by Stewart Lansley
The government has been aiming to shift the debate on poverty from relative poverty to life chances. So is it aiming to downgrade the goal of abolishing financial poverty?
In March 1999, Tony Blair surprised an audience of academics and policy advisers with an unexpected commitment – to halve child poverty within a decade and 'eradicate' it within 20 years. It was as unexpected as it was bold and raised a few eyebrows among those present. Few experts believed, given the scale of the task, that it would be possible to meet the targets in the timetable laid out. Between 1979 and 1997, the level of child poverty (measured in relative terms) had doubled. When Labour won a landslide victory in 1997, just over a quarter of children – 3.4 million – were estimated to be living in poverty.
In its first two terms, Labour made reducing poverty one of its principal social policy goals. Backed by a number of key policy innovations – from the minimum wage to the child credit scheme – relative child poverty fell to 2.7 million by 2004/5.
Then, during Labour's third term, for a mix of reasons – the level of resources, stagnant wage growth for many, and slowly rising unemployment from 2005 – this progress was halted and the child poverty figures climbed again to 2.8 million in 2008/9. As the government acknowledged, although some real progress had been made, the target of 1.7 million by 2010/11 was 'likely to be missed by a considerable margin'.
One of Labour's final legislative acts before it lost office was to make Tony Blair's earlier commitment legally binding with the passing of the 2010 Child Poverty Act. The Act set out four specific targets to be met by 2020:
In opposition, leading Conservatives (including David Cameron and the former leader, Iain Duncan Smith) had also made it clear that tackling poverty (and inequality) would be made a priority in government. As Duncan Smith put it in The State of the Nation Report: Economic Dependency published by the Social Justice Policy Group in 2006:
In modern times, poverty has been a difficult issue for the Conservative Party to deal with. However, as this Report makes clear, it is too important an issue to be left to the Labour Party. All forms of poverty – absolute and relative – must be dealt with.
The same report also went on to endorse a relative definition of poverty:
We should now say explicitly: Poverty must be defined in relation to changing social norms. We should reject completely the notion that poverty can be defined in absolute terms alone. Relative poverty matters because it separates the poor from the mainstream of society.
This commitment was echoed by Cameron himself: 'I want this message to go out loud and clear: the Conservative Party recognises, will measure and will act on relative poverty.'
In the same year – 2006 – Cameron also signed up to the target of ending child poverty by 2020. The Conservatives also voted for the 2010 Act, which was passed with all-party support. In the agreement signed in May, the Coalition partners further committed themselves to the aims laid out in the Act: 'We will maintain the goal of ending child poverty in the UK by 2020.'
Despite these outward commitments, however, some leading Conservative figures and their advisers have long harboured doubts about a number of aspects of anti-poverty policy, notably on the way it is defined. It is now clear that these doubts have been carried over into power. Shortly after the election, Iain Duncan Smith, appointed the Work and Pensions Secretary, gave the first of a number of hints of a rethink on the definition of and strategy towards poverty by declaring that he was unhappy with the official relative approach to defining poverty and especially the use of the median: 'You get this constant juddering adjustment with poverty figures going up when, for instance, upper incomes rise.'
Frank Field MP – one of Britain's leading authorities on the subject, and appointed by the Coalition to lead an independent review into poverty – shares this view. In an article in the Daily Telegraph a few weeks after Duncan Smith's comments, Field argued that the use of the median is essentially self-defeating:
Any candidate sitting GCSE maths should be able to explain that raising everybody above a set percentage of the median income is rather like asking a cat to chase its own tail. As families are raised above the target level of income, the median point itself rises. Not surprisingly, therefore, no country in the free world has managed to achieve this objective.
In fact, these assertions are false. If the rich get richer at a faster pace than the rest of society, as they have, there is no effect on the median – the poverty line, set at a fixed proportion of the median, stays exactly the same. If all households below 60 per cent of the median were to rise above it, poverty would be eliminated but the median would stay the same. Both Duncan Smith and Frank Field are, inexcusably or deliberately, confusing the median with the mean, which is the average income calculated by dividing the sum of all incomes by the number of people in the distribution. This can be seen easily by the income threshold approach interactive graph; this shows what happens to the level of poverty when household incomes are changed above and below the median and the mean.
Another sign of a fundamental rethink is that both Field and Duncan Smith have repeatedly asserted that fighting poverty is about more than income. As Field put it in his Telegraph article: 'Over recent decades, the Left and centre-Left's answer to poverty and inequality has been to spend more money, to redistribute from richer to poorer. Yet this central social democratic ideal is being tested to the point of destruction.' This has become something of a recurring theme. As Nick Clegg, the Deputy Prime Minister, described it in an article in November 2010, 'poverty plus a pound is not enough'.
In his Independent Review of Poverty and Life Chances report published in December 2010, Field argued that some of the money used to support children should be redirected from benefits to improving life chances. Following his report, on 21 December the government published a consultation document that set out 'a new approach to tackling poverty'. Again, a good deal of the document raised questions about the importance of monetary measures: 'The Government is clear that tackling child poverty requires more than simply treating the short-term symptoms of poverty or moving families across an arbitrary income line.'
To date the government has also failed to meet two of the requirements of the 2010 Act. First, it has failed to publish its child poverty strategy outlining the measures it intends to take to meet its legal obligation to eradicate child poverty by 2020. Due by the end of March 2011, within a year of the passing of the Act, its strategy has yet to be published. Second, under section 8, it should have established an independent Child Poverty Commission, which it should consult in preparing its strategies. Yet no such Commission has been established. In answer to these points, the Work and Pensions Minister, Maria Miller, has said in a parliamentary answer on 22 March that: 'Our plans for establishing a Commission will be set out in the Child Poverty Strategy, to be published shortly.'
The Conservatives, of course, are only too aware of the political sensitivities surrounding their approach to poverty, not least because of their past record. The line being promoted is that the government will seek to tackle poverty through a 'broader approach', one that prioritises an improvement in life chances.
No one could disagree with such an approach. Tackling poverty depends on a twin-based approach – raising low incomes and improving educational and work opportunities and upping parenting skills. The more successful the latter policies, the fewer people would end up in poverty. Labour accepted this by spending more public money on both income redistribution and on policies geared to improving life chances through programmes such as Sure Start, the New Deal, and Future Jobs Fund that were aimed at improving educational and parenting performance and access to jobs. The reduction in child poverty under Labour up to 2004/5 was in part down to the introduction of the minimum wage and more generous levels of child credit and family credit, but it was also helped by falling unemployment and policies to improve access to jobs through improved childcare provision and employment services.
There are two main factors underpinning the Coalition government's apparent rethink. The first is the cost of the 2020 commitments. As the Institute for Fiscal Studies has shown, meeting the relative target would require tax and benefit measures costing around an extra £19 billion a year at 2009 prices. Indeed, the cost constraint was recognised by Labour before they lost office:
while it may be possible to reach the targets in 2020 based on financial support measures alone, such a strategy would be both costly and unsustainable, because it does not tackle the causes of poverty. It is only through empowering and supporting families to lift themselves out of poverty that the 2020 targets could be met and sustained beyond 2020.
The second factor is ideological. In the past, some leading Conservative thinkers have shown a strong preference for measuring poverty in absolute rather than relative terms. In 1979, Sir Keith Joseph, Mrs Thatcher's first Education Secretary, argued that the needs of the poor should be defined in terms of subsistence needs only.
An absolute standard means one defined by reference to the actual needs of the poor and not be reference to the expenditure of those who are not poor. A family is poor if it does not have enough to eat… By any absolute standard there is very little poverty in Britain today.
A decade later, the then Social Security Secretary, John Moore, declared that we had reached 'the end of the line for poverty'. Moore believed that absolute poverty had been eliminated and that relative poverty was 'simply inequality' and therefore could be ignored.
This philosophy was effectively enshrined in Conservative measures between 1979 and 1997 and is the main reason why poverty and inequality soared during this period. In particular, from the early 1980s, benefits – from unemployment benefit to the state pension – were raised in line with prices not earnings. This meant that, although the average level of prosperity rose sharply under the Conservatives, benefit recipients found themselves slipping behind the rest of society. Labour greatly modified this approach. While they failed to meet their own targets on poverty, they at least ensured that, in contrast with the Conservative era, most sections of society shared roughly equally in the proceeds of growth, thereby keeping a firm lid on relative poverty levels.
Although the Conservative Party's approach is now less ideological, it is clearly planning some kind of a shake-up in existing anti-poverty strategy. Indeed, one of the remits of the Field review was to 'generate a broader debate about the nature and extent of poverty in the UK and examine the case for reforms to poverty measures, in particular for the inclusion of non-financial elements'. What the strategy document will contain is unclear, but the Coalition is more constrained than it would ideally like. The most radical option would be to abandon the Child Poverty Act targets altogether. This is unlikely. Such a move would require primary legislation and would provoke considerable flak for reneging on a strategy both Coalition parties voted for a year earlier.
What is more likely is a more subtle approach, change by stealth, downgrading or sidelining the relative target – as measured by the median-related threshold – and giving greater priority to other goals. This might involve a mix of giving greater weight to the absolute target and/or switching the emphasis from providing financial support to improving life chances by publishing a new set of life chance goals, with the Field Report providing the justification.
Some elements of this strategy are already in place. As part of its 2010 Spending Review, the government has already reduced some benefits and frozen the value of others. The level of child benefit, for example, has been frozen for three years, and will thus fall in real terms. It has also changed the inflation index (from the RPI to the CPI) to be used in uprating other benefits in a way that will mean they will rise more slowly than they would have done. These benefit cuts will increase the extent of relative and absolute poverty because they will lower the incomes of some groups of the poor, making meeting the 2010 targets an even more distant goal.
The government could take this further by freezing some benefits at their current real levels on a more permanent basis. In a BBC radio interview on the day of his report, Field proposed that ministers should consider withholding annual above-inflation increases in child tax credits and instead plough the money into early years' education. This would in effect mean that those dependent on benefit, both in and out of work, would steadily fall behind the rest of the population as long as their incomes rose in real terms. The government might try to justify such measures and the effective relaxing of the relative poverty target on two grounds. First, that the current measure using the median can lead to statistically perverse results. Second, that with limited resources, concentrating on life chances is a more effective way of reducing poverty.
In his final report published in December 2010, which concentrated on preventing poor children becoming poor adults, Field proposed augmenting current poverty targets by establishing just such a set of life chances indicators that will measure how successful we are as a country in making life's outcomes more equal for all children. These indicators included measures of parenting skill, the quality of the home learning environment and local nursery care. Field also made it clear that these measures should take precedence over financial measures.
A major limitation of the existing child poverty measures is that they have incentivised a policy response focused largely on income transfers. This approach has stalled in recent years and is financially unsustainable. A more effective approach is to use a set of measures that will incentivise a focus on improving children's life chances, and ultimately break the transmission of intergenerational disadvantage.
It is, of course, incontrovertible that poverty is more than a problem of lack of income, and Field is right to highlight the role played by these wider factors in determining the life chances of children. This has long been accepted and was acknowledged by the previous Labour government in its twin-based approach. Nevertheless, any move to sideline the financial and income targets, and especially the median income-related relative poverty target at the heart of the 2010 Act, would have a number of implications for the future of anti-poverty strategy and the extent of poverty.
The first implication of weakening the commitment to the relative target by, for example, freezing work and non-work related benefits in real terms, or raising the priority given to ending absolute poverty, would be to condemn the bottom fifth of society to living standards that would fall behind those in the middle of the income distribution. It would effectively sanction, as state policy, the acceptability of a rise in inequality between the bottom and the middle. This would mean a return to the experience of the 1980s and much of the 1990s when the poorest sections of society failed to keep pace with general rises in prosperity and a bigger and bigger income gap opened up between the poor and the rest of society.
Relativism is a core principle of a civilised society. Any rowing back on this principle would mean the introduction of a multi-speed Britain, with lower income groups condemned to the living standards of the past, as they slip further and further behind those higher up the income ladder. Successive surveys of public opinion carried out since the first 'Breadline Britain' survey in 1983 have looked at which items – from food and housing to clothing and leisure – are considered essential to avoid poverty in contemporary Britain. These surveys have found a clear public consensus that a minimum acceptable living standard is closely linked to contemporary and not past living standards. In the latest survey, conducted in 1999, the items identified by a majority as necessities that nobody should have to do without because they couldn't afford them included not just items like heating, medicines and visits to school such as for sports day, but 'having friends or family around for a meal', 'money to spend on self weekly', 'presents for friends/family yearly' and 'a holiday away from home'. These findings demonstrate that the public fully embrace the idea that poverty is relative. The poor should be guaranteed a minimum living standard that is determined by contemporary and not past lifestyles.
It is here that the debate about the use of the median is important. Using the median as the threshold certainly has limitations, in part because any measure of poverty based only on income is essentially partial. Income-based measures do not take into account financial resources other than income, or other financial deductions such as debt. And they don't take into account non-income-based resources, such as the level of service provision. The choice of the 60 per cent threshold has a limited evidential basis and is not directly related to measures of need or deprivation.
There have been two distinct approaches aimed at establishing an income threshold related to the income required to meet certain levels of need. These set poverty lines by an examination of people's living standards to establish what needs should be met and who falls below this standard. Adaptions of these methods have been used in the calculation of the second target contained in the 2010 Act. Moreover, there is some evidence that the level of poverty determined by these two quite distinct methods – the 60 per cent of median income and the thresholds based on more objective assessments of needs – generate very similar results.
More recently, as we have seen, the median has come under fire for another reason – that it can lead to perverse results on changes in poverty. The latest of these claims came in the 2011 Red Book – the detailed Treasury guide that accompanies the Budget Statement on 23 March:'
The way that child poverty is currently measured also means that policies that impact on median income can have perverse impacts on measured poverty by raising or lowering the poverty line. For example reducing the income tax paid by millions of lower earners or providing additional support to low-income pensioners could push up the poverty line and increase the number of children calculated as being in poverty.
The first implication of this statement is that if median incomes rise because, for example, wages in the middle of the income distribution are rising faster than at the bottom, this will lead to a rise in the poverty line and in the level of poverty. But this is not a perverse result – it shows that if living standards rise in the middle but not at the bottom, relative poverty will have risen. The second implication is that anti-poverty policies are self-defeating. Yet extra help for the poorest, for example, by increasing pensioner benefits does not affect the median income or the poverty line. There are some policies aimed at the poor – such as increasing child benefit or raising the tax threshold – that might also increase the median income and thus the poverty line, but again, this is not a perverse implication.
This does not mean that the use of the median cannot have some statistical quirks. It has been argued, for example, that it can lead to seemingly perverse poverty estimates over time, especially during recessions. Thus, relative poverty in the UK fell during the recessions of the mid-1970s, the early 1980s and the early 1990s even though real incomes among the poor fell on average so that absolute poverty rose. This is because median incomes fell, and the poverty line followed suit. The actual pattern, however, varied between different groups. It was true of poor pensioners, for example, whose benefits were protected in real terms. In the case of households with children, the rate of relative poverty rose during these recessions largely because rising unemployment and stagnant wages cut real incomes for those on the lowest incomes by more than the median household.
Because the median income indicator is a measure of the gap between low incomes and the rest, relative poverty would only fall in a recession if the incomes of the poorest fall by less than the median so that this gap narrows. This might happen if most benefits are protected in real terms or the poor only suffer mild falls in real incomes. If, however, the poor suffer a sharper drop in income than the middle during a downturn, relative as well as absolute poverty rises. It could be argued that this pattern is less a perversity than a direct result of attempts to protect the poorest during economic turbulence. Maintaining at least the real value of benefits on which the poor depend provides a floor below which living standards will not fall.
It is sometimes also argued that a similar distortion applies during periods of rapid growth. For example, it has been argued that because median incomes rose sharply during the period of rapid growth from the late 1990s, tackling poverty has become more difficult because the poverty line has been rising rapidly along with the median. But this is the point of a relative poverty measure. It reveals that if the poorest are falling behind the rest during a period of rapid prosperity growth, they may be absolutely better off but they are still relatively poorer.
Despite these criticisms, the median-related threshold has significant merits. As the income of the household that sits in the middle of the distribution, it is less susceptible than the mean to changes in incomes at the top or bottom. As the interactive graph in the earlier link shows, if those at the top enjoy increases in incomes while everybody else stays the same, the mean income would rise but the median would stay the same. That is equally true in the other direction. The median is unaffected by changes in the incomes of the richest and the poorest and there is no mathematical reason why relative poverty – defined in relation to the median – cannot be abolished. It is thus a much more robust measure than the mean.
The median is also a strong indicator of what is considered normal in contemporary society. Linking the poverty standard to a fixed proportion of the median means that it moves up and down as the median rises or falls. It thus provides a simple measure of how well society is faring in tackling relative poverty, of how the poorest members of society are doing in relation to others, and enables comparisons over time and between countries.
One of the key benefits of the median-related poverty target is that it sends a signal that the poorest should be allowed to share in the proceeds of growth. Relaxing it would mean condemning the poorest to the slow lane of economic progress. This is the central reason for linking the poverty standard to a measure of average incomes – it ensures that society sets a minimum living standard that is linked to overall improvements in prosperity.
Ironically, the median-related threshold was first introduced at the end of the 1980s during Mrs Thatcher's third term in office with the introduction of the Households Below Average Incomes (HBAI) series. It replaced the previous Low Income Statistics series, which adopted a measure based on supplementary rates plus a percentage. By introducing this series, the then government set up a measurement process that revealed the full effect of wider economic, social and policy trends.
Since then, 60 per cent of the median is a measure that has become increasingly widely used as a primary threshold of income poverty. It has, for example, been adopted across most countries in the European Union and is one of the most effective ways of holding governments to account on progress towards reducing poverty. Despite the criticisms, the strengths of the median threshold mean that it should be kept alongside the other measures contained in the 2010 Act.
The second issue is that although raising life chances through improved schooling and parenting and better access to child care are an important part of an effective anti-poverty strategy, and will ultimately reduce the numbers in income poverty, they are also highly resource intensive. Just as the government is cutting benefit levels as part of its attempts to cut the fiscal deficit, they are also cutting spending levels in areas that will impact on such policies. Despite the Coalition's commitment to retaining Sure Start, for example, it is now likely that sharp cuts in public spending will mean the closure of hundreds of Sure Start centres that are aimed at improving life chances.
Third, prioritising life chances over income reflects a very narrow view about the fundamental causes of poverty. Underlying the findings of the Field Review and the references by David Cameron to Britain's 'broken society' is the belief that poverty is caused mainly by a combination of family breakdown and economic or welfare dependency rather than wider structural economic and social causes. Inherent in the shift of emphasis to improving life chances is a traditional Conservative belief, one that may be about to take us back to the nineteenth century view that the blame for poverty lies in a lack of personal and social responsibility, that the poor themselves are largely to blame for their own situation. As Duncan Smith has put it, 'the nature of the life you lead and the choices that you make have a significant bearing on whether you live in poverty'.
It is this belief in a largely 'individualist' approach to explaining poverty and their emphasis on promoting marriage and the family that has led the Conservatives to criticise Labour, as Cameron has put it, for relying 'too heavily on redistributing money, and on the large, clunking mechanisms of the state'. Yet, without the extra income support provided by Labour, especially in their first two terms, it has been estimated that the level of poverty would have been six percentage points higher and the level of child poverty 13 percentage points higher. 
Moreover, the evidence suggests that, though they play a role, family breakdown and welfare dependency are not the primary explanations of poverty. The central factor driving high levels of child poverty since the early 1980s has been the rise of what might be termed a 'livelihood crisis' in which a significant and rising proportion of the population has been faced with heightened barriers to decent work or pay. As a study by the Joseph Rowntree Foundation into recurrent poverty concluded, 'People's personal characteristics have some impact on the risks of recurrent poverty but structural labour market factors remain the strongest influence, implying that this is where the focus of efforts should lie'.
It is the growing lack of job opportunities, a trend that preceded the recession of 2008/9, that has been squeezing family incomes and driving the decline in life chances and social mobility in recent times. In the two immediate post-Second World War decades, the problem of the mass unemployment of the 1930s had been largely cracked – the UK unemployment rate averaged 1.6 per cent. Since 1979, it has averaged 7.8 per cent, nearly five times as high. According to the Office for Budget Responsibility, the level of unemployment is expected to rise to 8.2 per cent in 2011 compared with 7.9 per cent in 2010. The number of under-25s unemployed is now close to one million, nearly 40 per cent of the total, while the level of long-term unemployment has been rising sharply
While unemployment has been spreading, relative pay at the bottom has been falling. The proportion of employees with hourly wages below two-thirds of the median nearly doubled from 12 per cent in 1977 to 22 per cent in 2009. In 1977, 5.3 million people earned less than £7.28 per hour. The spread of low pay means that work is less of an escape from poverty. Many of those working in low paid jobs have poor qualifications and skills and move in and out of work in a 'low-pay, no-pay' cycle. It is because of the rise of an increasingly low pay economy that the proportion of poor children living in working households has been rising sharply to stand at 61 per cent in 2009. The Joseph Rowntree Foundation found an all-time high of 2.1 million children in poverty in 2010 where their parents are already in jobs. It is the long term weakening of the labour market and the decline in economic opportunities that is one of the primary reasons why governments have had to work harder just to prevent poverty rising.
While the introduction of the National Minimum Wage in 1999 has built a floor into this sinking process, it has mitigated but not halted this broader trend. Although the level of unemployment is likely to fall as recovery gathers pace, the evidence from earlier recessions is that the level of concentration of unemployment, as measured by the extent of long-term unemployment, is likely to stay high for several years. However, the extent of low pay is likely to stay the same or even rise. For these reasons, reliance on improving life chances is likely to have minimal impact on poverty levels in the short run.
Finally, far from being independent of each other, the evidence is overwhelming that money matters and that low levels of income are strongly related to poor life chances. Despite this, the government's consultation document, published in response to the Field Report, argued that 'the evidence available suggests that simply increasing household income, through reducing income poverty, will not make a big difference to children's life chances'. While poverty cannot be seen simply in terms of financial measures, inadequate income affects every aspect of family and children's lives. Low incomes are heavily correlated with a lack of success at school, poor health, a lack of job opportunities and low family well-being.
A number of responses to the Field Report have argued that it underestimates the importance of poverty on life chances.A major study led by Jonathan Bradshaw of York University has found that a wide range of social outcomes – from mortality and morbidity to mental illness and suicide – are associated (and mainly causally related) to child poverty. Moreover, while improved education and family support can mitigate some of the impact of poverty, it is in itself insufficient. Kathleen Kiernan, Professor of Social Policy and Demography at the University of York, and colleagues, have used data from the UK Millennium Cohort study to assess the extent to which positive parenting mediated the effects of poverty and disadvantage. Kiernan concludes:
Despite the best efforts of their parents, children living in poverty and relatively disadvantaged circumstances still remain behind their wealthier, well-parented peers. … Children's achievement can be adversely affected by poor parenting; it can also be adversely affected by poverty. Directing efforts at only poverty or parenting, to the exclusion of the other, is unlikely to result in equitable outcomes.
Poverty, however measured, is now likely to rise over the next few years. In an independent forecast of the impact of wider economic changes and the government's public spending and fiscal policies on two of the 2020 targets, the level of relative and of absolute poverty, the Institute for Fiscal Studies concludes that:
Among all children and working-age individuals, we forecast a rise in relative poverty of about 800,000 and a rise in absolute poverty of about 900,000 between 2010–11 and 2013–14 … Meeting the legally-binding child poverty targets in 2020 would require the biggest fall in relative child poverty after 2013–14 since at least 1961.
When it is published, the government's poverty strategy will thus coincide with an economic situation – a mix of persistent unemployment, falling real wages, rises in some taxes, the freezing and, in some cases, cuts in, a number of benefits, and wider cuts in public spending – that is driving higher levels of poverty. Any attempt to move the goalposts by downgrading the importance of relative in favour of absolute poverty would not enable the government to claim a fall in poverty during their term in office, because the net incomes of the poorest fifth of the population are set to fall over the next few years.
The government may well downgrade the twin-based approach by giving greater priority to measures aimed at improving life chances, but in a situation of falling public spending, such measures would have a very limited impact. The likelihood is that with the livelihood crisis likely to deepen with a weak labour market, falling or static incomes and deteriorating public services, upward social mobility may be set to continue to weaken, whatever the government's declared aims, while the goals laid out in the 2010 Act look ever more distant at least for the duration of this Parliament.
Stewart Lansley is the author of Poor Britain (with Joanna Mack). His book, The Cost of Inequality, will be published by Gibson Square in June 2011. He is grateful to Jonathan Bradshaw, David Drew, David Gordon and Joanna Mack for helpful comments.