Saving private renters

Local Housing Allowance (LHA), which caps how much support is provided through the social security system to low-income private renters, was last relinked to the 30^th percentile of rents in April 2024, and has been frozen in cash terms ever since. But private rents have risen considerably since then. [1]  A low-income family renting a typical two-bedroom home faces a shortfall of £158 a month, and the gap between average rents at the lower end of the market and the housing support a private renter can receive via the benefit system is set to reach a record high of 23 per cent this October.

Governments of various persuasions have claimed that LHA rates feed through to rents, inhibiting policy makers from relinking housing support in the fear that the increase would pass through to landlords. But new analysis of the April 2024 LHA uplift suggests only a small share of increased support is captured by landlords, with a central estimate of around 10p in every pound of the uplift feeding through to higher rents at the bottom of the market. Consistent with this, the share of low-income private renters whose LHA fell short of their rent dropped from 67 per cent just before the LHA increase to 46 per cent a year later, and was still below its pre-uplift value in May 2026.

These findings strongly suggest that relinking LHA to rents at the 30th percentile mainly benefits low-income tenants, and that only a small part leaks into higher rents. This mirrors DWP-commissioned research on the 2011 cut to LHA, which found that around 90 per cent of that policy change fell on tenants. As a result, the Government should commit to relinking LHA to the 30th percentile of local rents on an annual basis at the Autumn Budget. This comes with a cost of £2 billion a year by the end of the Parliament but, if necessary, could be funded by increasing the Universal Credit taper.

The gap between Local Housing Allowance and market rents will hit a new high this autumn

The rising cost of living has hit many households hard in recent years and shows no signs of abating. But lower-income private renters have had to contend not just with the rising price of energy, food and other essentials, but also with increasingly burdensome housing costs as the amount of housing support they receive from the state has shrunk relative to rents.

Local Housing Allowance (LHA) determines the maximum amount of help that the 1.9 million privately-renting households in Great Britain on Universal Credit (UC) or Housing Benefit (HB) can receive from the state with their rental costs. LHA caps the amount of benefit paid for each property size in each of the 192 Broad Rental Market Areas (BRMAs) in Great Britain. When first introduced in 2008, LHA was set at the median rent for each local area and increased in line with local rents. In 2011, LHA was cut from the median rent to the 30^th percentile of rents. The automatic link to market rents was ended a year later, and rates have been frozen in nine of the past 14 years. But rents continue to rise when rates are frozen, meaning every month the LHA covers a little less of the market rent.

The impact of this is shown in Figure 1. By March 2020, four years into a cash freeze, LHA had fallen 14 per cent below the 30^th percentile of rents, on average. The April 2020 one-off rebase closed that gap, but rates were then frozen again and, by March 2024, rapid rent inflation had widened it to 23 per cent. Rates were re-linked to the 30th percentile again in April 2024, but have been frozen since.^[2] Rents have continued to climb since: by August 2026 the gap had reached almost 23 per cent, close to the March 2024 record.^[3] Across England, the rent of a two-bedroom home at the 30th percentile now sits £158 a month on average above the LHA rate.^[4] Small surprise, then, that in July 2026, Crisis and Citizens Advice estimated that fewer than 2 per cent of homes listed for rent in Great Britain could be covered by LHA alone.

Figure 1: The gap between Local Housing Allowance and prevailing private rents is close to a record high

The problem is not confined to the most expensive parts of the country. In more than half of BRMAs, the gap between the LHA rate and the prevailing market rent for a two-bedroom home is above £100 a month, with every English region containing at least one such area.[5] But unsurprisingly, the gap is widest in London: it reaches £324 a month in Inner East London and is above £250 a month in three other London areas.[6]

And things can only get worse. As Figure 1 also makes clear, we estimate that if rents keep growing at their recent pace of just under 4 per cent a year, which is also the rate the OBR assumes for private rents over the long run, then by October 2026 the gap will surpass the previous high of 23 per cent observed in March 2024 before the last rebasing.

The concern that landlords will capture LHA increases is inhibiting policy action

Even though the gap between housing support and market rents is reaching an increasingly untenable level, there is clearly a concern that if the state increases LHA, landlords may simply raise rents in response. Economic theory certainly suggests this could be the case, and this belief clearly has informed government thinking.[7] In 2011, for example, the Department for Work and Pensions said of the LHA cut made that year that “the purpose of reform is to influence rent levels and housing choices”, and its 2012 assessment cited local authority reports that “some landlords use the transparency of the arrangements to raise rents to the Local Housing Allowance level”. More recently, in July 2025, Angela Rayner alluded to this concern when she said that “the only way we are going to fix this [housing] crisis is not by giving more money to private landlords for people who should be in social housing”. Andy Burnham echoed the sentiment in a speech in June 2026, asserting “We are forced to chase rents in the private-rented sector through the benefits system”. (His solution, too, was to build more social housing).

But does this theory hold true in practice? The best UK evidence to date comes from analysis of a cut to LHA, not a rise. As discussed, in 2011, LHA rates moved from being set at the median rent to the 30th percentile. DWP-commissioned research analysing the incidence of this policy change found that around 90 per cent of this LHA cut fell on tenants in the short run, with only around one-tenth showing up in lower rents. Landlords bore more of the cut where the old rates had been most generous relative to what tenants needed.

If rises in LHA have the same impact on the market, then this research predicts that it would be tenants and not landlords who would primarily benefit. But the incidence of a rise could differ from that of a cut. Rents are sticky downwards: landlords rarely cut the rent of a sitting tenant. In contrast, raising the rent when demand rises is routine. So, what did happen to rents when LHA rates were increased in April 2024?

Landlords captured only a small part of the 2024 LHA rise

One way we can answer this question is to use data released by the Valuation Office Agency (VOA), which covers English BRMAs only. This data captures the rents of open-market lettings, and rent officers aim to remove lettings of tenants in receipt of housing support via UC or HB where they can identify them. Although this is unlikely to be an exact science, this data should, in theory, measure rents paid by non-recipients of LHA, allowing us to assess whether the uplifts to LHA have had spillovers into the rents paid by non-recipients.[8] (We turn to claimants’ own outcomes in the next section).

We start with the aggregate picture in Figure 2 which compares growth in rents at the 30^th and 60^th percentiles. (We select the 60^th as our counterpoint as this is sufficiently far from the 30^th as to be unaffected by changes in that segment of the market). Between April 2020 and April 2024, the 30^th percentile grew 16.5 per cent against 19.1 per cent at the 60^th. After the 2024 relink, rents at the two percentiles grew at a similar rate, both rising 16.2 per cent by April 2026. In other words, at the national level, neither instance of LHA relinking saw the bottom of the market pull away from the middle, consistent with landlords capturing little or nothing of either LHA uplift.

Figure 2: Rents at the bottom of the market did not outpace the middle after the increase to LHA in either April 2020 or April 2024

And we can learn more from looking at the variation across areas. The April 2024 relink raised LHA by different amounts in different areas, making it a useful natural experiment. Across the 150 English areas where the relinking took effect in full, the two-bedroom rate rose by between £5 and £62 a week, or by between 5 and 35 per cent, with areas where rents had grown fastest since 2020 seeing the largest uplifts.

We exploit this variation in Figure 3 which plots the cash rise in the 30^th percentile of two-bedroom rent between 2024 and 2026 minus the rise it would have had at the 60^th percentile’s growth rate, against the size of the April 2024 cash uplift for each BRMA (as before, the 60^th percentile acts as a control, capturing area-specific trends in rents unrelated to LHA changes).[9] To estimate the size of pass-through exactly, we use a regression framework, weighting areas by the size of the rental market and controlling for the rural share of postcodes and for London.

Figure 3: Areas that received larger cash uplifts to LHA rates in 2024 have seen only slightly bigger rent rises at the bottom of the market

If landlords had captured the LHA rise, the scatter would slope steeply upward, at close to one-for-one. As it is, areas with large uplifts saw only slightly bigger rises at the 30^th percentile than areas with small uplifts in LHA, and the line slopes upwards only gently, with a slope of 0.10 meaning that, for every £1 a week added to the two-bedroom LHA rate in April 2024, the 30^th percentile rent rose by around 10p a week more than it would otherwise have done over the following two years. The standard error is 0.08, so the data is consistent with anything from no spillover to around one-quarter. But what can be ruled out is a large spillover into rents that some seem to assume.^[10]

Government can rule out that landlords captured all of the 2024 LHA rise by raising rents

As mentioned above, the VOA data can tell us about rents of non-LHA recipients but not about what happened to the rents of LHA recipients themselves. The analysis above suggests that the spillovers to non-recipients were small, but it is theoretically possible that the rents paid by LHA recipients could have changed by more if the market for rental housing was strongly segmented between those who do and do not receive LHA. Although we do not have data on the rents paid by LHA recipients, we do know how many LHA recipients are facing a shortfall, meaning that the rent they pay is greater than the LHA in their area. Data from DWP tells us that between March 2024 and April 2025 the proportion of LHA recipients with a shortfall fell by 12 percentage points in the third of areas with the smallest uplifts, and by 20 points in the third with the largest (see Figure 4). The proportion with a shortfall rose slightly over the next 12 months but still remained below its March 2024 value in May 2026.

Figure 4: The proportion of claimants with a shortfall fell most where the April 2024 LHA uplift was largest

As a result, there were 154,000 fewer UC households with a rent above their LHA in March 2025 than in March 2024 in England, even though the number on LHA rose by 124,000 over the year.[11] This does not rule out that the rents paid by LHA recipients rose slightly as a result of the LHA rise, but it does emphatically rule out that landlords captured all of the LHA rise in higher rents, and is consistent with the idea that claimants’ rents move with the market, not with LHA levels.

For most claimants, the first pound of LHA uplift will go on essentials not on rent

This new analysis of the 2024 relink suggests that only around 10p in every pound of extra LHA showed up as higher rents at the bottom of the market, and that the uplift led to a considerable fall in the fraction of LHA recipients facing a shortfall. These both suggest that the uplift led to genuine improvement in rental affordability. This is a similar conclusion to the study of the 2011 cut, where around 90 per cent of the cut was estimated to fall on tenants; both agree that landlords did not capture most of the change.

But why is pass-through so small? Economic theory assumes that LHA works as a true rent subsidy, i.e. money that can only be spent on rent. But there are good reasons to conclude that, in recent years at least, it has not operated in that way for all recipients. In particular, on the eve of the April 2024 relink, 67 per cent of UC households receiving LHA in England had a rent that was higher than the LHA rate in their area, and so had to use other income to pay their rent. For these households, rent is already eating into the money for everything else, meaning that a higher LHA simply shrinks the shortfall. It seems unlikely, then, that a household receiving more LHA would use the extra money to look for a more expensive property and hence boost housing demand. Rather, a rise in LHA works more like a rise in the standard allowance of UC than like a pure rent subsidy.

Our supposition that many LHA-receiving households would spend any extra housing support on items other than rent is bolstered by the high proportion who are going without key essentials. In 2024-25, 58 per cent of working-age adults in private-renting families on housing support were materially deprived on the DWP’s measure, meaning their family could not afford a set of basic items (see Figure 5).[12] One-in-five (21 per cent) could not keep their home warm, one-in-eight (13 per cent) could not afford three meals a day, and one-in-eight (13 per cent) could not afford fresh fruit and vegetables every day (among all working-age adults the equivalent figures are 12, 6, 4 and 4 per cent). So, the majority of privately renting benefit recipients are going without several essentials, making it entirely plausible that a higher LHA rate will be used to relieve those pressures before families seek more expensive housing.

Figure 5: Private renters on housing support already go without essentials

Our view is that the outcomes in 2024 are a good guide to the likely impact of a relink in 2027, because the shortfalls are just as large: in May 2026, 58 per cent of UC households receiving housing support paid rent above their LHA rate, and an April 2027 relink that set LHA to the level of rents from the year to September 2026 would still be below the rents that most claimants are paying.

The Government should relink LHA to the 30^th percentile in the Autumn Budget, and commit to annual uprating in the future

Given the economic orthodoxy, it is reasonable for policy makers to ask whether an LHA uplift simply benefits landlords in the form of higher rents. But the evidence we have presented in this spotlight suggests that this theory does not play out in practice. Rather, we conclude that relinking LHA to the 30^th percentile of rents would largely benefit those the policy is meant to help (i.e. low-income tenants), and that only a small part is likely to leak into higher rents.

Looking towards Autumn Budget 2026, we propose two changes. First, the Government should relink LHA to the 30^th percentile of local rents in April 2027. A failure to act on this issue risks the gap between LHA and prevailing rents rising to an untenable 30 per cent by March 2028, much higher than the March 2024 figure of 23 per cent that prompted a reset at that point. Second, the Government should restore automatic annual linking to local rents so that LHA is relinked to the 30^th percentile each year. This would stop significant gaps opening between housing support and rents which are placing such strain on some of the poorest families in this country. (This would also change the forecast default: maintaining the value of LHA would become the baseline rather than showing up as a spending decision each time rates are reset).

Of course, this does come with a price tag. We estimate that relinking LHA to the 30^th percentile in April 2027 and maintaining the link thereafter would cost £2 billion a year (in current prices) by 2029-30, compared to the (entirely unsustainable) assumption of no nominal change.[13] That is not an insignificant sum especially in such fiscally straitened times. But if the relink has to be paid for from within the working-age welfare budget, one option to consider would be to raise the taper rate in Universal Credit from 55 to 58 per cent. That is not without its downsides too, but this pair of reforms together would have the virtue of making the social security system more equitable overall.

[1] The authors gratefully acknowledge advice and research support provided by colleagues Alex Clegg and Greg Thwaites. Any errors, however, are the authors’ own.

[2] As Figure 1 makes clear, the April 2024 relink did not fully close the LHA-rent gap. This is because the new rates were based on rents from the 12 months to September 2023, but between that time and when the new rate took effect in April 2024 rents had risen another 9 per cent. As a result, on the day of the relink the gap shown in Figure 1 still stood at 9 per cent.

[3] The gap in Figure 1 compares the ONS Price Index of Private Rents with an index of LHA. The LHA index starts at the April 2012 rent level, moves each April by the uprating decision, and resets in a relink year to the average rent over the twelve months the new rates were set on. The rent index covers average rents on all tenancies in Great Britain, while LHA is set on the 30th percentile of new lets in English Broad Rental Market Areas. So the gap is one national summary, not an average of local shortfalls. It assumes that rents at the 30th percentile grow in line with average rents.

[4] The £158 is the two-bedroom rent at the 30th percentile in the April 2026 List of Rents minus the frozen two-bedroom LHA rate, converted to a monthly figure and averaged across the 152 English areas weighted by Census 2021 private renter households. It is not the median shortfall across LHA recipients. Excluding the two areas held at the national cap it is £140 a month, and in the median area it is £125.

[5] Throughout this spotlight we present results for two-bedroom properties as this is the most common-sized property in the PRS (39 per cent of privately rented homes in England had two bedrooms in Census 2021, against 29 per cent with three and 21 per cent with one). However, we have also produced results for all other property sizes and the findings we present hold true for all with the exception of the shared accommodation rate, a small and volatile market where the estimate is noisy.

[6] The national caps held the two-bedroom rate at £412.86 a week in two of the 152 English areas, Central London and Inner North London, so the relink barely changed their rates. Across all five property sizes, seven of the 760 area-by-size rates sat at a cap. For example, in Central London, the two-bedroom gap is £1,200 a month. We exclude the two capped areas – Central London and Inner North London, where national caps set in 2011 hold LHA well below the 30^th percentile of those areas – from the pass-through analysis.

[7] Economic theory suggests that how much landlords gain from a housing subsidy like LHA depends on how responsive supply and demand for housing is. The most important factor is whether the supply of rental properties responds. If the stock of the sort of homes rented by people receiving LHA is fixed, then any extra demand from a subsidy feeds into higher rents, and landlords capture all of the subsidy. If supply can increase, then rents will not rise by as much, and the size of any rent rise will depend on how price-responsive are both the demand for, and supply of, the types of homes rented by people who receive LHA. It is important to note that none of this depends on individual landlords knowing which tenants receive LHA: the effect works through the market as a whole.

[8] This is a different approach to that used in M Brewer et al., The curious incidence of rent subsidies: Evidence of heterogeneity from administrative data, Journal of Urban Economics, November 2019 which drew on DWP data of claimants’ actual rents.

[9] Figure 3 is a partial regression plot. A plain scatter of the two variables would show a line whose slope differs from the regression estimate, because that estimate also allows for the rural share of postcodes and for London. So for each area we first take the part of each variable that those two controls do not explain, using the same weights as the regression, and plot those parts against each other. The slope of that plot is exactly the regression coefficient. We add each variable’s average back so the axes read in pounds a week.

[10] All estimates weight areas by the number of rents observed, control for the rural share of postcodes and for London, and measure the uplift and the response in pounds a week. We tested this result in many ways, and none changes the picture. Adding the proportion of private renters on LHA in each area or the proportion of claimants whose rent already exceeded their LHA, excluding London, changing the response window, using other bedroom sizes or other percentiles as the control, dropping any single area, or accounting for long-running differential rent trends all leave the slope small and well short of one.

[11] Source: DWP, Stat-Xplore, Households on Universal Credit, by local authority and Local Housing Allowance indicator, March 2024 and April 2025. The indicator splits private-rented UC households into those whose LHA covers their rent and those whose rent is higher. We take the counts at local authority level and allocate them to the 152 Broad Rental Market Areas by postcode share, because the BRMA field in Stat-Xplore drops 18 areas.

[12] Source: RF analysis of DWP, Family Resources Survey and Households Below Average Income, 2024-25, Great Britain. Working-age adults are in material deprivation if their family lacks five or more of 21 everyday items because they do not have the money for them or because they are not a priority on their current income. Private-renting families on housing support are those in the private rented sector receiving Housing Benefit or Universal Credit. UC receipt stands for receipt of the housing element, which slightly overstates the group. The sample is around 740 adults. DWP changed the deprivation questions in 2023-24, so these rates are not comparable with earlier years.

[13] This figure supersedes an earlier estimate of £2.5 billion first published in M Cavanagh et al., See it. Say it. Sort it: How to reset Britain’s economic policy, Resolution Foundation, May 2026. This update has means that the UC taper would need to be increased from its current 55 per cent to 58 per cent, rather than the 59 per cent we recommended in May 2026.