In March 1999 Tony Blair promised to abolish child poverty by 2020, and halve it by 2010.
Statistics out tomorrow are likely to show that this halfway target was missed.
The parties will have a big squabble over their respective records. Labour supporters will say that a lot of progress was made. Coalition supporters will wonder whether there was as much progress as there should have been, given the billions spent. And there will be another round of debate about changing the target.
You may well groan at this point. I apologise. I know it's something I have written about here and here and here. And Policy Exchange has published on it here and here and here. And we had a big debate about it with Frank Field just recently.
But please forgive me one more blog on this. While the coalition government have widened the definition of poverty and social mobility, it has so far stuck with the existing target left behind by the last government.
While I believe having a child poverty target is a good thing, the current target urgently needs to change.
Let's rewind to the start.
Blair's simple-sounding goal to abolish "poverty" was put into practice in the form of a very specific target: to reduce the number of children living in households with less than 60% median equivalised income.
The specific nature of this target is what has caused all the fuss. It effectively conflates poverty and inequality. While other targets were added later, this measure is still used synonymously with the word "poverty" by politicians and the media. It is the number that politicians are referring to when they talk about how many children are in poverty.
Given it is so widely and unthinkingly used in Westminster, there are some counter-intuitive things about such a definition of poverty:
- If the average income goes up, but the incomes of people lower down the income scale stay still, then the measure that the government is using says that “poverty” has gone up. In contrast, recessions can appear to "reduce" poverty because they reduce the average income, but leave the income of people at the bottom reliant on benefits unchanged.
- Such a definition means there are more people in "poverty" now than in the 1970s despite decades of material progress. Back in the silver jubilee year of 1977 only half of households had central heating. Now almost everyone does.
- And this definition means there is more "poverty" in Britain than much poorer but more equal countries like Slovakia, Greece and Korea.
These quirks should certainly make us careful about treating the 60% median relative poverty measure as if it is neatly synonymous with the word "poverty". In fact, asked what it means to be "in poverty" in Britain today, most people (70%) think it still means not having enough to eat, or a place to live. Few members of the public endorse the more expansive definitions of poverty favoured by many academics, which equates poverty with differences in relative incomes, or some people not having goods that other people have. Only 7% think that poverty is about not having things other people have, and only 18% think it means having nothing beyond your basic needs.
But these are not the main problems with the target. The main problem is that it is only measuring income, not services or needs. That means that:
- While giving people more cash benefits counts towards the current poverty target, giving children a better education does't.
- Likewise, increasing the number of people in work does surprisingly little to reduce poverty, because it pushes up the median income (in fact when non-parents take a job and leave the dole, child poverty goes up). So the current measure understates the importance of work in reducing poverty.
- If the government gives benefits to people who don't work, that counts as less poverty. But if it spends more on providing childcare for people who work on low incomes, that doesn't.
- Moving people on housing benefit from expensive private rented accommodation into a cheaper social housing makes poverty go up (because they need less in cash benefits).
- Government action to sort out the care system, or abusive parents, or drug and alcohol addiction – or any other causes of poverty – will not really show up in the current target in the near term.
To cut a long story short, if the government does doing things that reduce poverty over the long term, it is not "rewarded" by the current poverty measure nearly as much as if it just hands out benefit cheques.
Does this really affect government decision making?
Yes. For example, in his first emergency budget George Osborne decided to spend an extra £2 billion a year, saying this was to keep relative child poverty constant. The coalition government later squeezed childcare subsidies to pay for increases in child tax credits. But surely a conservative/liberal government should be incentivising work, rather than increasing benefits like child tax credit, which you can get even if you are not working?
But that's what Osborne has to do – as long as the current target remains.
Instead of a target that pushes politicians towards short termist solutions, and spending more on benefits, we need targets which target the root causes of poverty like worklessness and poor education.
I know I have said all this before. so in this blog I want to explore the implications of the current target a little, and have a look at what has been achieved so far. And I will try and say more about the limitations of benefits as a poverty fighting tool.
Who is in relative poverty and why?
Given the government’s current definition of poverty, which groups of people tend to be more or less poor?
Obviously people who are workless, or don’t have much work, are more likely to be in poverty.
Less obviously, because the target looks at "equivalised" household income, people who have more children are also more likely to be in poverty. Equivalisation works by taking the combined income of the whole household (all the adults income, from both benefits and work) and dividing it by the number of people in the household (children and adults are given different weights). The government, therefore, defines the “poverty line” as higher if there are more adults and children in the household.
To take an extreme example, a two parent household with seven children in it would require an income of about £1,000 a week to avoid being defined as in relative poverty. For this reason, the majority of large families are defined as being in relative poverty. The chart below shows how, as the poverty line goes up, the proportion falling below it goes up too.
Figures before housing costs, aggregate of 2006-2010
So how many people end up being defined as in relative poverty depends on how much work they do, how many kids they have, and how much money the tax credit and benefit system gives them.
The top two lines in the graph below show that the majority of children in workless households are defined as being in poverty (be they lone parent or couple, and however many children they have).
Very few two-earner couples, or lone parents who work, are defined as being in relative in poverty – unless they have a relatively large number of children.
For example, only 6% of two earner households with one child are considered to be in poverty. Interestingly, for a given number of children, "classic" one earner households are more likely to be defined as being in poverty than either couples, or lone parents.
How much work people do also matters. 15% of lone parents with one child in part time employment are in poverty. But only 9% of lone parents with one child in full time time employment
Figures before housing costs, aggregate of 2006-2010
What were Labour's reforms?
There were three main strands to what Labour tried to do on child poverty. Two of them I think were good, and one more questionable.
1) Attempting to increase employment, and increase the financial incentives to work.
The Government now spends £1.3 billion a year on working tax credits, which top up the wages of people in work, and so provide a stronger incentive to work. It spends another £1.5 billion on the childcare element of tax credits, which reduces the cost of (particularly lone) parents who go out to work. Labour also increased a number of other childcare subsidies, and in total, government now spends about £7 billion a year on childcare. This probably increased employment, particularly for lone parents, although lone parent employment was already rising beforehand. Labour also introduced the minimum wage which increases the incentives to work for lower paid people.
2) Attempting to improve education.
Government now spends a lot more on schools. Other things equal this would be a good thing, although structures and reform of the system are more important than how much is spent. Labour also introduced Surestart centres. While I am sceptical about how effective surestart is, early years education is certainly going to be an important part of improving education.
3) Large increases in benefits targeted on households with children
This is the more questionable aspect of the last Government's anti-poverty agenda. Labour rapidly increased spending on child benefit (now £12 billion a year) and introduced child tax credits (£27 billion a year). The overwhelming majority of total tax credit spending goes on the child tax credit, which is payable whether people are working or not. It is effectively a much bigger, means tested version of child benefit. About £7.6 billion a year is spent on tax credits for households who are not in work. It's good in so far as it tops up the incomes of both workless and working households. But unlike working tax credit, it has the disadvantage that it disincentivises work.
Spending on such benefits rose more quickly than average incomes in eight out of the thirteen years of the Labour government. For example, in the first three years of the child poverty strategy, a workless family with three children would have seen their benefits increased by over 30%. And during a brief period in which such benefits were not being rapidly increased (2005-2007) the relative measure of child poverty stagnated or rose slightly.
As Mike Brewer points out in a recent report for the Child Poverty Action group, "it it is extremely difficult to reduce a relative measure of poverty when the default position is that benefits and tax credits get increased each year by less than average wages."
Who benefitted the most from Labour's reforms?
Given this approach to fighting poverty there have been improvements from both increasing lone parent employment (some of which the last government can claim credit for) but most of the reduction in child poverty has been the result of increasing benefit spending. The largest proportional decreases in poverty rates have been for those with large families and those who are workless.
Figures after housing costs
Polls show the public are split on how the tax and benefit systems should respond to people having children. Some people think the more children you have the more money you should get in benefits. Others think that's a choice you made and that you have to take the consequences. The current target implicitly endorses the former view.
It would always gave been preferable and more sustainable to see poverty falling as a result of increasing employment. And given that the government now has a massive deficit (and has been borrowing a quarter of all the money it spends) there is now really no other option.
I would suggest that the emphasis should move from increasing transfers to encouraging work. For example, government could shift spending from non-work-related benefits like child benefit and child tax credit, and spend the money either on in work benefits (like working tax credit) or other measures to increase employment (e.g. improving the quality of jobcentres). But the current target would strongly discourage any politician from doing such a thing: in the short-to medium term it would make the child poverty numbers look way worse. Moving money from welfare spending to education (as Tony Blair promised to do) would look even worse given the current target. The beneficial effects this might have on relative poverty in the long term would appear decades later. Given that a week is a long time in politics, that's a big problem.
Benefit spending: the right tool for the job?
In principle, if we spend enough on benefits we can hit the target. George Osborne "just" needs to spend an extra £19 billion a year on benefits and we could hit the target tomorrow.
But that wouldn't be a good idea. Increased benefit spending has multiple side effects. Firstly it discourages work. It is disappointing that more progress wasn't made in reducing worklessness during the boom years, and more progress could have been made if resources hadn't been spent on non-work related benefits. The higher taxes that pay for such transfers impose a further cost on the economy.
And according to the Institute for Fiscal Studies, the large increase in benefits for low income people with children also increased fertility: "there was an increase in births (by around 15 per cent) among the group affected by the reforms." Having more children may be a good thing, but it will also have offset some of the poverty-reducing effects the increased benefit spending could otherwise have had.
Given they spent loads of money, enjoyed an unprecendently benign economy, and had a massive majority, it would be truly disappointing if nothing had been achieved by the last government. And there was indeed real progress. In fact I feel bad criticising their record, because their intentions were the very best. But I can't help thinking that we could we have done better with a different approach.
A new, more balanced approach might involve looking at a range of different measures – including unemployment and education directly. The government might want to think about in-work poverty, and worklessness in different ways, because they require different solutions. And it would be nice to start properly tracking some of the most potent causes of future poverty – like children growing up with addicted parents, or in the care system.
One way or another, they need a smarter approach, because things for this government will be much, much harder.