How tax reform would make rent controls feasible to deliver

Over the same period, the share of households paying these high rents has grown. As the stock of social housing was steadily depleted following the introduction of Right to Buy in 1980, and soaring house prices throughout the 1990s and early 2000s priced increasing numbers out of homeownership, more households had to make their home in the private rented sector (PRS). Over the 2 decades to 2020, the proportion of all households living in the PRS increased by 2.7 million homes, from 11% to 19% of the total housing stock — meaning almost 1 in 5 households in the UK now rent privately (Baxter et al., 2022). This includes a significantly higher proportion of lower-income households than historically lived in the PRS: in 1994/95, 40% of households on low incomes (bottom 2 quintiles of the income distribution) lived in social housing, with 12% living in the PRS. By 2023/24 this had changed to only 26% of low-income households in social housing and 20% living in the PRS.¹

The number of children and older people living in the sector has also increased significantly. Just under a quarter of all children now live in the PRS, up from 8% in 2000 (DWP, 2025); meanwhile the number of people aged over 65 living in the sector is forecast to increase from 500,000 in 2022 to over 2 million by 2040 (Independent Age, 2024).

This longer-term view of affordability pressures in the PRS is key to understanding why the very steep rent inflation of the past few years hit renters so hard. Though average rent-to-income ratios in 2024/25 sat at broadly the same level they have for the past 15 years — owing to the fact that the steep rent increases across 2023 and 2024 followed a period of high wage growth following the Covid-19 pandemic – this nonetheless means that renters have seen little to no growth in their disposable income during a period where other essential costs like food and energy have also increased significantly (Blower et al., 2026). The permanent squeeze on renters’ disposable incomes from rents taking up a persistently high proportion of incomes, in other words, leaves them much more exposed to the impacts of other essentials spiking in price.

It’s not surprising, then, that polling conducted for JRF by More in Common found that renters are more likely than mortgagors to not have any financial buffer left at the end of each month, and more likely to not be putting anything away as savings.² Nor is it surprising that renters spending a third or more of their incomes on rent are significantly more likely to experience material deprivation, and be in after-housing-costs poverty, than those spending a lower proportion of their incomes on rent (Blower et al., 2026).

The Office for Budgetary Responsibility (OBR) forecasts that private rents will continue to increase at the same rate as wages until the end of the decade (OBR, 2026). For renters, this means more of the same: rent continuing to take up a significant proportion of their incomes and eating into any much-needed growth in wages, making it harder to absorb unexpected increases to other bills, make progress on saving a deposit to buy their own home, put money aside to weather emergencies, or spend in their local economies.

This affordability pressure has rightfully focussed attention on how to address the high cost of rents households face. But to date, policy has found it difficult to make sufficient inroads into dealing with this issue.

Benefits crucial for low-income households, but won't solve rent affordability

A principal lever for addressing high housing costs is through the social security system, with help with housing costs for low-income households being available through Universal Credit (UC) and Housing Benefit (HB). These schemes have a vital role to play in supporting those who would otherwise be unable to afford the cost of renting privately — but on their own, they are inadequate for solving the problem of rent affordability pressures in the round, for 2 distinct reasons.

The first reason is that the way these schemes are designed and delivered means they often don’t offer the low-income households they are intended to support an adequate level of subsidy. UC and HB levels are set using Local Housing Allowance (LHA) rates, which are set at the 30th percentile of rents for properties of a certain size in a specific area. But we estimate that around 1 million households currently receiving UC or HB for their housing costs live in properties with rents >30th percentile of rents for similar properties in their area. This means the level of support they receive is less than the actual rent cost they need to pay to access housing in their area, leaving them to cover the difference from other income (which, for families who rely on Universal Credit to cover their living costs, may already be inadequate) (Blower et al., 2026). The benefit cap constrains this support even further, particularly in high-cost areas.

Moreover, although LHA rates are supposed to be set at the 30th percentile of local rents, the trend amongst recent governments has been to periodically freeze the rate rather than uprating it every year to reflect current market rents — largely in response to worries about the growing and unpredictable cost of the benefit owing to high and uncontrolled private rents. This means that even for households whose rent is ≤30th percentile of local rents when they first move into a property, the support they receive through UC or HB will erode over time if their rent increases and LHA is not uprated.³

Beyond issues with the adequacy of the support delivered to low-income households, the second reason to look beyond the benefit system for solutions to unaffordable PRS rents is that the impacts of high rent costs are felt much further up the income distribution that the benefit system is intended to target. Particularly in high-demand rental markets — London, other major cities and areas with high numbers of commuters — average rents are taking up more than 30% of the incomes of renters on median incomes (ONS, 2025).

Though the impacts are less acute than for those on lower-incomes, for this group rent costs are nonetheless a source of downward pressure on living standards, holding them back from getting on the housing ladder, starting a family, or building up their longer-term financial security — all of which have societal impacts beyond the significant number of households immediately affected.

This doesn’t mean that improving the level of support delivered through UC and HB is not important; ensuring that the help available through the benefit system is adequate to the costs faced by households is key to having an effective safety net. But it does mean that we can’t look to the benefit system to turn the dial on the problem of PRS affordability on its own.

Action on drivers of high rents is key, but can’t be delivered quickly

The second lever government has to address the cost of rent is through the supply of new homes. Structurally, high rent costs relative to incomes — and indeed, high house prices relative to incomes — stem from an imbalance between the supply of and demand for housing.

A combination of widening access to mortgage credit, a period of extremely low interest rates, population increases and changes in the way we consume housing have all increased the demand for housing space, while the elasticity of housing supply has remained low owing to the restrictiveness of the planning system, the structure of the development market, and low investment in public housebuilding in recent decades. This has in turn driven up prices: between the first quarter of 1996 and the final quarter of 2007, average house prices increased by more than 170% in real terms (Savills, 2022).

This imbalance has underpinned the dependable gains homeowners have seen in the value of their properties — and these gains, in turn, were the backdrop to the substantial expansion of the private rented sector during the 2000s and 2010s. The ability to charge rent on an asset that dependably appreciated in value made investing in becoming a landlord particularly attractive. Access to Buy-to-Let lending, with the ability to borrow against rental income, encouraged small-scale, amateur investors to buy up and rent out existing homes.

Renters are on the other side of this coin. Renting households who do not qualify for a home in the depleted social rent sector, and for whom homeownership is out of reach, are stuck having to pay a stubbornly high proportion of their incomes to rent access to somewhere to live, from those who have been able to acquire homes as investment assets and benefit from returns on those investments in the form of capital gains and rental income.

Government is already taking steps to address the supply side of this issue, making this a central part of their electoral offer and economic and housing strategies. They have set a target of delivering 1.5 million homes across the course of the parliament and have backed this up with major reform of the planning system; set out a new 10-year Social and Affordable Homes Programme and accompanying sector support (such as a new rent settlement); and launched a programme of New Towns — all of which are welcome.

However, there are limitations to the extent to which a supply-side approach can materially impact rent affordability for renters in the near-term. This is for 2 reasons. Firstly, any pro-supply reform faces a number of headwinds that make housebuilding difficult in the short run. High interest rates weigh on affordability for residential buyers and undermine the investment case for institutional investors, while recent significant build cost inflation (which may worsen due to the situation in the Middle East) raises the cost of construction and further impedes viability (Baxter, 2026). These factors are already showing up in a drop in new construction and are likely to act as a downward pressure on housebuilding for some time, and potentially worsen if – as predicted – interest rates rise again. 

Secondly, we need to be realistic about the timescales needed to materially reduce rent levels through new supply. Even if the Government gets close to meeting its target of 1.5 million homes over the course of the parliament, Office for Budget Responsibility (OBR) forecasts estimate that rents will continue to absorb all average wage growth until at least the end of the decade, leading to no material impact on affordability. Hitting the target, or even increasing it, would still be a decade(s) long project to bring down prices and rents. 

This does not mean we should disregard action to increase the supply of new homes, alongside other levers that may aide affordability — increasing supply is vital to ensuring sustainably affordable housing in the long run. In fact, it means the Government could and should go further — taking action to increase public housebuilding, as previous JRF work has argued, and through further reforms to the planning system. But it does leave a question as to what is to be done about the economic insecurity of today’s renters. If we are to answer this, we need an approach that can act more quickly and directly on high rental costs than the lever of new supply.

Despite this potential, the government has so far resisted calls from renter advocacy groups to consider introducing rent controls — which reflects the fact that rent control has as many detractors and sceptics as it does vocal advocates. This opposition often hinges on the argument that, through compressing the rental incomes of landlords and dampening the returns they see on their investment, rent control policies can lead to a rapid sell-off of properties that would harm tenants by creating a shortage of available homes, and a hit to investment in the maintenance of stock, leading to poorer quality homes.

This is a risk we should take seriously. As we’ve previously argued, a shrinkage of the size of PRS as a proportion of the overall housing stock isn’t something we should be afraid of in itself — indeed, given that the PRS is now home to millions more households than it was 20 years ago, for many of which the tenure is not a suitable long-term home, it should in fact be a goal of housing policy to shrink the PRS as a proportion of the overall housing stock (Baxter et al., 2022).

International evidence on the impacts of rent controls has generally shown a positive impact on homeownership rates (Kholodilin, 2024), and previous JRF analysis has shown that a contraction in the rental market can advantage first-time buyers in particular (Elliott and Baxter, 2025). But it is important that this doesn’t happen too rapidly. A sudden contraction in the availability of homes to rent could cause harm to tenants, particularly those for whom buying their own home may not be immediately within reach. Though there would be a reduction in net rental demand from the households who are able to move into homeownership, this could nonetheless happen in a geographically and/or temporally lumpy way creating a shortage of available homes for those who need to rent.

Mitigating this risk is in large part about policy design. It also depends, though, on the level of returns landlords are making. Price controls are best deployed in markets where suppliers of goods or services as a whole are making supernormal rates of return on their investments (which would generally indicate that suppliers have some kind of market power that a price control could be used to correct). Under such circumstances, it’s possible to cap prices, and therefore profits, without pushing suppliers below the rate of return needed to justify their investment (averting a number of the challenges outlined above).

If price controls are deployed where supernormal rates of return aren’t being made, on the other hand, they can push suppliers below the required rate of return and cause them to exit the market. This is what opponents of rent control fear will happen if the Government intervenes to cap rent prices at a level below market rates: that landlords will no longer be making an adequate return on their investment, causing many existing landlords to sell up and deterring new entrants.

The extent to which these concerns are warranted, therefore, hinges on a key empirical question: do landlords in the PRS generally make supernormal rates of return on their investments? To answer this question, we commissioned the Autonomy Institute to produce a first-of-its kind analysis of the rate of return landlords in the PRS have made on their investments — comprised of both rental income and capital gains, minus borrowing and operating costs — across the timepoints for which we have detailed information on landlord investments from the English Private Landlord Survey.

The key findings of this analysis, as they pertain to the question of whether landlords enjoy supernormal returns on their investments, are summarised below; the full analysis, including detailed methodology, are available on the Autonomy website (Garcia and Stratford, 2025).

Landlords on average made supernormal returns on investments

Average pre-tax return on equity (ROE) for landlords across the sector over the time period analysed have been strong: landlords as a whole have made average annual pre-tax returns of 8.5% in 2018 and 9.2% in 2021. In 2024 average returns fell slightly, sitting at 6.9%, reflecting the impact on the sector of the sudden spike in interest rates in 2022.

In each of the 3 years analysed, these average rates of return exceeded by some way the benchmark rate of return for the broader real-estate sector, which sat at 4.75%-4.95% over the 2018-2024 period. This means that landlords in the PRS were making significantly higher average pre-tax rates of return on their investments than the pre-tax returns enjoyed by investors across the real estate sector as a whole. Indeed, in each of the 3 years, only a minority of landlords weren’t reaching or exceeding this broader industry rate of return. In 2018, 96% of all landlords saw returns at or above this benchmark; in 2021 it was 99%. 2024 saw the rate fall somewhat — but even then, 81% of landlords still reached or exceeded the industry benchmark.

A significant majority of landlords in each year also exceeded the rate of returns available from comparable investments accessible to typical buy-to-let landlords, representing their ‘opportunity cost of capital’ — meaning the extent to which, at a given time point, landlords could see higher returns on their capital by withdrawing their equity from the PRS and investing elsewhere.⁶ 90% of all landlords exceeded this benchmark in 2018, rising to 99% in 2021. Benchmark exceedance fell somewhat in 2024 to 79%, but that still means a large majority of landlords made returns above the level available in comparable investments.

The analysis also compares landlord returns against an alternative benchmark, representing returns in higher-risk, higher-reward investments in equities funds (which are by nature more volatile, meaning the benchmark itself moves significantly year to year). In both 2018 and 2024 only a minority of landlords met or exceeded this higher-risk benchmark. In 2021, however, 96% met or exceeded it — which shows that at least under certain conditions, landlord returns outperform even the returns available from comparably higher-risk investments, in addition to being less volatile.⁷

The first and most significant finding of the analysis, therefore, is that at least for the period examined, a strong majority of landlords have indeed been making supernormal rates of return on their investments. This should give us confidence that at least from the perspective of economic theory, a price control, in the form of some kind of regulation of rents, could be accommodated in the market without significant distortive impacts on the supply of homes.

The second significant finding of the analysis is that capital gains are often providing the lion’s share of total landlord return on equity (ROE) — which means that changes to expected capital gains from stronger or weaker house price growth are likely to play a more significant role in the returns landlords make on their investments than changes in rental income.

The analysis shows that capital gains have been the dominant component of total ROE for landlords in most areas across the 3 time points. And where there have been significant changes in regional average ROE levels between 2018 and 2024, these have largely been driven by changes in house price growth, rather than rates of return on rental income. This is most pronounced in London: average annualised capital gains in 2018 sat at 7.4%, but by 2024 they had come down to just 1.1%. Other regions saw the reverse trend: in the North East, annualised capital gains increased from just 0.7% in 2018 to 3.7% in 2024. By 2024, London was the only region where average ROE was falling below the benchmark of the wider real-estate industry.

This is an important point when considering the potential for landlord disinvestment in the sector: upswings or downturns in house prices play a bigger role in determining the competitiveness of the overall rate of return landlords make than fluctuations in rental income. This shouldn’t be surprising; a big part of the reason why Buy-to-Let was such an attractive investment in the 2010s was because of expectations of dependable house price growth (and very low interest rates, themselves a key driver of house price growth).

This means that periods of stagnant house price growth can cause total annualised returns for landlords to dip for a time. But it also means that, in the absence of broader structural changes to the supply-demand imbalance underpinning historic house price growth, we’ve no clear reason to expect the strong capital gains aspect of landlord returns to change significantly in the future.

The 2024 results for London stand out in terms of how far the level of total annualised returns is negatively impacted by lower annualised capital gains compared to previous years. Though the South East and East of England also see reductions in returns on the capital gains side in 2024 compared to earlier years, the effect is significantly less pronounced than it is for London. This reflects the downward pressure that strained affordability (owing to very high house prices relative to incomes and, more recently, higher interest rates making borrowing more expensive), as well as particular issues with buying and selling flats, has put on house price growth in London in recent years.

Forecasts for house price growth in London suggest that prices in the capital over the next few years will regain some more upward momentum — albeit at a slower rate than the rest of the country. Savills forecasts for house prices in London to grow 13.6% between 2026 and 2030 — which if borne out would equate to an annualised rate of growth over the next 5 years of roughly 2.7% (Savills, 2025).

If annualised house price growth had been 1 percentage point higher in the period leading up to 2024 — that is, around 2.25% instead of 1.25% — landlords in London would have been meeting the benchmark for return levels in the broader real-estate sector in 2024.⁸ This suggests that if house prices in London do grow at the rate broadly forecast by Savills and others⁹ over the coming years, landlords in London may return in the not too distant future to a position of making total returns that exceed this benchmark.

Though total ROE is comprised of both rental income and capital gains — with capital gains, as we have seen, being the main driver of overall returns — the capital gain aspect of returns is only realisable upon sale of the property. This means that looking at returns on rental income in isolation is important for understanding the day-to-day profitability of landlords (that is, their ability to cover all their ongoing costs — mortgage interest and other borrowing costs, operating expenses and tax) — from their rental income.

Across all 3 years analysed, the vast majority of landlords are profitable on the rental income side of their investment: 93% of landlords in 2018 and 2021, and 95% in 2024, made a profit on their rental income. The fact that the proportion making an overall positive ROE on their investment (between rental income and capital gains) each year is even higher — 98.8% in 2024 and even higher in 2018 and 2021 — suggests it may be a feature of the sector that some landlords will accept making a loss on their rental income in exchange for making a strong, positive return on the capital gain on their property.

It’s also likely to be the case, though, that some of the landlords making a loss on their rental income at the point of each of the survey waves were on the verge of selling their properties as a result. Whether an individual landlord is able to weather a period of negative returns on their rental income will depend on their wider financial situation, in particular whether they have other sources of income from which to make up for a shortfall in their rental income in covering their costs. Where this isn’t the case, a period of negative rental income returns may be sufficient to make their investment unviable — even if total returns remain strong. Post-tax returns on rental income, therefore, are a key analytic for understanding the risk of landlord disinvestment.

The composition of the small minority of landlords in each year making a negative return on their rental income is primarily shaped by financing structure and interest rate exposure. During periods of elevated interest rates, highly leveraged landlords can be more or less shielded from elevated borrowing costs, depending on when they need to remortgage. In 2018 and 2021, there were significantly higher rates of landlords on fixed rate mortgages making a post-tax loss on their rental income compared to mortgagors on variable rates — reflecting the fact that those on variable rates benefited from the extremely low interest rates during this period.

In 2024, the trend reverses: a significantly higher rate of variable rate mortgagors are making a negative return on their rental income compared to those on fixed rates. Post-tax returns on rental income for unmortgaged landlords, by contrast, are strong across the time points; the proportion of unmortgaged landlords making a loss on their rental income is zero in all years.

But the changes did come with some other, undesirable consequences. One is that they reduced the ‘neutrality’ of the tax system, giving a tax advantage to corporate landlords who are still able to deduct their full interest costs from their tax liability. This creates an incentive for individual landlords to incorporate, and our previous analysis estimated that this tax advantage for corporate landlords could be worth over £800 million (Elliott and Baxter, 2025).

Another problem is the interaction between the Section 24 changes and interest rate volatility. The interest rate environment in which landlords are now operating is very different to when the reforms were introduced; borrowing costs have risen sharply in the past few years. When interest rates increase significantly, the share of rental income spent on interest payments for mortgaged landlords increases dramatically — which makes the impact of less favourable tax deductibility for mortgage interest more pronounced relative to overall profitability. At the extremes, landlords can be pushed into making a loss on their rental income by tax — as illustrated in the charts below.¹¹

In 2024, almost 20% of individual landlords on variable-rate mortgages, and 10% of those on fixed rates, were making a post-tax loss on their rental income.¹² We estimate that by 2030, as more legacy fixed rate mortgages come to an end and landlords have to remortgage at a higher rate, the proportion of mortgaged landlords making a post-tax loss on their rental income will nearly double, from 11.2% to 20.3%, driven for the most part by feed through of higher interest rates.¹³

Changing the way rental income is taxed would make tax system fairer and less distortive

Though the changes to Section 24 were an effective way of using a tax lever to dampen investor demand and put residential buyers in a stronger position, they are not effective at capturing a fair share of the profits being made by landlords on their rental income. Post-Section 24 changes, mortgaged landlords are taxed on their revenue (their rental income after accounting for operating costs, but before they’ve paid their financing costs) rather than their profit.

Unencumbered landlords, by contrast, do very well out of the current tax treatment of rental income: they pay a lower rate of income tax on the considerable profits they have left over after covering their operating costs, than the level of tax paid on income from work.¹⁴ The result is that the tax revenue raised from rental income as a whole comes disproportionately from landlords who may be making minimal profits on their rental income, instead of those who are making the highest returns — this is something the Government should seek to rectify.

Making changes to the tax treatment of rental income is sensible in its own right and has broad support across the political spectrum. A recent report from CenTax recommends that to address the irrationality in the way rental income is taxed, the Section 24 changes should be reversed, to allow mortgaged landlords to once again deduct their full mortgaged interest costs from their taxable income. At the same time, they recommend that NICs be applied to rental income, to equalise the tax treatment of rental income with the tax treatment of earnings.

These changes would remove the distorting impact that arises from differential tax treatment of individual and incorporated landlords, where landlords currently have an incentive to incorporate to avoid paying tax on their mortgage interest payments. They would also make the tax system fairer by applying the same tax treatment to income from different sources.

When we look at a more fine-grained level at how these changes would shape the distribution of returns made by landlords, moreover, it becomes clear there would be additional advantages.

Tax changes would shield landlords most exposed to interest rate shock, raising revenue from those making higher profits

If we take the distribution of landlord post-tax returns on rental income from 2024, restore the ability of mortgaged landlords to deduct their full interest costs from their tax liability and instead apply full NICs (employer and employee) to all taxable rental income, the share of mortgaged landlords making a loss on their rental income drops from 11.2% to 7.6%.¹⁵

At the same time, it compresses the returns of those at the top of the distribution, reducing the number of landlords making >3% returns on their rental income. The overall tax take from rental income in 2024 under this scenario would have been 11% higher, coming from unencumbered landlords paying more tax and mortgaged landlords paying slightly less.¹⁶

When we project ahead and look at the impacts our proposed tax changes would have in 2030, these effects become more pronounced. As we saw above, the gradual pass-through of higher interest rates over the next few years will mean that under current tax arrangements, the proportion of mortgaged landlords making a loss on their rental income will be 20.3% by 2030. But if we take the same projection, reinstate full mortgage interest deductibility and apply NICs to rental income, the proportion making a loss would be reduced to around 13%.

The overall tax take would be higher in 2030 with the tax changes than it is projected to be under current tax arrangements — though the increase would be smaller than the increase to tax receipts that these changes would have generated if they had been applied in 2024. The reason for this is that as more mortgaged landlords roll on to higher interest rates, their taxable income given full interest deductibility will decrease.

Comparing current tax arrangements to our proposed changes, those without mortgages would see the share of their gross rental income spent on tax in 2030 increase from just over 26% to around 33%. Mortgagors, by contrast, would pay a lower proportion of gross rental income on tax compared to under the current tax arrangements (around 18%, down from 22%).

Unencumbered landlords would still keep a much larger share of their gross rental income as profit under the tax-change scenario compared to mortgaged landlords (43% compared to 18% for mortgagors), which reflects the considerably lower costs faced by unencumbered landlords. But the tax changes capture a greater share of their profits, bringing them down from their current level of around 50% profit on average.

Taken together, our modelling shows that reforming the tax arrangements for rental income would have the effect of redistributing the tax burden on landlords in a way that both protects those with the thinnest margins on the rental income side of their business (mortgagors exposed to higher interest rates) and generates more tax income from those making the largest profits from rent. In other words, it acts both on the risk exposure and high profit ends of the distribution.

An additional benefit of our proposed changes is that they create a taxation regime that is better equipped to accommodate changes in the fortunes of mortgaged landlords that may arise in the future. If interest rates do come down significantly, increasing the profit margins of leveraged landlords as a smaller share of their rental income is needed to cover their borrowing costs, they will pay a higher rate of tax on these increased profits through the application of NICs. So there is an important future-proofing aspect to our proposals: they build a system that will effectively capture a higher and fairer share of the profits landlords make, regardless of interest-rate volatility.                                                

Redistributing tax across landlord types smooths impact of constrained rental incomes, with rent control

Our modelling also indicates that reforming the taxation of rental income would also mitigate some of the impacts that introducing rent controls could have on mortgaged landlord viability. We have shown how our proposed tax changes would go a significant way towards shielding mortgaged landlords from the worse interest rate situation in 2030. Crucially, this shielding effect holds up even when we model rent control policies constraining rental incomes between now and 2030.

If our proposed changes to the tax system were adopted today, and a soft form of rent control that limited annual in-tenancy rent increases at annual wage growth was also introduced, our modelling shows there would be no discernible impact on the proportion of landlords making a loss on their rental income.

With a considerably more ambitious form of rent control that capped rent increases both with and between tenancies (capping rents in-tenancy at CPI and capping between tenancy increases to CPI + 2%), the proportion of mortgaged landlords making a loss would increase slightly by 2030 compared to a scenario of tax reform and no rent control(from roughly 13% to 17.3%).

But crucially, this is still lower than the proportion that will be making a loss by 2030 with no rent control, if current tax arrangements are maintained (20.3%). This remains true even with a version of rent control that fully froze rents during tenancies in addition to controlling them at CPI + 2% between tenancies — showing that our proposed tax reforms would go a long way towards shielding the sector from the impact of constrained rental incomes.

Rent control policy needs careful calibration — but plays effective role alongside other policy levers

Our analysis in this report has shown both that rent caps are viable given the level of returns landlords typically make on their investments, and that the transition to introducing them can be eased through reforms to the taxation of rental income. However, there are many important questions about the precise policy design of an effective system of rent control that have been beyond the scope of this report, but which need careful consideration.

Most obviously, there is the core question of rent control design — that is, the level at which caps on rent increases should be set. Our modelling found that rent control regimes that only limit increases within tenancies, unsurprisingly, are much less impactful in terms of making rents more affordable relative to incomes over time. Capping in-tenancy rent increases at CPI with no controls between tenancies, for example, would only constrain rent growth to around the lower-baseline forecast for rent growth over the next 5 years without any rent control.

For this reason, the indicative design we have used in most of our modelling in this report — of rent increases capped at CPI within tenancy and CPI + 2% between tenancies — caps rent increases between tenancies as well, though at a higher level than within tenancies.

Having a more generous cap between tenancies maintains a level of price signalling in the market, which is important for understanding where targeted action on supply may be needed in very high-demand areas. It may also be advantageous to build into a rent control regime the ability to flex cap levels depending on changing factors in local areas — for instance, acute increases in demand, or local supply elasticities.

Another key issue is the impact that rent controls could have on the rate of housebuilding. This is certainly a risk we should be aware of; driving up the rate of construction is necessary and important. This points to a need to design rent caps in a way that is mindful of this. We are confident that this is possible as credible international academic evidence showing that cap-based models — particularly those with carve outs for new-build homes — have little to no impact on wider rates of housebuilding (Kholodilin and Kohl, 2023).

Our future work will consider the role of exemptions for new supply and major renovation work, as well as what supporting policies are needed alongside rent controls to tackle the underlying structural causes of high rents — for example, action to ease the planning burdens facing new supply, targeted investment in social and affordable housing, and targeted acquisitions from exiting or distressed landlords.

We also need to consider the interaction between rent control and devolution — for example, whether it would be desirable to give Mayors the ability to set caps at lower or higher levels depending on their local housing market context. This could be done in return for declaring rent pressure zones, in which greater efforts are undertaken to increase housing supply (such as by increasing the ease with which areas are densified or via targeted funding for social and affordable housebuilding).