Rented sector reform: Housing (Scotland) Bill: business and regulatory impact assessment

Annex B. Summary of Literature on Rent Controls

Rent controls have been instituted in various forms internationally, including temporarily here in Scotland under the 2022 Act. However, apart from this brief emergency legislation, there is little comparative contemporary evidence from the UK to draw from to help understand the likely impacts of longer-term regulation. Therefore, it is necessary to look more widely to explore potential outcomes for key stakeholders within the housing system. As summarised in a recent review, a longitudinal study of rental market regulations by Kholodilin^[147] found that there were three key features of contemporary rent controls. These are “rules regarding rents for new rent contracts (i.e. the initial rent), rules concerning the updating of rents within a contract (i.e. rent increases) and, exceptions which either exempt or provide for enhanced regulations.”^[148] A scoping review of rent controls in OECD countries^[149] described a fourth feature, namely how extensive the coverage was across the PRS.

Broadly speaking within the literature, those rent controls that include a nominal long-term ‘freeze’ to any increase are referred to as ‘first generation’ rent controls, those that allow capped increases but where the cap applies to both sitting and new tenancies are termed ‘second generation’ rent controls, and those in which rental increases are capped only for sitting tenants are referred to as ‘third generation’ rent controls, although there is some variation in how individual studies differentiate between second and third generation.

A recent briefing paper on international experiences of rent control by CaCHE found that international experiences of rent controls are nuanced and context dependent making comparative analysis difficult, although there are lessons to draw from existing experiences.^[150] A review by Whitehead and Williams in 2018^[151] found that the contextual history of regulation is important in determining “how controls evolve and their impact on both landlords and tenants”. So international experiences of rent controls contain not only different configurations of regulatory ‘features’, but also the way they operate and their outcomes are dependent on the unique local context, and that context may include historical regulations, cultural attitudes towards renting, and the relative size and dominance of the PRS, among other factors such as the existing and emerging regulatory landscape.^[152]

In a briefing paper prepared for the Scottish Government,^[153] researchers from CaCHE outline that, globally speaking, later-generation rent regulations can be found in Canada, Europe and few other national contexts; many other countries have no controls; while a number of countries including Africa, South America, India and some parts of Europe continue to operate first-generation controls. It is also worth noting that not all rent controls are instituted at a national level, as they are often found at a regional or city/town level.

There are ideological assumptions that underpin different types of studies exploring the impacts of rent controls. For example, at one end there is the ‘free market’ perspective which sees housing as a free-market commodity, in which rent controls may even be viewed as form of ‘theft’ from property owners/landlords. At the other end of the spectrum, studies may take a position that rented housing should be viewed primarily as a home for the tenant who has a right to somewhere they can live and flourish.^[154] The (sometimes unstated) ideological position taken by these studies may also influence how ‘success’ is measured.

In a systematic review exploring the international literature on rent controls from the last 20 years undertaken by CACHE,^[155] the majority of economic studies were found to be heavily dependent on particular assumptions about the system they were evaluating, and the authors concluded from the literature that “rent control can have a negative, positive or neutral effect on one or more aspect of housing and related markets, depending on your modelling assumptions”. There is a general consensus among mainstream economists that ‘first-generation rent controls’ (those rent controls which constitute a long-term ‘nominal freeze’ on any rental increase) have the potential for negative system impacts, most notably incentivising landlords to leave the sector.^[156] However, other evidence reviews have pointed out that outside of mainstream economics, opinions are not as consistent,^[157] and the evidence base for later-generation rent regulation (i.e. where rent adjustments are capped in a more responsive way in line with other market conditions) is more mixed.^[158]

In general, there is evidence to suggest that there are benefits to these types of later-generation rent controls to sitting tenants; however, without adequate consideration there is the potential for other groups, such as new tenants, to be impacted negatively via price increases occurring between tenancies, or by limitations on property availability in the ‘medium term’.^[159] Additionally, there is evidence from econometric studies to suggest that other unintended consequences have been observed, such as: low-income tenants who are as much in need as those who end up in rent-controlled properties may be excluded and end up in poorer-quality accommodation; tenants in rent-controlled properties may have to make informal ‘side payments’ to secure properties; or tenants in neighbouring uncontrolled areas may face rising rents due to control-induced shortages.^[160] However, there is also evidence that for tenants, rent controls can lead to stabilised rents and better affordability. In a recent landlord and tenant engagement questionnaire on the detail of proposals for future rented sector reform, there was a very clear consensus from PRS tenants and tenant organisations that they felt that rent control proposals would bring the most general and financial benefits to tenants of all the proposed rented sector changes.^[161]

Evidence of potential negative impacts of rent controls identified include the potential for lower investment by landlords into areas like property maintenance, and the potential shrinking of the sector via landlords exiting the market. A recent policy briefing paper on rent controls prepared by CaCHE for the Scottish Government said:

“Rent control leading to reduced labour movement is a common part of the academic economic evidence but there is less clarity over its impact on property maintenance (and we know that well-designed non-price regulation may be able to weaken negative maintenance effects, e.g. through the right to repair). We think that the traditional view about rent controls reducing supply is less convincing, in part because it depends on the ‘bite’ of the control relative to expected returns for landlords, but also because it makes simple assumptions about what suppliers do in terms of exit from the market, who they sell to, or whether they are less responsive to controlled rents (there is certainly evidence that where the structure of landlordism in an area comprises more small scale landlords they can respond quite differently to larger and better capitalised landlords).”^[162]

There is limited robust evidence about the characteristics of landlords in Scotland but there is some evidence to suggest they are likely to be wealthier and have larger incomes than average in the population, and certainly than average for PRS tenants.^[163] A 2019 review of housing wealth inequalities in Scotland did find, however, that there is diversity within the socioeconomic profiles of landlords.^[164] This suggests that some landlords may be exposed to cost-of-living pressures, even if not to the same extent as tenants on average. In recent research, landlords in Scotland have expressed concerns about the impact of rent controls on their profitability, as well as about the wider context of what they view as increasingly restrictive sector legislation.^[165] Similarly, in a Scottish Government questionnaire on the detail of proposals for future rented sector reform,^[166] among those PRS landlords and landlord organisations who responded, rent controls were perceived as being potentially the least financially beneficial of all rented sector proposals for landlords.

Other potential unintended consequences that have been noted in previous rent control regimes have included restrictions on mobility for tenants and potential costs associated with robustly instituting and maintaining rent controls.^[167] However, as a 2022 study looking at international evidence on rent stabilisation including the recent experience in Ireland^[168] points out:

“[m]ore recent regulatory regimes have looked to offset these side-effects with better calibration including the usage of exemptions for new supply, allowances for maintenance investment and other mechanisms similar to the policy calibration found in the current Irish policies. The impact on supply is now likely more context specific and likely to depend on the specific regulatory set up in the operating jurisdiction, as well as the other market specificities in each country.”

The authors of this paper recommend the use of targeted exceptions to help avoid negative consequences in the sector.

As mentioned above, the CaCHE review of rent control evidence pointed out that potential impacts of rent control measures are highly context dependent. This is because different rent control regimes include a variety of regulatory ‘features’ as well as underpinning structures related to enforcement and tenant security, which are intended to engage with local circumstances, achieve different regulatory aims. , and curb potential undesired impacts on different stakeholders. Such heterogeneity, makes it difficult to anticipate the full effect of new price regulation on a rental system, particularly when combined with other legislative changes (such as are outlined in this Bill). The CaCHE review concluded that there was a “need to be cautious regarding the appropriateness of international comparisons (even within welfare regime clusters) and the relevance of contemporary evidence for the UK.”^[169] They suggest that other aspects of the rented sector regulatory system, such as tenancy length and how regulation is enforced, form a key part of how rent controls function; however, these are not always sufficiently accounted for within the literature.

In a 2018 review of international rent control evidence, Whitehead and Williams^[170] summarise the goal of ‘second-generation’ rent controls’ (or those that include control of rent increases both within and between tenancy controls) as being:

“to allow some mitigation of cost increases for landlords and thus reduce the incentives for them to under-maintain their properties, while retaining some limits on the size of rent increases in order to help tenants in markets typically characterised by shortage.”

A 2019 review of rent stabilisation evidence by the same authors^[171] found that allowing for between-tenancy increases (sometimes called ‘third-generation rent controls’) could help to provide more certainty for both tenants and landlord during the lease period, but could also take account of market pressures at the start of the new tenancy. However this approach does not aim to ‘hold down’ rents but to reflect longer-term trends, and they also argued for the importance of strong enforcement in this type of regime. Describing this type of regime in their earlier review^[172] they state:

“The impact of higher initial rents varies according to how long a tenant actually remains in a tenancy. Tenants who stay longer than the average will end up paying ‘too little’ in rent and those who stay for a shorter period than average will pay ‘too much’. A benefit for landlords is that controlled rents and rent increases tend to reduce turnover.”

The CaCHE review^[173] also noted that a counter-augment that has been advanced against rent stabilisation policies that allow landlords to reset the rent is that they:

“don’t provide significant improvements in affordability for those in regulated properties, rather they simply reprofile and frontload rental payments: landlords know that they aren’t going to be able to increase the rent during a tenancy so the incentive is to maximise the initial rent set (Arnott, 2003). The extent to which landlords will be able to adopt this strategy will depend in part on local market conditions. Where it can be operated it means that mobile households will be disadvantaged relative to households who stay in the same tenancy for longer periods.”

The possibility that rent controls which apply only to sitting tenants might incentivise landlords to increase rents on new lets above the rate they would have otherwise has been raised in the context of emergency rent controls in Scotland under the 2022 Act, both in public commentary ^[174] and in feedback from landlords in stakeholder engagement.^[175] However, there has not been a robust study of this effect, in part because of data limitations on cross-country comparisons, mainly due to the differential availability of data on existing rents across UK countries.^[176] Looking at Zoopla data on average new-let rents only,^[177] there is some evidence that new-let rents have increased at a faster rate in Scotland than other parts of the UK over the period October 2022 to January 2024,^[178] with a cumulative increase of 15.3%. However, increases in new-let rents in other parts of the UK have been nearly as high, with average rents rising by 13.7% in Wales, and by 13.5% across the UK as whole. Furthermore, sitting tenants in other parts of the UK have not been protected by a rent cap. Given that the rent cap in Scotland for sitting tenants was 0% and then 3% over this period,^[179] i.e. significantly below the growth in new-let rents across the UK, and that around two-thirds of tenants in Scotland stay in their rental for more than a year,^[180] the higher growth in new-let rents is potentially consistent with the average increase in rents across all tenancies (both existing and new tenancies) being lower in Scotland than the rest of the UK.

A further way in which rent controls which apply to sitting tenants only could result in perverse incentives, which has also been the subject of public commentary in the context of the emergency rent cap in Scotland,^[181] is the case of joint tenancy arrangements. Upon the departure of one tenant, the remaining co-tenants may be required to sign a new tenancy lease, which means that they would no longer be protected by rent cap measures which apply to existing tenancies.

The importance of understanding contextual interactions with regulatory detail has been noted in various studies because of the potential to create such perverse incentives. A 2023 scoping review of rent control evidence from OECD countries,^[182] in addition to noting the two examples from Scotland discussed above, warned that perverse incentives that could lead to unintended consequences were found in other countries’ rent control systems. These included open exceptions for refurbishment and remodelling (which could incentivise landlords to prematurely carry out works so as to become unregulated), and allowing landlords to ‘bank’ rent increases (i.e. saving up rent increases over multiple years then charging them all at once) which could undermine security of tenure for tenants.

The CaCHE briefing paper on rent controls^[183] found that a review of current international examples of rent controls in operation include levels of non-compliance that emphasise the importance of strong enforcement. They also cite evidence that suggests that tenure constraints (i.e. disallowing a shift from the PRS to short-term lets) may be necessary to prevent tenure shift in rent-controlled areas. The authors note that robust data and research evidence related to the PRS are both vitally important for the “credibility, durability and, in the case of data, for operational policy development”.

Finally, when considering international evidence on rent regulations, Whitehead and Williams^[184] conclude that, in practice many landlords do not attempt to maximise rents in favour of keeping longer term tenancies, which allows them to benefit from rent stabilisation:

“[international examples] make it clear that regulation is not inherently good for tenants and bad for landlords. By providing a clear framework it can be a win-win situation, at least for mainstream landlords and tenants alike. It can address market failures for both tenants and landlords, while reducing risks for both. It can give landlords more consistent rent rises and reduce financing costs and risks for landlords, while providing better security of tenants. However, achieving a better solution depends on the detail of each initiative as well as the institutional and market environment.”