THE independent Office for Budget Responsibility (OBR) was established in 2010 to monitor the public sector’s finances. Twice a year – alongside each Budget and Spring Statement – we produce detailed forecasts for the coming five years, assessing the likely impact of policy decisions and expected developments in the economy. We then use these forecasts to assess the Government’s performance against the fiscal targets that it has set itself for the management of the public finances.

In scrutinising the outlook for public spending, an important component is spending on social security benefits and tax credits – cash payments that governments make to individuals or families with lower incomes and/or specific needs. In 2016-17, the UK Government spent £217 billion on these payments, equivalent to 28 per cent of total public spending and 11 per cent of national income. The Government has set a ‘welfare cap’ on some of this spending. The OBR has been asked to assess compliance with this cap.

Welfare pics


There are two parts to this guide.

  • First, we present an overview of total UK spending on social security benefits and tax credits, which looks at how much is spent, what it is spent on, and which factors have caused this to change in the past and are expected to cause it to change in the future.
  • Second, we look at each line of social security and tax credits spending in turn, addressing similar questions of how much is spent, how that has evolved over time and what we expect to happen over the next five years. We end by looking at universal credit, which is expected to change the welfare spending landscape over the coming years as it replaces six existing benefits and tax credits with a single payment.

Parliament has asked us to focus on how much the Government spends, not on how well it spends it. So we do not discuss how benefits and tax credits affect the distribution of income or measures of living standards and poverty – important though those issues are. The figures presented here are consistent with our March 2018 forecast, covering the five fiscal years up to 2022-23. Each fiscal year runs from April to March.

What counts as welfare spending?
What is welfare spending spent on?
What drives welfare spending?
Recent trends in welfare spending
The outlook for welfare spending
The welfare cap

A pdf version of this guide is also available to download here:

An OBR guide to welfare spending

January 26, 2017 – 11 MB

Back to top

What counts as welfare spending?

THE UK public sector is estimated to have spent a total of £771 billion in 2016-17 (equivalent to 40 per cent of GDP). Within this total it spent around £484 billion (about 25 per cent of GDP) on the ‘welfare state’, broadly defined, including health, education, social services and housing, as well as social security and tax credits. Most of the remainder went on other public services (e.g. defence and transport) and interest payments on government debt.


Social security and tax credits spending amounted to £217 billion (11 per cent of GDP or around £8,000 per household). The Government’s welfare cap excludes spending on the state pension and those benefits linked most closely to the ups and downs of the economy. Spending subject to the cap totalled £119 billion in 2016-17 (6 per cent of GDP).

Back to top

What is welfare spending spent on?

AT any one time over half of all families receive income from at least one benefit in the welfare system – and most people will receive one or more welfare payments for well over a third of their lives (including child benefit when young and the state pension when retired).

Of the £217 billion spent on welfare payments in 2016-17, around 59 per cent was paid to pensioners, with state pensions the largest single item at £92 billion. Personal tax credits – mostly for families with children – cost £27 billion and housing benefit – three quarters of which is paid to people of working age – cost £23 billion. Together these payments made up around two-thirds of all welfare spending.

Disability and incapacity-related benefits accounted for a further 8 and 7 per cent of welfare spending respectively, followed by child benefit (payable for most children from 0 to 18 years of age) at 5 per cent. Pension credit made up a further 3 per cent of welfare spending.

Despite its relatively high profile, spending on jobseeker’s allowance – paid to those who are unemployed and meet certain criteria – accounted for only £2 billion or a little under 1 per cent of total welfare spending. Equivalent spending on universal credit in parts of the country where jobseeker’s allowance is no longer available amounted to £740 million.

The remaining 11 per cent of welfare spending included spending on a range of benefits in Northern Ireland, which are administered separately, and a number of smaller benefits including carer’s allowance, statutory maternity pay, income support and winter fuel payments.

Back to top

What drives welfare spending?

WELFARE spending is driven by factors that affect the number of people receiving each welfare payment (the ‘caseload’) and the average amount paid to each recipient (the ‘average award’). In addition to the basic decisions that governments take about eligibility criteria and the generosity of each type of payment, factors influencing welfare costs include:

    • Demographic and economic trends. Overall welfare spending per person is higher at both younger and (particularly) older ages. State pension spending has been pushed higher by the ageing of the population (which has raised the proportion of adults over the state pension age), and an increase in the proportion of women working (which has increased the number of people eligible for the full pension). The cost of unemployment benefits rises and falls with the ups and downs of the economy. A rising share of the population renting rather than owning their home has increased the housing benefit caseload, while the shift from the social- to the private-rented sector has raised the cost per claimant due to higher rents paid. Most importantly for the system as a whole, changes in inflation typically drive the uprating of most welfare payments. If inflation is higher than earnings growth, the welfare bill will tend to rise relative to national income. (This can be offset by government policy decisions, as has been the case in recent years for most working-age welfare payments).


  • Major reforms to the welfare system. These often lead to unexpected changes in spending. For example, the cost of tax credits rose faster than expected in the mid-2000s, as earnings grew more slowly for tax credits recipients than in the wider economy, and as childcare costs increased significantly. Reforms to unemployment benefits in the second half of the 1980s, designed to reduce their cost after the recession of the early 1980s, pushed the incapacity benefits caseload up sharply. This prompted major reform of incapacity benefits in 1995.
  • Changes in take-up rates – the proportion of people eligible for a benefit who claim it. The introduction of pension credit in 2003 was accompanied by a campaign to raise take-up; the caseload increased by more than 50 per cent between 2002-03 and 2005-06. Similarly, benefit take-up among low-income families with children increased from around 60 per cent for the family income supplement to 90 per cent for the current system of tax credits.
  • Wider public policy decisions. For example, lower spending on social housing may have put upward pressure on the housing benefit bill, by increasing the proportion of recipients paying higher rents in the private-rented sector.

Back to top

Recent trends in welfare spending

WELFARE spending has increased fourfold in cash terms over the past 30 years and has more than doubled in real terms, after adjusting for inflation. But, as a share of national income, there has been no clear trend. Our estimate of welfare spending fluctuates with the ups and downs of the economy, averaging close to 11 per cent of GDP between 2016-17 and 2022-23. It climbed to over 12 per cent of GDP after the late-2000s recession – a smaller increase than in the early 1990s – but has been falling since 2012-13 and is forecast to continue falling over the next five years.

Spending on all age groups fell as a share of national income through the boom of the late 1980s, but then increased in the recessions of the early 1990s and late 2000s. Spending on working-age recipients appears to have been the most cyclical, reflecting the link with unemployment. Spending on pensioners went through a long period of relative stability prior to the last recession before rising relatively sharply during it, while spending on children saw a marked rise in the 2000s. Spending on each group is forecast to fall as a share of national income over the next five years, although for pensioners it is set to rise again from 2021-22 onwards.


Looking just at the proportion of welfare spending going to each group, the mix tilted first towards pensioners from the mid-1980s (up 5 percentage points), before tilting back towards working-age people through the early 1990s recession (up 8 percentage points). Spending then shifted towards children from 1997-98 to 2010-11 (up 6 percentage points), and finally back towards pensioners and away from working-age people since 2010-11 (by 3.5 percentage points by 2016-17, with a further 2 percentage point shift towards pensioners forecast in the next six years). The 57 per cent of welfare spending expected to go to pensioners in 2022-23 would be the highest proportion in at least 30 years.


In terms of the breakdown of welfare spending as a share of national income on specific benefits:

  • During the 1980s and early 1990s, spending fluctuated with the economic cycle, with spending on unemployment and incapacity benefits falling then rising. Spending on state pensions also fluctuated as the uprating of awards varied relative to the strength of average earnings growth;
  • From 1997-98 to the late 2000s recession, spending on tax credits increased significantly as they absorbed other benefits (such as the child allowances previously paid with income support and jobseeker’s allowance) and were used to reduce child poverty. There was also a flow of people away from unemployment benefits and onto disability benefits, reducing the cost of the former and increasing the cost of the latter. The share of GDP spent on state pensions and child benefit fell, as they were generally uprated by inflation at a time when prices were rising less quickly than average earnings and national income; and
  • The welfare bill jumped between 2007-08 and 2012-13 during the last recession and the start of the recovery. Uprating the state pension and tax credits in line with – or by more than – inflation protected their purchasing power while increasing their generosity relative to both earnings and the size of the economy. Higher unemployment and weak earnings growth increased the number of people claiming jobseeker’s allowance and housing benefit.


We expect spending on most welfare payments to fall as a share of national income on current policy, with tax credits falling proportionally the most and disability benefits increasing (as described in the next section).

Back to top

The outlook for welfare spending

OVER the next five years spending on welfare is forecast to rise 13 per cent in nominal terms and 4 per cent in real terms but rise by only 1 per cent in terms of real spending per person. Our preferred measure – spending as a share of GDP – is expected to fall by 0.3 percentage points. This is a smaller decline than that of the preceding 5 years, which saw spending as a share of GDP fall by 1.2 percentage points as Government policy lowered the generosity of working-age benefits.

This 0.3 per cent of GDP fall is driven by reduced spending on tax credits, housing benefit and universal credit, with smaller falls across other items. The fall in spending on the state pension partly reflects the rising state pension age – which offsets the effect of uprating the state pension by the comparatively generous ‘triple lock’ (the higher of wage growth, inflation and 2.5 per cent).

The majority of these falls in spending are the result of policy changes announced in four statements to Parliament. Three took place in the 2010-2015 Parliament:

  • the post-election June Budget 2010, when the Coalition Government announced measures that were expected to reduce welfare spending by around £9 billion in 2014-15, the bulk of which came from switching the uprating of most working-age benefits and tax credits from RPI or Rossi to the lower CPI measure of inflation; freezing the uprating of child benefit for three years; and reforms to tax credits, housing benefit and disability benefits;
  • Spending Review 2010, in which measures were expected to reduce welfare spending by around £7 billion in 2014-15, mainly from withdrawing child benefit from higher-income families, limiting contributory employment support allowance claims to one year; and further tax credits reforms; and
  • Autumn Statement 2012, in which measures were expected to reduce welfare spending by around £4.5 billion in 2017-18, driven largely by a three-year cap of 1 per cent on the uprating of most working-age benefits.

In the Summer Budget 2015 at the beginning of this Parliament, the new Conservative Government announced a range of policy measures that were expected to reduce welfare spending (mostly for working-age recipients) by around £12 billion in 2019-20 – the year in which it was aiming to secure a budget surplus. Two measures to cut spending on tax credits were reversed in the November 2015 Autumn Statement, reducing savings by around £3 billion in the short term, but only around £0.5 billion by 2020-21 when most tax credits recipients will have moved to universal credit (UC), which is now less generous following cuts announced in July 2015. While the ‘pay to stay’ housing policy was also reversed in the November 2016 Autumn Statement the overwhelming majority of the £12 billion worth of cuts remain in place.

The main sources of the remaining cuts include:

  • the four-year freeze in the uprating of most working-age benefits from 2016-17 to 2019-20, estimated at the time to save £3.9 billion by 2019-20;
  • the cut in UC work allowances, estimated in July 2015 to save £2.9 billion in 2019-20;
  • limiting the child element to 2 children and removing the family element/premium for flows into tax credits, UC and housing benefit, estimated in July to save £1.6 billion in 2019-20; and
  • policies that result in cuts to housing benefit announced in July 2015 – reducing social sector rents by 1 per cent a year for four years, estimated in July 2015 to save £1.9 billion in 2019-20.

We also expect spending on disability benefits and incapacity benefits to fall as a share of GDP as ongoing reforms to both systems are completed.

These cuts are being implemented against a baseline that reflects the cuts in welfare spending announced in the 2010 to 2015 Parliamentary session, which were estimated to amount cumulatively to £21 billion in 2015-16. But trends in the economy (notably the weakness of earnings growth and periods of high price inflation), delays in implementing reforms and other factors, meant that welfare spending in 2015-16 was only around £5 billion lower than had been forecast in June 2010. Spending on items now subject to the welfare cap was actually higher – not lower – in 2015-16 than forecast in June 2010.

Our latest long term projections show spending on state pensions and other pensioner benefits rising – the former from 5.1 per cent of GDP in 2022-23 to 7.1 per cent in 2066-67. This reflects an ageing population and increased generosity as a result of the ‘triple lock’ on uprating. Spending on other welfare benefits is projected to be relatively flat as a share of GDP over the long term.

Back to top

The welfare cap

IN Budget 2014 the Government set a cap on the total amount forecast to be spent on selected benefits and tax credits from 2015-16. (It is separate to the ‘benefit cap’ that caps the overall level of welfare payments payable to any individual family). The Government has excluded spending on the state pension from the welfare cap – which it claims is “better planned and controlled over a longer time period” – and jobseeker’s allowance and associated housing benefit payments – as “the most cyclical elements of welfare”. This leaves just over half of social security and tax credits spending subject to the cap.


The cap was reset in the July 2015 Budget and we made a formal assessment of performance against it in the November 2016 Autumn Statement, which showed it being breached in all years. The Government then introduced a new welfare cap, which was set to only apply in 2021-22. This cap was re-set in November 2017 following the beginning of a new Parliament, with the latest cap applying to spending in 2022-23. We will monitor progress against it until then.

Spending subject to the cap is forecast to rise by 8.4 per cent from £119 billion in 2016-17 to £129 billion in 2022-23, while spending outside the welfare cap is forecast to rise by 18 per cent from £101 billion to £119 billion. The difference is explained in large part by the basic state pension rising in line with the ‘triple lock’, while many benefits subject to the cap have been frozen in cash terms until 2019-20.

Table 1 : Welfare cap spending

Back to top

State Pension

THE state pension is the largest single item of welfare spending making up over 40 per cent of the total. The system for pensioners who retired before April 2016 comprises the basic state pension (providing a basic level of retirement income) and the state second pension (related to earnings). Since April 2016, these have been replaced by a ‘single-tier’ (flat-rate) pension for newly retired pensioners. The state pension age (SPA) is set to rise to 66 for both men and women by 2020.

Spending on the state pension has risen in cash terms from £17 billion in 1985-86 to £92 billion in 2016-17. Relative to GDP, spending fell during the late 1980s and rose in the late 2000s recession and subsequent recovery. More recently spending has stabilised again.

Spending fell as a share of GDP because the state pension was generally uprated in line with inflation at a time when prices were rising less quickly than average earnings and national income. However, rising life expectancy, and post-war baby-boomers reaching state pension age (SPA), increased the proportion of adults over the SPA from 23 per cent in 1985 to 24 per cent in 2015 (an extra 2 million pensioners), while a higher proportion of women in work increased eligibility for the full basic state pension.

Since 2010, the basic state pension has been uprated by the ‘triple lock’ whereby it is uprated by the higher of CPI inflation, average earnings growth or 2.5 per cent. This increased the generosity of the state pension by 3 per cent relative to RPI inflation over five years and by 10 per cent relative to earnings.

Spending on the state pension is forecast to rise by less than 1 per cent of GDP between 2017-18 and 2022-23. Continued upward pressure from an ageing population is only partially offset by the SPA for men and women rising to 66 by 2020.

Back to top

Personal tax credits

PERSONAL tax credits comprise the working tax credit – payable to families with someone in work (typically for 16 hours or more a week) – and the much larger child tax credit – payable to families with children. The working tax credit also subsidises childcare costs. Awards are based on family circumstances and means-tested against family income.

Spending on personal tax credits accounted for 13 per cent of total welfare spending in 2016-17. It has risen sharply, from £1 billion (on its predecessors) to £27 billion in cash terms and from 0.3 to 1.4 per cent of national income between 1985-86 and 2016-17.

In 1999-00, the working families’ and disabled persons’ tax credits (WFTC/DPTC) were created – subsuming family credit and disability working allowance, while also increasing average awards. The current ‘new tax credits’ were created in 2003-04, absorbing both WFTC/DPTC and the child allowances in income support and jobseeker’s allowance. Support was also extended to families on higher incomes and families without children while the children’s tax credit was abolished. The increase in average awards and the widening of the scope significantly increased spending.

Spending subsequently increased during the late-2000s recession, as the child element was uprated faster than earnings and inflation, substantially increasing generosity relative to earnings and the size of the economy. At the same time earnings grew more slowly in the tax credits population than in the wider economy, further raising spending.

Measures announced in June 2010 cut support from higher up the income distribution meaning the caseload fell by around 20 per cent in 2012-13. Average awards were also cut as the uprating of various elements moved from RPI to CPI inflation, childcare support was cut from 80 to 70 per cent of eligible costs, the withdrawal rate was increased from 39 to 41 per cent, and additional payments for babies were also cut. These cuts were only partially offset by an increase in the value of the child element.

Spending on tax credits is forecast to fall further from 1.4 per cent of national income in 2016-17 to 1.1 per cent in 2022-23. This largely reflects the four-year freeze on the uprating of tax credits from 2016-17 to 2019-20, alongside measures that cut support for larger families.

In reality spending on tax credits will fall faster, because it is one of the elements of the welfare system that will be replaced by universal credit over time (as described at the end of this guide).

Back to top

Housing benefit

HOUSING benefit is financial support available to people on low incomes who rent their homes from private- or social-sector landlords. Unlike many benefits, there is no fixed amount available to each claimant. The value of the award depends on an estimate of ‘eligible’ rent and other household circumstances. Housing benefit is administered by local authorities.

In 2016-17 housing benefit cost around £23 billion, 11 per cent of total welfare spending and 1.2 per cent of GDP (up from 0.8 per cent in 1985-86). Housing benefit for those claiming unemployment benefits tends to rise and fall with the economic cycle, while caseloads for people with a disability or health condition have risen markedly since the late 1980s – driven by trends in incapacity and disability benefits. Caseloads for pensioners have fallen since the 1990s, partly due to higher home ownership rates among pensioners, as well as growth in pensioner income.

Since the early 2000s, the share of spending accounted for by claimants in the private-rented sector has risen. As private rents are on average higher than social rents, that shift has put upward pressure on welfare spending. These rents have risen faster than earnings and inflation over the past decade.

The number of claimants in employment has risen from 0.4 million at the end of 2008 to 1.0 million by November 2017 (from 10 to 22 per cent of the total). This reflects the combination of strong employment growth but also weak earnings growth, as well as the continued rise in the proportion of households renting their homes.

Spending on housing benefit is forecast to fall as a share of national income – from 1.2 per cent in 2016-17 to 1.0 per cent in 2022-23 – as average awards grow more slowly than GDP. That is mainly the result of policy measures.

Housing benefit will be replaced by universal credit over time.

Back to top

Disability benefits

DISABILITY benefits provide financial support to assist with the costs of daily care and/or mobility needs, based on the extent of the recipient’s disability. This has been provided through disability living allowance (DLA) since 1992. For working-age claimants, this is now being replaced by the personal independence payment (PIP), which has different eligibility criteria. Attendance allowance provides support to pensioners who require substantial and regular care.

In 2016-17 spending on disability benefits accounted for 8 per cent of total welfare spending. Spending has risen from around £1 billion in 1985-86 to £22 billion in 2016-17, and from 0.3 to 1.1 per cent of national income over the same period.

The introduction of DLA led to rapid growth in caseloads and spending as a result of wider eligibility, the introduction of claimant self-assessment, and a rise in take-up.

The introduction of PIP for working-age claimants in 2013 was forecast to reduce spending, as people’s eligibility is ‘more rigorously’ reassessed. That process of reassessment continues, with the evidence to date pointing to smaller-than-expected savings (echoing the experience of reforms to incapacity benefits, as described in the next section). In recent forecasts, we have revised up how much we expect will be spent on PIP in the coming years as more people are expected to receive payments and the average amount paid to each claimant will be higher than initially expected.

Back to top


Incapacity benefits

INCAPACITY benefits are income-replacement benefits available to people unable to work due to sickness and/or disability. Employment and support allowance (ESA) has replaced incapacity benefit as the main component. Other components include a dedicated element of income support and the severe disablement allowance.

In 2016-17 spending on incapacity benefits was around £15 billion – around 7 per cent of total welfare spending. Spending on incapacity benefits has risen in cash terms over the past 30 years, but it has been on a downward trend in real terms (adjusting for inflation) and relative to the size of the economy since the mid-1990s, almost halving as a share of GDP. incapacity

Spending on incapacity benefits rose sharply between the 1980s and the mid-1990s, as caseloads rose from 1.3 million to 2.8 million between 1984-85 and 1995-96. This reflected periods of high unemployment and industrial restructuring. The rise in spending prompted major reform of the system in 1995, tightening eligibility and reducing generosity. As a result, spending fell as a share of GDP.

ESA replaced incapacity benefit in 2008 and was originally expected to deliver significant savings via stricter ‘work capability assessments’. Delivery problems have meant that savings have fallen well short of initial expectations, prompting successive upward revisions to our spending forecasts.

The volume of assessments has been lower than expected with significant delays and backlogs. In addition, completed assessments have resulted in a higher proportion of people being found eligible for support and fewer declared ‘fit for work’ than originally expected.

The majority of ESA will be replaced by universal credit over time.

Back to top

Child benefit

CHILD benefit is a cash payment payable for each child. Historically, it provided universal support for parents or guardians bringing up children, but since 2013 it has been subject to a tax charge for earners over £50,000.

In 2016-17 spending on child benefit was around £12 billion or 5 per cent of total welfare spending. Spending on child benefit has risen in cash terms, but has fallen relative to national income over the past 30 years as it was generally uprated in line with inflation at a time when prices were rising less quickly than average earnings and national income. This decline was occasionally offset by one-off policy decisions to increase generosity.

Child benefit has been means-tested since 2013, via the ‘high income child benefit charge’. This removes eligibility for child benefit from families with at least one parent earning more than £60,000 and reduces awards on a sliding scale for those with one earning between £50,000 and £60,000. Alongside recent decisions to freeze the uprating of child benefit until 2018, uprating with CPI in later years means that spending on child benefit is forecast to fall as a share of national income through to 2022-23.

Back to top

Pension credit

PENSION credit was introduced in 2003, replacing the ‘minimum income guarantee’ and before that income support for the over 60s. It tops up the income of older people to a minimum level, through the ‘guarantee credit’, while a ‘savings credit’ provides extra support to those who have saved for their retirement so that the guaranteed income does not remove the incentive to do so.

Spending on pension credit was around £6 billion in 2016-17. Over the past 30 years, spending on pension credit and its predecessors has risen in cash terms, fluctuating between 0.3 and 0.5 per cent of GDP. The introduction of pension credit did see spending increase, with government campaigns to encourage take-up contributing to a rise in the caseload of more than 50 per cent between 2002-03 and 2005-06.

Between 2017-18 and 2022-23 spending on pension credit is expected to fall as a share of national income. The rise in the state pension age to 66 will reduce caseloads, while the new single-tier state pension has been set above the level of the standard minimum guarantee.

Back to top

Jobseeker’s allowance

JOBSEEKER’S allowance (JSA) provides financial support for the unemployed on the condition that they take steps to find a job. Claimants are required to sign a commitment to seek work and to take part in a work programme after claiming for a certain period. Recipients can also claim other benefits – notably housing benefit if they rent their accommodation.

Spending on unemployment benefits (around £2 billion in 2016-17) is small relative to total welfare spending – just 0.9 per cent of the total. This has risen and fallen with the economic cycle, as shown in the chart below. Spending on unemployment benefits has fallen from a peak of 1.4 per cent of GDP in the 1980s to just 0.1 per cent now.

Spending increased substantially following the early-1980s recession, reaching a real terms peak in 1986-87 when claimants topped three million. Since 1986 eligibility criteria have been tightened a number of times, reducing caseloads and spending. Further smaller peaks in spending followed the early-1990s recession and, much smaller again, the late-2000s recession. We forecast spending on jobseeker’s allowance to be relatively stable as a share of GDP over the next five years.

The majority of jobseeker’s allowance will be replaced by universal credit over time. By 2017-18, around 40 per cent of the caseload had already moved to universal credit.

Back to top

Universal credit

UNIVERSAL credit (UC) combines the following six welfare payments into a single monthly payment administered by DWP:

  • income-based jobseeker’s allowance;
  • income-based employment and support allowance;
  • income support;
  • working tax credit;
  • child tax credit; and
  • housing benefit for working age people.

Once fully rolled out, UC will pay more than £60 billion a year to around 7 million claimants. UC is intended to simplify the working-age benefits system. It was not originally designed to save money, but successive policy changes mean that it is now expected to cost less than the payments it is replacing, with savings reaching £1 billion by 2022-23 and £2½ billion excluding the cost of transitional protection payments for some cases that are moved from the old to new systems at DWP’s discretion.

These savings represent the net effect of various features of UC that reduce and increase spending relative to its predecessors. Gross savings arise from operational differences – from reductions in error and fraud, the introduction of the minimum income floor for the self-employed, and the abolition of the income disregards in tax credits – as well as policy designed to make UC less generous than its predecessors, such as the July 2015 decision decision to reduce work allowances. Offsetting these gross savings are the costs associated with higher take-up of payments – due to the simplification of the welfare system – and more generous support for some at low hours of work, others via a lower withdrawal rate, and also childcare costs.

The savings from UC are expected to build up with the rollout of the benefit. The process was originally due to have finished by 2017-18, but various delivery challenges have meant repeated delays. Our forecasts for the UC caseload in specific years have therefore been revised down repeatedly, while the end-point of the rollout has been pushed further into the future. Based on current plans, the rollout is expected to be near complete by 2022-23 with close to half of the caseload having migrated to UC by 2019-20.

As UC continues to roll out over the next five years, it will have an increasingly large impact on spending on the benefits it is set to replace. While the effect on spending on UC’s predecessors was small in 2016-17, spending is expected to be around £3 billion lower in 2017-18 and £8 billion lower in 2018-19 as larger numbers of claimants migrate to the new system.

We detailed the real-world and forecasting challenges posed by the transition to UC in our January 2018 Welfare trends report.

For more information on welfare spending please see: