No half measures

The Government’s long-awaited Child Poverty Strategy is due next month, close to, or contemporaneous with, the Autumn Budget. There have been some welcome announcements already: the over-indexation of the Universal Credit (UC) standard allowance for the next four years, for example, and the extension of free school meals to all children in families on UC in England. But despite these actions, child poverty rates are still forecast to soar, from 31 per cent (4.5 million children) in 2024-25, to 34 per cent (4.8 million) by the end of 2029-30.

Using the latest economic data, our modelling shows that if the Government wants child poverty to fall it has no choice but to place income support at the heart of an “ambitious” Child Poverty Strategy. There is simply no form of partial repeal of the two-child limit that would result in child poverty rates being lower in 2029-30 than they were in 2024-25. But in one fell swoop, the Government could reduce the number of children growing up in poverty by 330,000 today and save a further 150,000 children from that fate by 2029-30, if it were bold enough to scrap the two-child limit in full.

Repealing the two-child limit at the Budget would be an unequivocal step in the right direction by the Government. However, even then, their ‘child poverty headroom’ – the gap between our projection for 2029-30 and the 2024-25 baseline – would still be slight (0.4 percentage points), leaving the Government at risk of not achieving any reduction in child poverty over the Parliament if economic conditions were to change.  

The wait is almost over: after more than a year of discussion and debate, the Government is expected to publish its Child Poverty Strategy next month, close to, or contemporaneous with, the Autumn Budget. The link to a fiscal event is important. Previous Resolution Foundation research has shown that even with optimistic assumptions about (higher) parental employment rates and (lower) housing costs in the future, child poverty is still set to remain on a stubbornly upward trajectory over the course of the Parliament. The unavoidable conclusion is that no Child Poverty Strategy will be credible unless it is accompanied by significant additional state support for family incomes.

Without further action, child poverty will hit a historic high by the end of the Parliament

The Government, of course, is far from oblivious to this reality, and has already announced policies to support lower-income families. In June’s Spending Review, for example, it announced that all children living in families in receipt of UC in England would be eligible for free-school meals (FSM) from September 2026, reducing the number of children in poverty by an estimated 50,000-100,000 by 2029-30. Added to this, at the Spring Statement, the Government committed to over-index the Universal Credit (UC) standard allowance for the next four fiscal years, confirming this in the Universal Credit Act 2025 in July. But at the same time it also said this would be accompanied by a large cut in the value of the Universal Credit health element for new claimants from April 2026, and current Universal Credit Health recipients will not benefit from the above-inflation rise in the standard allowance, offsetting any child poverty reduction effect as a result.[i]

Figure 1: Under current policies, child poverty is set to reach record highs by 2029-30

However, as Figure 1 shows, welcome as they are, these policy changes do not radically shift the child poverty outlook over the forecast period.[ii] Using the latest economic data,[iii] we estimate that in 2024-25 (i.e. when the Government took power), relative child poverty measured on an after-housing costs (AHC) basis stood at 31 per cent, equivalent to 4.5 million children.[iv] Rates have likely risen since, and although the FSM extension in 2026-27 will act to push rates down that year, it is working against such a strong child poverty headwind that the effect is sadly short-lived. As a result, in the absence of further policy decisions to support lower-income families, we estimate that rates will hit 34 per cent in 2029-30, higher than observed at any time since 1961. Numerically, that would be equivalent to 4.8 million children in poverty (the relatively small increase in this number from 2024-25 reflects a declining child population in the UK over this period).[v]

Partial repeal of the two-child limit is not what an “ambitious” strategy looks like

As we have documented before, the most cost-effective policy choice to drive down rates of child poverty would be to scrap the two-child limit in its entirety.[vi] In one fell swoop, a decision by the Government to remove the limit would result in an estimated 480,000 fewer children growing in poverty in 2029-30 than the counterfactual, at a cost of £3.5 billion (or £7,280 per child lifted out of poverty). According to the Government’s own statistics, the majority (63 per cent) of households that would benefit from the repeal of the two-child limit have three children; over half (54 per cent) of those that would be affected are working households; and 40 per cent contain at least one adult or child with a health condition or disability.

That result is clearly an attractive proposition for those who care about child poverty; the price tag far less so. Small wonder, then, that it is rumoured the Government has been exploring options for partial reform of the two-child limit, most notably lifting the limit for working families only; moving to a three-child limit which would restore benefit entitlement for third children but not for any fourth or subsequent child; and scrapping the limit in full but then paying the UC child element at a reduced rate for third and subsequent children. However, as Figure 2 shows, all of these partial-repeal alternatives would leave child poverty rates higher at the end of the forecast period than at the beginning (just under 32 per cent compared to just under 31 per cent in 2024-25).[vii]

Figure 2: Child poverty will be higher in 2029-30 than 2024-25 if the Government does not fully scrap the two-child limit

And there are other problems with these options too: the creation of sharp cliff edges (lifting the two-child limit for those in work only); the continued disconnect between need and entitlement (lifting the two-child limit for those in work only and the three-child limit); and retention of the troubling non-consensual conception exemption (ditto).[viii] Even the arguably most logical half-measure – repealing the limit but paying the child element at a lower level for third and subsequent children given potential economies of scale – has its own issues. First, research contests the idea that the cost of a child falls considerably for third children onward. Second, children’s benefits are already paid at a lower level today than they have been historically: the higher child element for first children was abolished for children born after April 2017, a cut worth £554 in 2025-26, and the UC child element for second children is now £309 per year lower (8 per cent) in real terms than it was when first introduced in 2013.

Fundamentally, all the partial repeal options are inconsistent with manifesto and subsequent promises to develop “an ambitious child poverty strategy”. Although the Government has yet to indicate explicitly which year it will measure progress against, clearly child poverty rates must come down across the Parliament for it to be able to claim success. But whichever year it chooses – 2023-24 (technically the most accurate – but the hardest to meet), 2024-25 (our central case), or 2025-26 (the least defensible option – but the easiest to meet) – none of the compromises with respect to the two-child limit look set to bring child poverty rates down Figure 1below the baseline year in any meaningful way.

There is no good reason to delay the repeal of the two-child limit

These findings leave the Government in a decidedly unenviable position as it approaches the Autumn Budget, with external estimates expected to show the Chancellor must come up with significant tax rises or spending cuts before funding any new measures in the Child Poverty Strategy. But the line that the two-child limit can only be repealed “when economic conditions allow” looks increasingly threadbare for three key reasons.

First, as previous Resolution Foundation research has shown, there are plausible sources of additional revenue that the Chancellor could exploit to cover the costs of scrapping the two-child limit, such as a salt and sugar tax.[ix] Second, costs in 2029-30 will be the same whether the government acts now or towards the end of the Parliament, and so it is a fiscal fiction to promise action later but not account for it. But third, and arguably the most compelling reason to act swiftly with respect to the two-child limit, any further delay has significant human consequences. As Figure 3 shows, the number of children growing up in poverty will simply keep increasing until the two-child limit is repealed. We estimate that by the end of this fiscal year, 29,000 more children will be growing up in poverty than in 2024-25 as a result of the policy, with all the economic, social and human costs that entails both now and in the future. Moreover, there is no practical reason the two-child limit could not be repealed immediately following the Budget: changes can be made rapidly to the UC system, as the £20-a-week uplift introduced at the outset of the pandemic ably illustrated.

Figure 3: An additional 29,000 children will be living in poverty next year if the two-child limit is retained

Repealing the two-child limit in full is the most important step, but would still leave the direction of child poverty in the balance

There is broad consensus that scrapping the two-child limit in full is the most powerful policy lever the Government can pull to drive down child poverty rates, and the most cost-effective one too. But plausibly the Government’s programme cannot stop there: without further policy action, we estimate that child poverty rates will have crept back up close to 2024-25 levels by the end of the forecast period in 2029-30, leaving the Government with little ‘child poverty headroom’ against a 2024-25 baseline – just 0.4 percentage points. Moreover, our modelling is sensitive to forecasts especially with respect to earnings and rents: in particular, the assumption for future earnings growth used in our projections is in line with the OBR’s projection from March 2025 and may turn out to be conservative.[x] In the event of faster earnings growth, child poverty would rise more quickly as the median income would pull away faster from those at the bottom.

Given this, in Figure 4, we model the impact of two additional reforms that would place significant downward pressure on child poverty rates were they to be enacted in this Parliament, and would renew the link between benefit entitlement and need. First, as we have highlighted before, many families that stand to gain in the first instance from repeal of the two-child limit will then be hit by the benefit cap instead. Removing the benefit cap is therefore an important corollary to lifting the two-child limit. Second, the current freeze in value of Local Housing Allowance (LHA) bites very hard on lower-income families with children that live in the private rented sector and receive support from the state with their housing costs. Rapid rent rises since the last LHA rebasing in April 2024 mean that the gap between support and real rents is already untenably large. But it is a permanent relinking of LHA to local rents rather than sporadic uprating that is key to sustained child poverty reduction.

Figure 4: Relinking the Local Housing Allowance to the 30^th percentile of local rents on top of scrapping the two-child limit would achieve a significant reduction in child poverty this Parliament

Taken together, repealing the two-child limit in full, lifting the benefit cap and permanently relinking LHA rates to local rents would be a powerful programme to reduce child poverty over the Parliament and beyond. We estimate that the combined effect of all three would drive down child poverty rates from 31 per cent in 2024-25 (32 per cent in 2025-26), to 29.5 per cent in 2029-30. That would be equivalent to 660,000 fewer children growing up in poverty in 2029-30 than will be the case absent of any policy action.

But looking further out – as the Government has indicated it will do in the Child Poverty Strategy where it will set out a ten-year plan – the picture is less propitious. Rates will drift up inexorably without a longer-term structural shift to uprating working-age benefits in line with earnings rather than prices. The Government has committed to uprating the UC standard allowance above inflation for the next few years; permanently pegging that to earnings and extending to other benefits is required to get the projection for child poverty to remain flat in the long term. Such an approach that may seem radical for the UK but is the norm in countries such as New Zealand, Germany, Belgium and the Netherlands.

Conclusion

The Government is confronted by an unenviable set of fiscal choices this autumn as it finalises both its Child Poverty Strategy and its Autumn Budget. But it also faces a significant political choice as well. On the one hand, it is clearly concerned about the threat of angry headlines if it is seen to go ‘soft on welfare’, especially in the wake of two previous social security U-turns (on cuts to the Winter Fuel Payment and disability benefits). On the other hand, there is significant political jeopardy in failing to take bold action too. First, significant numbers of Labour backbenchers hold the promise to reduce child poverty very dear, and will no doubt hold their leaders closely to account for that as well. Second, the Scottish Government’s action to provide additional children’s benefits and to protect families against the two-child limit from April 2026 mark a clear dividing line in approach.

But as it seeks the political courage it needs to truly set child poverty on a downward course over this Parliament and beyond, Figure 5 should also give the Government pause for thought. On current policies, child poverty is set to rise by more than half a percentage point a year over the course of this Parliament, which would be the worst outcome on record for any Labour-led government in over 60 years. Partial reform of the two-child limit brings the growth rate down, but it is only the full repeal of the limit that is sufficient for child poverty to fall. Decisively turning the tide on child poverty, if current economic forecasts prove correct, would require action on other fronts too. As the Child Poverty Strategy enters the final furlong, the time has come, as the Education Secretary recently put it, for the Government to “do the right thing”.

Figure 5: Without significant action, the Government will preside over the sharpest increase in child poverty by any Labour Government in the last 60 years

[i] The Government has also announced a number of initiatives designed to improve the lives of lower-income children in particular in recent months, such as investment in Best Start Family Hubs which are akin to the Sure Start centres of the 2000s, and a library in every primary school by 2029.

[ii] Data citation: Department for Work and Pensions, NatCen Social Research. (2021). Family Resources Survey. [data series]. 4th Release. UK Data Service. SN: 200017, DOI: http://doi.org/10.5255/UKDA-Series-200017.

[iii] Full details of our projection methodology are available in the Annex of A Clegg and A Corlett, The Living Standards Outlook 2024, Resolution Foundation, August 2024. We use the OBR’s projection for earnings growth from its March 2025 Economic and Fiscal Outlook. Our starting point in accounting for inflation is the Bank of England’s August 2025 CPI forecast, but we create a specific deflator for ‘after housing costs’ income, in line with DWP definitions. This removes any housing costs from CPI to avoid double counting their impact. Private rents are assumed to rise in line with average earnings in future, with a 12-month lag. ONS’s Price Index of Private Rents is used in our nowcasting, but no regional variation in price changes is assumed beyond September 2025.

[iv] The Government has yet to indicate which year it will use as its baseline to measure success in its Child Poverty Strategy. Technically, the most accurate date to choose would be 2023-24, reflecting its inheritance from the year prior to taking power as well as the most recent year of outturn data for the official poverty statistics. In this spotlight we use 2024-25 as our central case, not least because early indications from Government suggested this was their preferred option. Picking 2025-26 on the ground that this is the year in which the Strategy is being published would flatter their figures (given we estimate child poverty is higher this year than in the two previous years) but is clearly not defensible as a measure of progress over the full Parliament.

[v] In 2024-25, there was an estimated 14.0m children in the UK, a figure that is projected to fall to 13.4m by 2029-30. See: ONS, National Population Projections. Because of this, it is more appropriate to focus on the projection of the proportion rather than the number of children in poverty.

[vi] The two-child limit is the rule that all families receiving means-tested benefits are not eligible for support with any third or subsequent child born from April 2017 onwards, bar those subject to limited exceptions.

[vii] In numerical terms, we estimate that moving to a three-child limit would reduce child poverty by 350,000 in 2029-30 compared to our projection for that year based on current policy; a lower child element for third and subsequent children (at two-thirds of the standard child element) would bring the number down by 340,000; and scrapping the two-child limit only for families in work would reduce child poverty by 330,000 by that date. These numbers are different from those in our work in May 2025 as they have been updated using newer FRS data, updated economic assumptions and the latest policy assumptions.

[viii] The non-consensual conception exemption has been shown to be intrusive for claimants, to have serious issues with implementation and has thrown up legal issues around safeguarding and ambiguous legal definitions. Moving to a system that does away with any need for such a clause should be (another) key motivation for ending the two-child limit in full. For a more detailed discussion of the pros and cons of these and other partial reform options, see A Clegg and A Corlett, Limited Ambition: An assessment of the rumoured options for easing the two-child limit, Resolution Foundation, May 2025.

[ix] See Chart of the Week here for the distributional impact of scrapping the two-child limit while introducing a salt and sugar levy.

[x] The Bank of England has more recently forecast faster earnings growth, for example.