Official statistics and the 'neighbours from hell'

I have written today to the UK Statistics Authority to raise some questions about the government’s figures on “troubled families”. In December the Prime Minister explained:

Today, I want to talk about troubled families. Let me be clear what I mean by this phrase. Officialdom might call them ‘families with multiple disadvantages’. Some in the press might call them ‘neighbours from hell’. … We’ve always known that these families cost an extraordinary amount of money, but now we’ve come up the actual figures. Last year the state spent an estimated £9 billion on just 120,000 families – that is around £75,000 per family.

The same figures have been repeated in a series of government statements, including material from the Department of Communities and Local Government, the Home Office and the DWP.

The UK Statistics Authority exists to guarantee the integrity of official statistics in the UK. They have established a range of criteria for integrity, transparency and quality, but among other requirements they state that departments should

  • “Ensure that official statistics are produced according to scientific principles”
  • “Publish details of the methods adopted, including explanations of why particular choices were made.”
  • “Issue statistical reports separately from any other statement or comment about the figures and ensure that no statement or comment – based on prior knowledge – is issued to the press or published ahead of the publication of the statistics.”

That is not what’s happened here. “We’ve come up with the actual figures”, the PM’s statement says, and policy has been rolled out from that starting point. Some explanation of where the figure of 120,000 families come from appeared in a note from the Department of Education, though it was not publicized; there have been trenchant criticisms from Jonathan Portes and Ruth Levitas, on the basis that there is no connection between the indicators used to identify troubled families and the problems of crime and anti-social behaviour. The basis of the costings is still not publicly available. I’ve asked the Statistics Authority to consider whether there has been a breach of their Code of Practice.

Some figures to correct

On Thursday, when I reviewed some of the Daily Telegraph‘s figures on benefits, I took it that some of the wilder statements were down to the Telegraph reporter. Discovering similar figures in the Daily Mail made me wonder if they didn’t come from a common source, and on checking I found that they did: the main source is the Secretary of State for Work and Pensions.

In his speech to the Conservative conference at the beginning of October, Iain Duncan Smith said this:
“just before the election, in one year, Labour spent £90 billion on working age welfare – the same as the entire education budget for that very same year. … To put this in perspective, by 2010 this increase in welfare spending cost every household in Britain an extra £3,000 a year in tax.”

The Cabinet Office report for the Coalition, The State of the Nation (2010), claimed that “At least 12 million working-age households receive benefits each week (including tax credits and Child Benefit) at a cost of around £85 billion per annum” (p 10). With 20.4 million households in the UK, the cost per household was £4166. The DWP’s figures for the cost of working age benefits in 1997-98 are £31.3 bn in nominal terms and £46.4 billion in real terms. However, those figures exclude Child Benefit at £7 billion and Family Credit, the precursor of Tax Credits, at £2.3 billion; so the nominal figure for 1997 is £40.6 billion, which at the time with 18.7 million households was £2167 per household. So it is not true, even taking the figures in nominal terms, that households in 2010 were having to pay an extra £3000 a year for ‘working age’ welfare.

A second point raised in the Telegraph was the cost of “troubled” families. I questioned at the time whether this was an ‘official’ statistic’; but I have since found that it has been used by the Prime Minister, the DWP, the Home Office and the DCLG, and I have to concede that it is. The Prime Minister said last year:
“We’ve always known that these families cost an extraordinary amount of money but now we’ve come up the actual figures. Last year the state spent an estimated £9 billion on just 120,000 families – that is around £75,000 per family.”

There are two key figures here. The first is the number of families, which comes from a confusion between families with disadvantages and families which cause trouble. Professor Ruth Levitas, of the University of Bristol, has made a detailed analysis of the claims. She comments: “The most charitable explanation is that their research is statistically incompetent.” The second is the estimate of cost. The only detail I have about the costing is a statement from the Home Office that £2.57billion of this is attributed to crime and criminal justice. I haven’t been able to find any basis for the costings, and Ruth Levitas didn’t either.

More on larger families

I took part today in a phone-in on Radio Scotland. Three main issues were raised.

The first issue is an assumption that people opt to live a life on benefits and have children as a way of increasing their benefits. This is a misunderstanding. Larger families tend to be older families (because it takes time to have a larger family). They come to benefits for other reasons – typically unemployment, disability or divorce – after a change in circumstances. Much the biggest group of people who get long term benefits are older people with disabilities. They are are less likely to have young children than others.

The second issue is an assumption that people who have lots of children gets lots of benefits, and those benefits will be lost if people work. For the most part, that isn’t so; it’s not how our benefits system works. Most of the allowances for dependent children on income-tested benefits have disappeared. People get Child Benefit whether they work or not. The main effect of having a larger family is either to change the calculation of Housing Benefit and Tax Credit, or to change the size of house that a family needs. Their rules apply to people in or out of work. If entitlements to larger families were cut, it would directly affect entitlements of people in work – which would have the opposite of the effect that Iain Duncan Smith is claiming. It would also create a penalty for combining families, another perverse incentive.

The third part of the story is that the money needs to be used elsewhere. Apart from the obvious rejoinder, which is that it won’t be used elsewhere – the government is trying to save £10 billion – it won’t save much either. What it will do is to create greater hardship for a particular group of claimants – 300,000 out of nearly six million. That 300,000 is responsible for more than a million children; on paper the cuts would apply to a minority of the children, but of course they would affect the welfare of all of them. I was troubled in the discussion by the repeated assertion from Alex Johnstone MSP that “welfare needs to be a safety net”, by which he means that it should be confined to the role of a safety net. It needs to be much more than that; it needs to protect people from disruptive changes that will push them into hardship. The proposal threatens to make matters much worse.

"Benefits encourage problem families, says Iain Duncan Smith"

This is the page 1 headline from today’s Daily Telegraph. Duncan Smith is suggesting that allowances for more than two children should be largely suspended. This is shaping up as a major assault on families who receive benefits.

The Telegraph article claims that

  • “many of those receiving money for large families had drug and alcohol problems… There are about one in five households where no one works and 1.5 million children are growing up with a parent addicted to drugs or alcohol.”
  • “Official figures show that 120,000 of the most troubled and difficult families cost the taxpayer about £9 billion a year.”
  • “Every household is now spending the equivalent of £3,000 a year in tax for welfare payments.”

They comment that “The benefits system is supposed to be a safety net — not a lifestyle choice to encourage people to have so many children they will always have to be dependent on the state”.

Here, I am something of a disadvantage – their job is much easier than mine. I am supposed to respond with evidence; they can just make things up. The actual evidence is not very strong, but here is what I’ve been able to dig up.

  • The figures for parents with drug and alcohol problems are reviewed in Children’s needs – parenting capacity, TSO 2011, pp 36 ff. They report estimates that 705,000 children live with a parent dependent on alcohol in the UK, and that in England, Scotland and Wales up to 360,000 children have a parent who misues drugs. Those figures may overlap, but they come nowhere near 1.5 million. There is no direct connection between these factors and being out of work.
  • The definition of ‘troubled families’, which I have considered in this blog before, is primarily based on material disadvantage, not on the presentation of social problems. No justification has been given for the estimated cost.
  • The Telegraph’s figure for “welfare payments” seems to be a rough estimate based on total expenditure for benefits. (Correction, 28th October: This figure is actually a mis-quote from the Secretary of State, who claimed that working-age benefits were costing £3000 extra over the period of the Labour government. The original claim is also inaccurate, and I have considered it in more detail in a subsequent entry for 28th October.))
  • Benefits can hardly be described as a ‘lifestyle choice’ if people don’t choose to stay on them. I have reviewed the figures for long-term dependency on this blog before. Most people who are dependent are pensioners; among non-pensioners, the vast majority are disabled; hardly anyone is continuously in receipt of Jobseekers Allowance.

The key point made by Duncan Smith is that benefits create perverse incentives for large families. With that in mind, I’ve reviewed the evidence about large families on benefits. The evidence is not strong; the best figures I was able to find from the DWP stop in 2007. The figures come from the DWP tabulation tool, based on their 5% sample of claimants.

The first question is how many ‘large’ families there are. There were 1,380,000 families with children in receipt of benefits – the other claimants were families without children. Of those, 190,000 had 3 children, 71,000 had four and 35,000 had five or more; that is 21% of the total. This is higher than the proportion of families with 3 or more children in the population as a whole, but we are talking about less than 300,000 families. Larger families on benefit tend to be older families; in the majority of cases (58%) their youngest child was at least five years old.

Further note, 27th October: Channel 4 has found better, more up to date information resulting from a Freedom of Information request. In May 2011, there were 1,354,280 families with children in receipt of benefits: 194,220 had 3 children, 76,310 had four 25,980 had five, 8780 had six, 3200 had seven, 1080 had eight, 360 had 9, 130 had ten and 50 had 11 or more. That comes to 310,110.

The second question is whether benefits encourage people to have children. I cannot tell from the figures how many people have children while they are on benefits, but at least I can say something about the age of their families.

Duration on oldest
current benefit		 All families with children At least one child under 5 % with a child under 5
Less than 2 years 499,600 272,100 55%
2-5 years 340,600 169,400 49.7%
5 years or more 539,700 136,800 25%
Total 1379,900 611,300 100%

The table shows that, while some families do have children while in receipt of benefit, families who get benefits for longer periods are markedly less likely than others to have children. There are obvious reservations to make about these figures – for example, I do not have the precise ages of the children, and I cannot break down families by the age of the parents – but there is no reason here to accept that people on benefits are being encouraged to have children.

Scotland and the EU

First Minister Alex Salmond has been accused of lying in a TV interview, when he said that yes, the government had sought the advice of legal officers about Scotland’s position in the EU in relation to debates and documents. In the Parliament, Salmond defended himself with chapter and verse about which documents he meant. I have just checked them out for myself, and I think Salmond has the right of it. Two of the three papers he cites, Choosing Scotland’s Future (2007, p 24) and Your Scotland your referendum (2012, p4) do make statements about Scotland’s position in Europe and will have passed the law officers, even if they are somewhat thinner than a proper legal consideration might offer. The main problem the Scottish government might have in giving a fuller account would not, I suspect, be the question of the confidentiality of advice; it would be that obtaining such advice would be a breach of the Scotland Act, which deliberately and explicitly prevents the government from contemplating the breakup of the United Kingdom. Following that line of enquiry has only been made legally possible following the Edinburgh agreement.

The more important question is where an independent Scotland would stand in relation to the EU. A helpful article last month by Alan Trench in the Guardian explains that while Scotland’s position is uncertain, it is debatable whether the EU could deny a Scottish application without breaking its own rules. Europe has a federal structure, in which every citizen is a citizen of Europe as well as of the member state; denying access to Scotland would deny citizenship to EU citizens.

Further note, 1st November: The press have caught up with this argument this morning, with an honorary member of the Commission confirming that EU citizenship cannot be withdrawn and that terms of entry would be negoatiated on that basis between a referendum result and independence.

The age of consent

Three news stories recently have focused on circumstances where under-age girls have been used by older men. In the Savile case, a celebrity regularly abused vulnerable young women. In Rochdale, a group of girls was systematically exploited by a cabal of pimps. In the third case, a fifteen-year old girl absconded with her teacher. There are several recurring themes here: illegality, the conduct of older men, breaches of trust, and in two cases at least, the failure of responsible authorities to protect the girls. The combination of celebrity and the role of the BBC has made the Savile case most prominent, but to my mind the Rochdale case is the most disturbing. The Times reported: “Policeman yawned loudly as girl, 15, described how she was forced to have sex.” The police had categorised this as a ‘lifestyle choice’.

Part of the problem here hangs on the way we think about ‘consent’. A 15 year old cannot legally give consent – but would any of these situations really have been acceptable if the girl was 16? There are reasons to think that the ‘age of consent’ is misnamed: we do not apply the same tests to a relationship with a boy or girl of the same age, and the law makes a distinction in the terms of the offence according to the age of the other person. The primary purpose of an age limit is to protect young people from exploitation from older people. And if we think about the issue in terms of the conduct of older people, rather than the independent judgment of the young person, the age at which protection from exploitation is withdrawn is too low.

Multiple disadvantage in Scotland

The Think Tank Demos has published a report about multiply disadvantaged families in Scotland. Families are described as “multiply disadvantaged” if they meet four or more of seven criteria:

  • low income
  • worklessness
  • no educational qualifications
  • overcrowding
  • ill health
  • mental health problems
  • poor neighbourhood.

There are 131,000 households who are multiply disadvantaged by this definition. Of those, 52,000 are pensioners, and 55,000 are households without children; 24,000 are households with children. The category of pensioners is based on 3 criteria of 6, because it excludes worklessness; overcrowding is not really a useful indicator either.

This is not the same test as the Westminster government uses for its definition of “troubled families”. That test is based on them meeting five of the following seven criteria:

  1. having a low income,
  2. no one in the family who is working
  3. poor housing,
  4. parents who have no qualifications,
  5. where the mother has a mental health problem
  6. one parent has a long-standing illness or disability, and
  7. where the family is unable to afford basics, including food and clothes.

The overlap between the criteria does make it plausible to suggest, though, that someone who is “multiply disadvantaged” in the Demos report will probably also score four or more on the “troubled families” score. As the Demos authors note, there is no implication that families who are disadvantaged in these terms present problems for other people – but there is no reason to suppose that from the criteria for “troubled families” either. And there has to be some suspicion about the tenor of a report which goes on to tie the characteristics of poverty to alcohol, drug use and child neglect – none of which applies to most, or even to many, of the families identified through these statistics.

The finding that there are only 24,000 families with children in Scotland who are multiply disadvantaged even on as few as four indicators does raise some questions about the direction of policy, which has tended to focus on the characteristics and culture of poor people as something set apart. There are some points to draw from the figures:

  • Scotland is a society where more than one person in six of working age receives an ‘out of work’ benefit, and a quarter of Scotland’s children are in low income households – but the vast majority of people in this position are not ‘multiply disadvantaged’ by the definitions in this report
  • showing that social problems are more prevalent than elsewhere does not mean that they are actually likely – most people who are ‘multiply disadvantaged’ do not have them
  • if the number of families who might be said to be multiply disadvantaged by these criteria is small, the numbers who might after that be said to suffer ‘intergenerational deprivation’ in these terms is, necessarily, smaller still
  • while multiple disadvantage is a legitimate cause of concern in itself, it is neither typical of poor families or commonplace.

It makes sense to design policies that can effectively reach people who are most disadvantaged. Poverty in Scotland is much more widespread, however, and it makes no sense to make such policies the basis for anti-poverty strategy more generally.

Concerns about disability benefits

The Hardest Hit Campaign is a coalition of groups concerned about the impact of targeting people with disabilities for cuts. Their report, The Tipping Point, records the concern of many people with the prospect of losing current benefits. It follows hard on the heels of Holes in the Safety Net, a report anticipating losses to disabled people from the introduction of Universal Credit, published by Citizens Advice, TheChildren’s Society and Disability Rights UK. Pat’s petition asks the government to “Stop and review the cuts to benefits and services which are falling disproportionately on disabled people, their carers and families”.

Two thirds of the benefits paid by the DWP go to pensioners, and those are currently protected. Of the remaining benefits, nearly half go to people who are sick or disabled. The cuts are falling on this group, because the government has left itself with little else to cut.

Who benefits?

In Parliament this week, David Cameron rejected arguments that benefits for people with disabilities will be lower. This is from Hansard on 17th October:

Mr Bain: … Last week the Prime Minister promised that work would always pay, but this morning Baroness Grey-Thompson and the Children’s Society have revealed that his current plans for universal credit next year will mean that up to 116,000 disabled people in work could lose as much as £40 a week. Does not that say everything about how this divisive Prime Minister always stands up for the wrong people? At the same time as handing huge tax cuts to 8,000 people earning over £1 million a year he is going to penalise some of the bravest strivers in our country.

The Prime Minister: The hon. Gentleman raises an extremely serious issue; let me try to deal with as fully as I can. The money that is going into disability benefit will not go down under universal credit; it will go up. The overall amount of money will go from £1.35 billion last year to £1.45 billion in 2015. Under the plans, no recipients will lose out, unless their circumstances change. All current recipients are fully cash-protected by a transitional scheme. On future recipients, we have made an important decision and choice to increase the amount that we give to the most severely disabled children, and there will be a new lower amount for less disabled people. That is a choice that we are making. As I have said, we are increasing the overall amount of money and focusing on the most disabled. That shows the right values and the right approach.

Some of the points here are unclear. There is no such benefit as ‘disability benefit’, so it is hard to tell just what the PM is referring to – it is not ESA or the Income Support that is going to be included in Universal Credit. I am not sure which benefit cost £1.35 billion last year, but my best guess is that it was Disability Living Allowance for children, which was forecast in the last budget to be £1.31 billion. If that’s right, the figure is irrelevant – this is not a benefit that is going to be affected by Universal Credit at all. Current recipients are only protected ‘unless their circumstances change’, and that will include movements into and out of work, and reaching school leaving age.

Probably more important than the precise figures, however, is the general message. The PM says here that the government is “focusing on the most disabled” and that this is “the right approach”. But this approach was apparently rejected by Lord Freud, the responsible minister, in last month’s evidence to the Work and Pensions Committee, when he explained that the priority was to liberate disabled people by giving them opportunities to work (see my note on A zero-sum game, 11th October). Is the government giving priority to those who are most disabled, or those who might be able to work? They are not often the same people.

Taser happy

It has been reported that police in Lancashire tasered a blind man by mistake. He was carrying a white stick and it was taken for a samurai sword.

I’ve followed this sort of case for a time, largely because I’ve undertaken previous work on complaints for HMICS (the Inspectorate of Constabulary for Scotland) and the Police Complaints Commissoner for Scotland. The main organisation dealing with these issues in the US is the National Association for Citizen Oversight of Law Enforcement, or NACOLE; they regularly report issues of this kind from the US. So, before Taser use in the UK became widespread, there were reports from NACOLE of a UCLA student tasered for refusing to leave the library, a great-grandmother tasered for arguing with a traffic policeman, and a deaf man tasered because he’d stayed too long in a store’s toilet and failed to come out when the shout of “police” was made. Given the track record, it was all too predictable that there’d be similar incidents in the UK as we took up the same technology.

Complaints against the police are generally levied against individual officers; there is a presumption that the officer who uses inappropriate force is individually at fault, and the complaint is routinely framed in these terms. This has things the wrong way about. The complaints that members of the public have about the conduct of officers are almost always complaints about the service, and it is never open to an individual member of the public to act against an individual officer. The problems stem from policy decisions, and it will only be through policy that adequate safeguards and controls can be developed.