater choices and fosters human dignity, especially for those on lower incomes

– it is easier to administer

– there could be substantial admin savings from having a less complex welfare system

– it has potential to promote entrepreneurial citizens’ action

– it provides a much stronger safety net (depending on level of income set)
– it will greatly reduce waste and fraud as it’s very hard to claim fraudulently
– a BI scheme faces the future in terms of automation
– can be phased in

I stand by those claims-but with some caveats. Firstly, it seems to me that it is perfectly feasible to introduce a BI at around £2900-£3800 per annum. Secondly, it seems to me that further changes to the tax system might allow that to be raised a little further.

However, I can’t see a way to generate the very large BI payments that some call for, and I think at these levels it is clear that BI is only one component of an integrated approach to modern, fair work. As I said in Part I, it seems to me

‘… that BI needs to be part of a wider package of fair work which includes things like pay ratios, higher taxes, how we tax wealth and capital, and worker empowerment.’

My work in preparing Part 2 simply confirms that. We need to think more radically about wages, fairness, pay and empowerment in our system of work.

I also think we need to consider a wider range of UK benefits in future designs for a UK BI, such as the passported benefits or Council Tax reduction.

But, overall I am excited about BI schemes, and convinced, that done well, they really can make a difference. Thanks for reading.

 

Notes

 

(1)  The Royal Society of Arts Creative citizen, creative state: The principled and pragmatic case for a Universal Basic Income 2015

(2) Institute for Fiscal Studies A Survey of the UK Benefits System IFS Briefing Note BN13 2016 Accessed at https://www.ifs.org.uk/bns/bn13.pdf

(3) HM Treasury Budget 2016 report, accessed at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/508193/HMT_Budget_2016_Web_Accessible.pdf

(4) Centre for Research in Social Policy Loughborough University accessed at http://www.lboro.ac.uk/research/crsp/mis/results/

 

Author: DaveGorman

An Englishman longtime in Scotland, interested in new ideas for liberalism that recognise our challenges in the 21st century. Loves clouds, ideas, environment and applying liberal thinking to make things better. Speaking in a personal capacity of course.

Leave a Reply

Your email address will not be published. Required fields are marked *