Back in 2011, David Cameron vowed to crack down on what he called Britain’s “Shameless culture”.

Too many problem families were following in the chaotic footsteps of the Gallaghers, the feckless anti-heroes of Channel 4’s sitcom Shameless, and costing the state £9bn a year in the process, he said.

But the government was accused of shamelessly misleading the public about the nature of the problem.

Now ministers insist the Troubled Families scheme has been a success, and they want to extend it take in 400,000 more families. What’s in the small print?

The analysis

In the aftermath of the riots that rocked English cities in 2011, the prime minister said he wanted to see the lives of 120,000 of the most troubled families in Britain “turned around” by 2015.

The rhetoric in Mr Cameron’s speeches at the time made it clear that these families were not just troubled – they were causing trouble.

But it turned out that the criteria originally used to identify these problem families had nothing to do with crime or antisocial behaviour.

To be counted among the 120,000, families had to tick five of seven boxes relating to poverty and deprivation.

Critics like the economist Jonathan Portes and sociology professor Ruth Levitas raged against the misleading use of language across government on this issue.

Ministers were claiming, without any evidence, that people suffering from social problems were also more likely to cause trouble for their neighbours.

The government stuck to its guns and clung to the figure of 120,000 – even after changing the definition of a “troubled family” used to calculate it.

By 2012, you needed to tick three of four quite different boxes to qualify for help:

Councils were given detailed advice on how to identify the worst families, but were warned not to contradict the Westminster line that the total number should still add up to 120,000:

So despite the complete change of definition, the number of troubled families who needed intensive help was always destined to be 120,000. Someone must really like that number.

The money

There is good evidence that offering intensive support to families with multiple can make a big difference to their lives and save public money.

But it’s not cheap. Estimates of the average cost of such work per family ranged from £14,000 (still a figure endorsed by the Department for Education) to £25,000 when the policy was being drawn up.

The government decided £10,000 per family was enough, and it would only stump up 40 per cent of that: £4,000 per household.

Councils would have to make up the shortfall and they only got the £4,000 on a payment-by-results basis.

The government told local authorities: “We hope that you will take advantage of the opportunity to invest this relatively large amount of resource into a small amount of families to achieve lasting change for them.”

This was at a time when Whitehall grants to councils were being cut, with some of the poorest urban councils taking the biggest hit.

Now DCLG is planning to cut the amount of funding per family that comes from central government. A spokesman said families would get “a bit less” but could not give us any more detail.

Turned around

Nevertheless many councils have taken up the challenge, and the government says that the lives of 53,000 of the most troubled families have been “turned around” – which means truancy, antisocial behaviour and child crime rates in the family have dropped, or that an adult has found sustained work

This was announced with some fanfare, but it means that Mr Cameron is less than halfway to his target in the final year of the three-year scheme. The Department for Communities and Local Government (DCLG) says it is working with 110,000 families as we speak.

The government has now broadened the criteria for the people who are placed on the scheme.

Now families only need to meet two out of six criteria: crime or antisocial behaviour; truancy; “children who need help”; unemployment; domestic violence and abuse; a range of health problems.

There are said to be around 400,000 more families who tick two of these six boxes, 40,000 of whom will be dealt with now, a year earlier than planned.

Is the 400,000 based on more solid research than the 120,000? DLCG tells us there is a study that backs it up, but they aren’t going to publish it yet. They are “committed to doing so in due course”.

[Update: We understood incorrectly that the 40,000 families who come under the new definition would count towards Mr Cameron’s 120,000 target, but DCLG tells us this DEFINITELY WILL NOT happen. So the new families can’t be used to help hit that target. The government has to “turn around” all of the 120,000 original group to keep the promise.]

The verdict

Dodgy statistics have dogged this project from the start, the biggest problem is that the definition a “troubled family” keeps quietly changing.

This means, among other things, that it’s difficult to stand up the government’s claims that the most troubled families cost the state £9bn a year.

That figure comes from a report that uses a different, earlier definition of what a troubled family is.

We don’t know if this new figure of 400,000 more families needing help is similarly dubious, because the government has not published the research that claims to backs it up.

DCLG also won’t give us any details of the new funding model, but a spokesman told us the money paid for each family from central government coffers was going to be “a bit less”.

Still, the basic idea of problem families getting intensive support is still broadly backed by local government, charities and MPs from all parties.

The Labour MP Graham Allen, an early intervention pioneer who has written reports on the subject for the government, praised the work of the project so far but added: “The more people you include in the programme, the thinner the resource to turn them around on an inter-generational basis.

“We should avoid confusing giving help to difficult families with really in-depth change.”