Category Archives: skepticism

Truth, lies and tax credits #1 – Osborne’s odious comparison

Osborne’s claim: Summer Budget 2015

The original Tax Credit system cost £1.1 billion in its first year. This year, that cost has reached £30 billion

The claim was repeated in the Telegraph a few days ago.

A system that cost just over £1 billion in its first year ended up costing £30  billion a year.

A thirty-fold increase in costs would certainly suggest things gone awry. But a thirty-fold increase is hard to reconcile with the government’s own research.

in most years [1997-2011] spending on working age people and children accounted for between 12 per cent and 13 per cent of Government spending

The truth is that Osborne is comparing two very different things.

The history

When Labour came to power in 1997, government provided support for families with children:

  • via the (at that time) non means-tested child benefit
  • via family credit, a means-tested benefit for working families with children 1
  • via the married person’s tax allowance, and the additional person’s tax allowance available to unmarried parents
  • via child personal allowances paid as part of income support and income-based jobseeker’s allowance, these being means-tested benefits paid to out-of-work claimants

In addition, a relatively small number of people with disabilities received in-work support via disability working allowance.

The 1999 scheme – the first tax credits

In October 1999 the Labour government introduced two new benefits:

  • working families tax credit, replacing family credit
  • disabled person’s tax credit, replacing disability working allowance

These were essentially a re-badging of family credit and disability working allowance. Structurally very similar, they were substantially more generous than the benefits they replaced:

  • increased amounts could be paid for adults and each dependent child
  • the income figure at which benefit began to be withdrawn was increased
  • the taper rate at which benefit was withdrawn due to rising income was reduced from 70% to 55%
  • child-care support began to approach realistic levels

It’s this 1999 scheme, which covered low-income working families only, to which Osborne refers when he talks about expenditure of £1.1 billion.

What he doesn’t say is that that figure:

  • only covered six months – from October 1999 to March 2000
  • only covered three months of tax credit payments for a typical claimant

The latter applied because family credit was awarded for six months at a time. Claimants were not permitted to end their award early in order to switch to working families tax credit. In October 1999, on average, a family credit claim had three months left to run. By the end of March 2000, former family credit claimants had therefore only enjoyed the increased income from working families tax credit for an average of three months.

Children’s tax credit

In 2000-2001 the government replaced both the married person’s tax allowance , and the additional person’s tax allowance with the short lived children’s tax credit, a tax allowance aimed specifically at parents.

The 2003 scheme – working tax credit and child tax credit

In April 2003 Gordon Brown introduced a new tax credits scheme, structurally very different to anything seen before in the UK.

Child tax credit replaced all of:

  • the per child allowances in family credit; and
  • the child personal allowances in income support and income-based jobseeker’s allowance; and
  • the children’s tax credit – this effectively became the family element of child tax credit which could be paid to claimants earning well over £50,000.

Working tax credit replaced the remaining parts of working families’ tax credit and disabled person’s tax credit. It extended entitlement to some workers who were neither parents nor disadvantaged due to disability.

The personal allowances for children which formed part of income support and income-based jobseeker’s allowance were not immediately removed from existing claims – migration to child tax credit took place over several years, with the cost of child tax credit rising accordingly. Now nearly all state child support other than child benefit is delivered as child tax credit.

The comparison

So £1.1 billion represents the cost of:

  • in-work support only
  • covering a six month introductory period
  • with only three months on average being paid under the new scheme

£30 billion: the full-year bedded-in cost of:

  • in-work support for low income workers and their families
  • support for children in non-working families
  • replicating support previously given to parents as a tax allowance

And of course no adjustment has been made for inflation or changes in the composition and numbers of households.

None of this is to say that tax credits did not become more generous – and undoubtedly the 2003 scheme had many faults. But Osborne’s comparison is neither useful2 nor fair.

  1. Family credit cost about £2.5 billion per year prior to the introduction of working families tax credit 

  2. Of course it’s politically useful 

On a Citizen’s Income Scheme

The citizen’s income is an unconditional non-means-tested allowance paid individually to all citizens.

On the recommendation of Anthony Painter at the RSA, I took a look at a citizen’s income scheme proposed by the Citizen’s Income Trust (CIT). Their scheme covers most basic needs, but not housing costs or benefits paid because of disability. The CIT includes indicative costings, and claims that the scheme is broadly cost-neutral once tax changes are taken into account.

Housing costs are excluded because of their variability: a national flat-rate citizen’s income sufficient to support private-sector London rents would be far in excess of that needed elsewhere.

A key objective of the CIT scheme (as with most citizen’s income schemes) is to eliminate or greatly ease the poverty trap created by the sharp withdrawal of means-tested benefits as other income rises. The scheme I looked at does not meet this objective.

The CIT assumes that housing benefit and council tax support remain in situ – pretty much as they were in 2012 before the recent cuts. Both are means-tested. In broad terms, where income exceeds a claim specific amount, housing benefit is reduced by 65p for each additional pound of net income (gross income less tax/NI). Typically, 20p or more of council tax support is also withdrawn for each additional pound of net income.

Suppose I am a tenant responsible for rent and council tax. I have my citizen’s income, intended to cover my basic needs other than housing. I then receive means-tested housing benefit and council tax benefit on top. Because I don’t have any other income, my housing and council tax are covered in full.

I then get a part time job and my income goes up by £100 gross on which I pay £32 tax (employee’s national insurance is bundled into income tax in the CIT scheme, personal allowances are abolished.) This leaves me with £68.00. Assuming my housing benefit and council tax support are calculated as they were in 2012, my housing benefit is then reduced by 65% of this, i.e. by £44.20, and my council tax support by 20%, £13.60. Total deductions are £89.80, making me just £10.20 better off. This compares rather poorly with universal credit (plus the separate council tax reduction), and is little different to the pre-universal credit schemes.

What would it take for the CIT scheme to eliminate the poverty trap?

You would need to pay everybody’s rent and council tax in full. Merely retaining housing and council tax support is insufficient.

Alternatively, fix the housing system so that everyone has access to well-maintained, secure and affordable housing.

Winners and losers

People with inherited property and modest trust funds would do well from the CIT scheme. Lone parent tenants in need of childcare would do very badly. Unlike the current system, the CIT scheme excludes:

  • support for childcare costs.
  • support for owner-occupier housing costs
  • additional amounts in means-tested benefits paid to people with disabilities or caring responsibilities

So if a CI scheme doesn’t spring the poverty trap, what’s the point?

Ultimately, the prospect of a decent life for all. But CI schemes conflate several different hopes:

  • reduced administrative costs
  • low marginal deduction rates (combined impact of lost benefit and increased tax as income rises) CIs are means-tested in practice, it just happens via income tax
  • removing couple penalties – all entitlements are individual
  • income redistribution (popular on the left)
  • payment in lieu of government provided services eg health (popular on the libertarian right)
  • removing conditionality (no need to sign on, look for work etc)

Payment in lieu of services appeals on the grounds of maximising personal autonomy, but once services are devolved to the person in this way, it’s much easier for them to be cut – reducing the CI is easier than sacking care workers, just as allowing a private steel company to go bust is easier for the state than eliminating part of a nationalised industry.

Reduced administrative costs are often cited as a major benefit of CI. Although there are savings to made, I suspect they’re commonly exaggerated. Many schemes will bring more people into the tax system. Any scheme which keeps separate housing benefits and benefits for people with disabilities (as in the CIT scheme), will equally keep much of the administrative cost. The Green Party’s CI scheme includes an additional £80.00 for lone parents, rightly acknowledging the additional costs of single parenthood. But adding it to the scheme means that there now has to be a bureaucracy determining the status of single parents – and pairs of single parents who become one family will lose £160 per week.

Removing or significantly reducing conditionality is perhaps the most immediately achievable goal – though it goes against the prevailing establishment view that being poor is a condition deserving of punishment in itself.


Dr Malcom Torry from the CIT has kindly forwarded me some new intermediate CI schemes developed by the trust. These retain much of the existing benefits system – and therefore much of its complexity, but usefully discuss how progress might be made towards a citizen’s income in the short to medium term.

Nielsen’s web placebo considered harmful

“If your poem isn’t working, cut off the last line. Repeat as necessary.” – Michael Donaghy

There’s a curious statistic floating about the internet. It’s not always expressed in exactly the same words, but it comes from Jakob Nielsen.

“On the average Web page, users have time to read at most 28% of the words during an average visit; 20% is more likely.”

Almost every part of this claim might benefit from elucidation. Average web page? Average visit? The time users have? Those reassuringly specific numbers?

Nielsen based his findings on data from a German study. 25 academic staff had their web browsing habits monitored over an extended period, and the authors provided Nielsen with their data – just under 60,000 page visits. He then excluded all page visits:

  • lasting more than 10 minutes or fewer than four seconds
  • to pages with fewer than 30 words

Assuming a reading speed of 250 WPM, he graphed the maximum percentage of words readers could have read, given the time they spent on each page, against the number of words on the page, noting that this percentage falls very rapidly as page length rises. Further noting that the average (presumably mean) page length in the data set was 593 words, he looked at the corresponding point on his Y axis and arrived at his 28% upper bound. 20% is then a guesstimate to allow for user orientation before reading begins.

We now have some definitions:

  • Average web page: a web page of 593 words.
  • Average visit: the average time some staff at an academic institution spent reading web pages 593 words long.
  • Have time: the time the reader happened to spend on the page.

A very bad thing?

But let’s suppose that people really are reading 20-28% of web pages. Generally this news is delivered mournfully, often prior to some hectoring on why we should be taking a butcher’s knife to our content. I haven’t seen anyone recommend a specific number of iterations, but it’s clearly a procedure with rapidly and literally diminishing returns.

But why would we ever expect anyone to read 100% of a page in the first place? Could 28% actually be a generous percentage? Here, perhaps, there’s a bit of sleight of hand. When we talk about an average page, average takes on its usual connotations of regular, typical, standardordinary. Content you might hope some users would read from start to finish. But the study Nielsen got his data from looked at all web browsing, not just content intended to be read in full – presumably including, amongst other things:

  • search results (the median time spent looking at Google search results was reported as being eight seconds, so many of these would have been included by Nielsen)
  • navigation pages and directories
  • calendars, weather forecasts and other services where we may just want a single piece of information.

For search results and navigation, we would generally be happier if users found what they wanted quickly. If the first Google result is the one I want, I may have only read five percent of the page. That isn’t a sign that Google needs to work on creating stickier content, it’s a sign that Google’s search algorithm is working well. Similarly, weather forecasts and calendars may just be glanced at, but if four seconds or more elapse before the user moves on, they get scooped into the data-set.

A web placebo?

Perhaps the (mis)use of this statistic is most charitably thought of as aphoristic, as a useful provocation rather than something meant to reflect an underlying truth – as a web placebo. But when such shibboleths harden into points of doctrine, it’s probably time to consider whether they do more harm than good.