Select Committee on Work and Pensions Written Evidence

Memorandum submitted by the Joseph Rowntree Foundation

  The JRF is one of the largest social policy research and development charities in the UK, spending about £10 million a year on a research and development programme that seeks to better understand the causes of social difficulties and explore ways of overcoming them.


    —  Benefits and tax credits now need to be looked at in combination

    —  The central purposes of the benefit system are being re-examined, and while this could bring a certain amount of simplification, there is also a risk of greatly increased complexity.

    —  A barrier to simplification is that we have no starting point of agreeing what is an adequate minimum income.

    —  Work incentives have improved overall over the long term, but remain a key issue. Trade-offs between employment incentives, in-work incentives and poverty alleviation need to be delicately managed.

    —  Two particular issues to address in terms of entitlements are the existence of discontinuities and the complex differences in uprating policies.

    —  The coexistence of tax credits and benefits create difficulties of timing and complexity.

    —  The benefits system continues to make a sharp distinction between people "in work" and "out of work", and thereby potentially deters some people from engaging in certain forms of economic activity, or pushes them into a hidden, informal economy.

    —  It is worth exploring approaches that could reduce the importance of means testing, including the strengthening of some universal benefits and measures affecting the need for council tax benefit and housing benefit.


  1.  An inquiry into simplification of the UK benefits system is a welcome but ambitious enterprise. A repeated theme in social research carried out by JRF and others is that claimants find the system complex and confusing, and that this reduces the value of the benefits system as a whole and in some cases deters people from using it. Yet over the years, efforts to simplify the system have proven elusive, partly because they have conflicted with other objectives, and partly because each reform has been overlain on an existing set of entitlements and structures. A bold reform to "start again" and create a new and greatly simplified system would undoubtedly have huge transition costs and create losers—which is not necessarily an arguments against contemplating it. However, without such a "big bang" reform, efforts to simplify certain parts of the system need to be carefully managed, taking account of potential impacts elsewhere.

  2.  When speaking of the "benefits" system, we usually refer to payments to help support the income of people with one or more of the following characteristics: incomes that are low relative to their needs (especially means-tested benefits); reduced earnings due to a situation such as unemployment, incapacity or retirement (especially insurance based benefits); and increased needs due to a situation such as disability or having children, meriting a transfer from people without these needs even for the non-poor (universal benefits). A significant complication since 1998 and especially since 2003 is that personal tax credits share key characteristics with payments labelled as "benefits". In particular, the 2003 reform made tax credits part of the system of providing income for people outside work, for families with children. It would be hard to find a logic other than pure labelling that distinguished these tax credits from benefits in terms of overall purposes, and thus it is impossible to talk sensibly about simplifying benefits without considering tax credits as part of the equation.

  3.  This submission makes selective points about the benefits and tax credit system that has emerged from JRF's recent research and analysis. It does not try to give a comprehensive analysis of the system, nor make a general proposal for reform.


  4.  In recent years, the understanding of the purposes of the benefit system has evolved, moving away from the idea of "social insurance"—a system of protection that people pay into when times are good, and draw out of when things go wrong. Increasingly, this has been replaced by the notion of a means-tested "safety net" designed to avoid destitution on condition that people behave in certain ways. The latest formulation of this ethos is in the Freud Report's orientation towards "reducing dependency". This puts a high emphasis on creating behavioural incentives, with less attention paid to what level of benefits will provide a decent standard of living. Freud appears to be suggesting a simplified set of benefit levels, with a personal allowance sufficient to cover basic living costs for client groups, yet also raises concerns about whether this might erode work incentives.

  5.  Whatever one thinks about this balance, it is important to note that the current orientation of the benefits system does militate towards complexity. The desire to target resources only to those who most need them creates complex forms of means testing. The desire to keep the basic level of benefits down and only provide extra resources for those with extra needs also helps create a more complex system than if basic entitlements were more generous. This is seen for example in the number of people who have to claim housing benefit, which must related to means and needs, in contrast to other countries where some or all of housing costs can be afforded from general benefits. The desire to influence behaviours also creates new forms of complexity, in terms of the conditions that have to be met in order to qualify for benefits. The difficulty with the suggestion by Freud that benefits be simplified into one or two rates is that if these rates are kept to a minimum, extra needs for individuals could go unmet.


  6.  Any reform aiming to ensure that the benefits system offers an adequate basic income on a simple basis to those who need it faces a fundamental question that has not been systematically asked by any contemporary UK Government: how much is adequate? The lack of a method for defining a socially acceptable minimum has been particularly evident in recent years, when poverty measurement has been expressed mainly in terms of minimum relative incomes, while benefit upratings have been largely based on price inflation and thus have lost ground relative to contemporary norms.

  7.  The Joseph Rowntree Foundation's current project on minimum income standards will for the first time provide a comprehensive measure of a minimum income standard for Britain. Led by judgements made by members of the general public of a list of items that everyone should be able to afford, and informed by expert knowledge about matters such as nutritional needs for a healthy life, this will provide a baseline of minimum income needs for each family type. This will suggest a simple foundation informing a future benefit system, but it will not get rid of complexity. Saying what minimum income we would not want any family type to fall below is not the same as saying that this income will be sufficient for all families of that type, as extra needs caused for example by disabilities or living in a remote area may mean that some families would have to have supplements to a basic income in order to achieve an acceptable standard of living. More information about the Minimum Income Standards project, which will report early next year, can be found at


  8.  Turning to a consideration of the existing benefits and tax credits system, one can distinguish issues about the administration, labelling, delivery and take-up of benefits from issues about the overall structure of entitlements. A first question is whether entitlements are fair, serve their key objectives and avoid creating disincentives to work.

  9.  An important aim of the benefit system is to provide income to people when they are not working, without creating a perverse incentive to remain outside work. According to analysis carried out for the Foundation by the Institute for Fiscal Studies,[29] there has been an overall improvement of work incentives in the past quarter century. This is not surprising, given (a) that the basic level of many out of work benefits have not risen in line with earnings and (b) that families with low incomes inside work have benefited from supplements to their incomes (first as benefits and now as tax credits) which have grown greatly in value. However, the IFS also points out that improvements in work incentives have been uneven, and have declined since 2000.

  10.  In structuring the system, governments have had to balance the priorities of:

    (a)  Providing an adequate income baseline;

    (b)  providing incentives to work rather than not; and

    (c)  maintaining incentives and rewards to increase earned incomes within families, whether by having more people within the family working, increasing an individual worker's hours or investing and training or career development to improve pay.

  Some of the trade-offs involved are illustrated by Figure 1. In section A of the diagram, there is only a small difference between minimum incomes inside and outside work. Reducing the latter would deepen poverty; increasing the former, as section B, by giving tax credits to people in work on low incomes, reduces the "employment disincentive" but increases the "in-work disincentive", by raising withdrawal rates on increased earnings. This can be addressed, as in section C, by reducing the rate at which tax credits are withdrawn, but that would cause people further up the income scale to receive tax credits and extend the number having relatively rapid withdrawal of income.

  11.  A first impact of tax credits has therefore been to shift disincentives from the incentive to work to the incentive to improve earnings. However, the extension of the Child Tax Credit (CTC) to include non-working families in 2003 has potentially affected the incentive to work for families with children. In particular, the more generous uprating regime applying to the CTC than to the Working Tax Credit (WTC) could reduce over time the size of the reward for working relative to baseline incomes. The IFS work shows that in fact, both the incentive to work and in-work incentives, while higher today than in 1979, have reduced somewhat since 2000.

  12.  Another piece of modelling analysis, supported by JRF, looking at the lifetime effects of the tax and benefits system, showed that the poverty trap is more pervasive than might be concluded from a static picture.[30] This work concluded that only better jobs with better pay and earnings could allow people to escape lifetime poverty, and that while tax credits can give a temporary boost, many people claiming them have unstable working lives that cause them to dip in and out of poverty.

  13.  The recent experience of setting entitlements thus shows that it is hard to achieve all objectives simultaneously. However, it also raises some particular issues that could be addressed more directly in the development of benefit policy—notably the nature of "discontinuities" in entitlement and the system of uprating.


  14.  One kind of discontinuity in our present system occurs at the transition from out-of- work to in-work support. People working just a few hours a week lose benefit pound for pound (above a small disregarded portion of income), but once they are working 16 hours (for lone parents) or 30 hours (for others), they become eligible for the Working Tax Credit. In the case of lone parents, the tax credit can potentially be worth about as much as Income Support, meaning that net income for working 16 hours can be greatly higher than for working 15 hours. At present, a JRF project led by One Parent Families is modelling options for making the benefit system work better for people who take on "mini-jobs".[31]

  15.  Another discontinuity, arguably, is at the other end of the tax credit system, at the point at which entitlement to tax credits runs out. Here, a recipient can go from a net income withdrawal rate of nearly 70% to a withdrawal rate of 33% by earning a pound more. A radically different structure proposed by supporters of a "citizens income" would give people a basic minimum income and have a constant rate of taxation thereafter. This would mean a higher withdrawal rate for most taxpayers, but an end to all situations where people lose most of their income at the margin (excluding the effect of Housing Benefit and Council Tax Benefit withdrawal). If the two main objectives of the tax and benefit system were to avoid very high disincentives and to maximise simplicity, such a system would have its attractions. Its big difficulty, however, would be a much higher "headline" basic rate of tax. In 2006, JRF together with IFS calculated the combined income tax and national insurance contribution rate required, in a "flat tax" system incorporating tax credits, as 46% (compared to the current basic rate tax plus NI rate of 33%).[32] While such a reform seems politically inconceivable, it was interesting to note that no middle-income families or families with children would lose substantially from such a system: the only big losers would be higher-income families without children.


  16.  One of the greatest, yet least noticed, anomalies in the present benefits system is the way in which rates are increased from one year to the next. Regimes for different benefits and tax credits vary from increasing in line with earnings to not increasing at all, with at least five variants, as shown in the following table:

Current uprating "default" basis
At least in line with average earnings
Inflation or £2 a week, whichever is higher
RPI inflation
RPI excluding housing costs
No increase

Pension Credit guarantee element and Child Tax Credit child element
Retirement Pension
All contributory benefits and some non-contributory ones including Child Benefit.
Income Support, Jobseekers Allowance, Housing Benefit, Council Tax Benefit.
Child Tax Credit family element, income disregard for Working Tax Credit

  These differences go relatively unnoticed because they do not, for the most part, involve recipients becoming worse off. The differences are only important over very long periods of time, but here they have been very important indeed. Where benefits are uprated in line with prices rather than earnings, this may mean only foregoing a 2% rise in a single year, but potentially a 50% rise in a generation. Indeed, over the past 25 years, relative poverty has deepened in large part because basic benefits (eg the single rate of Income Support/Supplementary Benefit) were as much as 40% lower, relative to earnings in 2006 than in 1981.

  17.  A simplified benefits system may potentially reduce these anomalies. A current JRF project based at the University of Essex is looking into the implications of different systems of benefit uprating, modelling various options and making international comparisons.[33] To a large extent, differential uprating policies may be seen as a method of gradually restructuring entitlements over time without creating too much disruption to the system, but if so there is a case for making these decisions and the rationale for taking them more explicit.


  18.  Claiming the benefits to which one is entitled has long been a complex business within the UK benefits system. Tax credits were intended to bring improvements but appear to have done the reverse. For many people they have brought one extra piece of complexity without bringing them out of other systems—such as housing benefit and income support—which have complexities of their own.

  19.  Other organisations making submissions are likely to bring details of how benefit administration works out in practice. JRF looked particularly at the issue of the timing of assessment and payment at the time of the introduction of the 2003 tax credit reforms, drawing on international experience.[34] This review suggested that annual assessment is a difficult thing to achieve without an extremely "light touch" approach to earned income, which is characteristic of Canada (where previous years' earnings are used without subsequent adjustment) but not Australia (which is more like the UK system. The experience of the system seems to confirm that we need either a lighter touch tax credit regime or a better integration with the benefits system.


  20.  The present UK tax and benefit system makes a sharp distinction between people who are "in work" and those who are "out of work". This is based to some extent on whether people are working more than a certain number of hours, and as mentioned above if this is less than 16 they are treated as being part of the out of work benefits system. One consequence of this is that they tend to have a disincentive to do any work at all, and to be treated literally as "inactive" and to some extent "incapacitated".

  21.  JRF studies on social participation and on the informal economy suggest that such a sharp dichotomy between "working" and "inactive" people is not helpful. A study on "co-production" looked at the process whereby clients work alongside professionals as partners in the delivery of services. One of its conclusions was that the benefits system needs to be able to provide incentives for those outside paid work to get more involved in their neighbourhoods without endangering their basic income.[35]

  22.  Another report[36] looked at people's involvement in the informal economy. Often such involvement came about because of difficulties in moving easily in and out of the formal economy, which those interviewed would have preferred to be involved in. Participants believed that the tax and benefit system created disincentives to returning to formal work. In particular, fears about inefficiencies in the system made participants wary of starting formal work. Participants had often experienced delays in receiving benefits and failure to reinstate Housing Benefit on change of circumstances. As a result, some were not willing to take what they perceived to be a risky leap into the uncertain world of formal employment. Many also felt they would be no better off when working formally rather than claiming benefits, despite the introduction of measures such of tax credits, as they would lose a range of other benefits such as free prescriptions and free school meals if they started formal work. This suggests that work incentives remain an important issue.


  23.  There is wide consensus that a growth in means-testing has made it harder for the benefits system to do its job well. This is partly because of stigma, partly because of complexity introduced by the testing of means and partly because of potential disincentives to earn or to save. On the other hand, the desire to target resources at measures that will reduce poverty often leads to extra means testing.

  24.  While a trade-off between targeting and complexity will always exist, there are some ways in which it may be possible to reduce means-testing while still maintaining a certain amount of focus. One is by putting resources into universal benefits whose claimants are more likely than average to be poor. An example of this is child benefit for larger families. JRF work has shown that such families are so much more at risk of poverty than the average family that raising child benefit for additional children can give comparable value for money in reducing child poverty to income-targeted measures.[37] Similar arguments apply to certain disability related benefits.

  25.  It may also be possible in some cases to reduce the need to claim some means-tested benefits. Council tax is a case in point. A recent JRF study on the impact of the tax on low income families[38] showed that users were exasperated by the sheer complexity of council tax benefit and of its interaction with other benefits: "They couldn't make it more complicated if they tried", said one user. Creating a more automatic reduction of council tax to people on low incomes as suggested by the Lyons Review, and/or ensuring that the structure of the tax makes it less likely that low income families face high bills, might reduce this burden of means-testing within the benefits system. Similar arguments can be applied to housing benefit. In particular, the creation of more affordable housing, targeted at families in need but with access not narrowly means-tested, could reduce the burden placed on a means-tested housing support system. It is worth remembering in this context that the large growth in housing benefit dependency two decades ago came about in large part because of the reduction of revenue subsidies in the public sector and of rent regulation in the private sector.

5 April 2007

29   Mike Brewer and Andrew Shepherd (2006), The poverty trade-off: work incentives and income redistribution The Policy Press. Back

30   Martin Evans and Jill Eyre, 2004, The opportunities of a lifetime: Model lifetime analysis of current British social policy Bristol: Policy Press. Back

31   For details see Back

32   Donald Hirsch, 2006, Flatter taxes: rich giveaway of new deal for poor? Online report at Back

33   For details see Back

34   Peter Whiteford, Michael Mendelson and Jane Millar (2003) Timing it right? Tax credits and how to respond to income changes York: Joseph Rowntree Foundation (out of print, but available online as free pdf at Back

35   David Boyle, Sherry Clark and Sarah Burns (2006) Hidden work: co-production by people outside paid employment, York: Joseph Rowntree Foundation. Back

36   Dennis Katungi, Emma Neale and Aaron Barbour (2006), People in low-paid informal-work: `Need not greed', Bristol: The Policy Press. Back

37   See Donald Hirsch (2006) What will it take to end child poverty?, page 57. Back

38   Michael Orton (2007) Struggling to pay council tax: A new perspective on the debate about local taxation by, York: Joseph Rowntree Foundation. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2007
Prepared 26 July 2007