The renaissance of rent control: Magic bullet or market distortion?

 

The related notion of inequality refers to rent control exaggerating or reducing already existing economic inequality between social classes and ethnic groups. In situations of misallocation, rent control has the potential to exacerbate inequality by disproportionately favouring more affluent households. Nevertheless, considering that lower-income households are more likely to be tenants, whereas higher-income households tend to be homeowners and landlords, rent control might actually contribute to reducing inequality. By setting rent limits, it can effectively lower housing expenses for lower-income households while diminishing rental income for higher-income households.

 

Rent control can also affect the socio-economic and ethnic composition of communities. Segregation arises when individuals are geographically separated based on factors such as race or social status. Rising market rents can drive out the poorest households, thus reducing the income and racial heterogeneity of affected neighborhoods and increasing the segregation. From a theoretical standpoint, rent control possesses the potential to both heighten and mitigate segregation. On the one hand, by generating a surplus of demand in comparison to supply, rent control can lead to dwellings being assigned based on landlord preferences, which might inadvertently foster segregation. On the other hand, by reducing rental burden, rent control can enable lower-income households to reside in more attractive neighborhoods, thereby lessening social segregation.

 

 

Connected to segregation is also the aspect of neighbourhood quality perception. When current residents have a preference for residing alongside individuals of similar social status and ethnic background, any influx of diverse individuals might lead them to perceive a decline in their neighbourhood’s quality. As observed, rent control holds the potential to influence segregation in either a positive or negative direction, consequently impacting how the neighbourhood’s quality is perceived.

 

Theoretical arguments 

The theory of rent control implies that it can reduce residential mobility, which measures how long tenant households stay in the same place: the longer this time, the lower the mobility. Under rent control, people occupying dwellings with low fixed rents have fewer incentives to leave. This can have some negative labor-market implications.

The effect on homelessness means that rent control could possibly lead to either fewer or more people living on the streets. On the one hand, rent control could theoretically reduce the rental burden of the lower-income households and, thus, reduce the probability of landlords evicting their tenants of controlled dwellings for non-payment of rent. It will not extend its protection to the fragile households living in uncontrolled dwellings, though. On the other hand, the reduction in the supply of rental dwellings due to rent control can result in some people having a tough time when looking for an available dwelling and, hence, increase homelessness.

 

Net welfare denotes the difference between benefits and costs of rent control. Typically, in the literature, the benefits include lower rental burden for tenants in regulated dwellings, while costs comprise an increased rental burden for tenants in unregulated dwellings and decreased revenues for landlords. Sometimes, dead-weight losses that arise due to higher search costs borne by tenants are also considered. Ideally, any policy’s net welfare change for the entire society resulting from its implementation should be positive, otherwise the policy does not make sense. Moreover, it is imperative to compute both short-term and long-term welfare consequences. A policy that improves the net welfare in the short run, but erodes it over the long run, is essentially useless and its adoption can only be explained by the myopia of policy makers or by the electoral politics.

 

 

Tax base effects describe changes in tax revenues caused by the implementation of rent control. This impact can materialize through two primary mechanisms. First, the imposition of rent limits diminishes landlords’ earnings, thus, reducing the state’s taxation revenue derived from their profits. Second, rent control has the potential to diminish the value of properties under its regulation, consequently leading to a reduction in the revenue obtained from property taxes. These tax effects should be taken into account in the calculation of the overall net welfare resulting from rent control.

 

Rent control can possibly affect inflation. Indeed, rent index is the largest component of the consumer price index. Therefore, by imposing caps on rent increases the government could decelerate overall price growth.

 

The literature also investigates the impact of rent control on evictions of tenants by the landlords. It is assumed that, in the absence of protection from eviction, the landlords are more likely to evict tenants. By doing so they are able to set higher rents for the new tenants.

 

Here, I elaborate on the impacts that do not fall under the categories mentioned earlier. Commuting times denote the time individuals require to journey to their workplace and return home. These periods can extend due to decreased residential mobility: individuals often opt to remain in their existing regulated residences rather than relocating nearer to their workplaces, leading to increased commuting time from their homes to their jobs. 

The marriage effect refers to the potential impact of rent control on the demographic decisions made by the people. For instance, a lack of rental housing can cause young people to postpone their marriage, since many cultures often require them to live separately from their parents. Finally, side payments represent various unofficial payments, such as key money, that can be fostered by the introduction of rent control.

 

Empirical evidence

Overall, I could find 206 works on the effects of rent control, among them 112 empirical published studies. The latter are the main focus of this study. This is perhaps the most comprehensive review of the rent control literature encompassing the period between 1967 and 2023.

 

These are the rent control effects that occupy the most prominent places in the literature with more than 6 published studies devoted to it. The left bar shows the number of studies that found a negative (positive) effect of rent control on the corresponding variable. The height of the bar in the middle corresponds to the number of studies that did not find a statistically significant effect of rent control on the variable. For the sake of completeness, along with the number of published studies (greenish shading) I also show the number of unpublished studies (grey shading).

 

The most prominent effect of rent control is, unsurprisingly, its impact on controlled rents; that is, on rents paid by the tenants of those dwellings subject to rent control. The picture is rather unambiguous: 36 out of 41 published studies (53 out of 60 published and unpublished studies) point to a statistically significant negative effect. Thus, rent control is quite effective in capping rents. 

 

By contrast, according to the studies examined here, as a rule, rent control leads to higher rents for uncontrolled dwellings. The imposition of rent ceilings amplifies the shortage of housing. Therefore, the waiting queues become longer and would-be tenants must spend more time looking for a dwelling. If they are impatient or have no place to stay (e.g., in the houses of their friends or relatives) while looking for their own dwelling, they turn to the segment that is not subject to regulations. The demand for unregulated housing increases and so do the rents. Only one published study — Bonneval et al. (2021) — finds no statistically significant impact of rent control on uncontrolled rents. The study uses real estate property manager’s accounting books data for Lyon between 1890 and 1968 and applies difference-in-differences regression for panel data.

 

The estimated effects of rent control on rental prices exhibit considerable variation across diverse studies. For controlled rents the range is between -57 % and -1 %, whereas for uncontrolled rents it is between -2 % and 14.8 %. The reason for such a variation lies in the different research setups. Certain studies focus on immediate, short-term effects, while others delve into the cumulative, long-term consequences of rent control measures. The average effect of rent control on controlled rents is -9.4 %, while that on uncontrolled rents is 4.8 %.

 

Unfortunately, only based on these results it is virtually impossible to evaluate the overall effect of rent control on housing rents. To do this, a careful analysis of the distribution of housing units across the controlled and uncontrolled sectors is needed. This distribution will depend on a number of factors, including the design of rent control policy. Moreover, the price effects can die out or increase over time. This evolution can be different for controlled and uncontrolled dwellings.

 

The impact on residential mobility appears to be quite clear: nearly all studies indicate a negative effect of rent control on mobility. Two potential reasons for this phenomenon are put forward. Initially, residents living in controlled dwellings have limited motivation to relocate. They possess concerns that finding a residence of similar quality at such a low rental cost might be challenging. This situation can yield unfavourable outcomes for the job market, as reduced residential mobility translates to less adaptable responses to shifts in the labour market. When economic conditions worsen in their city, tenants in controlled dwellings are less inclined to move to areas with more promising employment prospects. Secondly, diminished residential mobility could be attributed to heightened tenure stability. Through rent regulation, this policy alleviates the financial strain of tenant households, consequently reducing the likelihood of eviction. Additionally, rent control legislation is often adopted simultaneously with rules protecting tenants from arbitrary removals. As a result, tenants remain in their residences for longer time, thereby boosting their satisfaction. 

 

Likewise, the influence of rent control on new residential construction and supply seems to be similar. Approximately two-thirds of the studies indicate a negative impact, while several studies discover no statistically significant effect whatsoever. Two potential reasons underlie this variability. Firstly, variations in the design of rent control policies can matter. For example, newly constructed housing could be exempted from control, thus remaining unaffected by rent control regulations. Secondly, the choice of the dependent variable can also affect results of the analysis. Rent control can influence the construction of rental dwellings while leaving owner-occupied properties untouched; in fact, the quantity of owner-occupied dwellings might even increase, thereby compensating for any decline in the number of completed rental units. However, it is common to analyse the overall construction impact, often due to limitations in data availability. Furthermore, if private construction experiences a decline, governmental intervention becomes a possibility. This could involve the construction of public housing or financial support for private investors engaged in social housing development. Consequently, the total number of completed dwellings can remain steady or even rise, potentially leading to a misinterpretation of rent control’s impact as beneficial.

 

The published studies are almost unanimous with respect to the impact of rent control on the quality of housing. All studies, except for Gilderbloom (1986) and Gilderbloom and Markham (1996), indicate that rent control leads to a deterioration in the quality of those dwellings subject to regulations. The landlords, whose revenues are eroded by rent control, have reduced incentives to invest in maintenance and refurbishment, thus they let their properties wear out until the real value of the dwellings decreases and becomes equal to the low real rent. According to Gilderbloom (1986) and Gilderbloom and Markham (1996), moderate rent control does not impact housing quality. In a theoretical study, Lind (2015) shows that quality of housing will not suffer if the allowed rent increases are pegged to improvements made to the dwellings by landlords. When only unpublished papers are considered, the effects are mixed: half find negative, the other half no effects.

 

 

In the case of homeownership effects, the picture is a bit less clear cut: there are multiple studies pointing in different directions. In particular, the relationship appears to be blurred when only unpublished studies are considered. Nevertheless, the majority of studies predict an increase in the homeownership rate due to rent control. This can be explained by the desire of landlords to get rid of those properties that bring them insufficient rental revenues. Therefore, the landlords sell their dwellings or convert them into condominium ownership. By contrast, Gyourko and Linneman (1989), Lauridsen et al. (2009), and Bourassa and Hoesli (2010) find a negative effect of rent control on homeownership, explaining it from the perspective of tenants in controlled dwellings: they are less inclined to become owners, given their protected position. These studies are heterogeneous in geographical terms: Denmark, Switzerland, and the USA. Only one published study finds no statistically significant effects of rent control on homeownership (Werczberger 1997). 

 

Thus, empirical investigations do substantiate the hypotheses derived from the theoretical literature regarding the effects of rent control. While rent control does succeed in reducing rents within controlled dwellings, it also generates several adverse consequences that work against its intended purpose.

 

In this study, I examine a wide range of empirical studies on rent control published in referred journals between 1967 and 2023. I conclude that, although rent control appears to be very effective in achieving lower rents for families in controlled units, its primary goal, it also results in a number of undesired effects, including, among others, higher rents for uncontrolled units, lower mobility and reduced residential construction. These unintended effects counteract the desired effect, thus, diminishing the net benefit of rent control. Therefore, the overall impact of rent control policy on the welfare of society is not clear.

 

The analysis is further complicated by the fact that rent control is not adopted in a vacuum. Simultaneously, other housing policies — such as the protection of tenants from eviction, housing rationing, housing allowances, and stimulation of residential construction (Kholodilin 2017; Kholodilin 2020; Kholodilin et al., 2021) — are implemented. Further, banking, climate, and fiscal policies can also affect the results of rent control regulations.

 

Nevertheless, at least ideally, policy makers should take into account the multitude of these effects and their interactions when designing an optimal governmental policy. Researchers would readily support this by providing their expertise.

 

Konstantin A. Kholodilin is a senior researcher at the DIW Berlin in Germany. He specialises in real-estate economy, spatial econometrics and time series analysis of business cycles. This is an excerpt from “Rent control effects through the lens of empirical research: An almost complete review of the literature” published in the Journal of Housing Economics, Volume 63, 2024. This paper is under a Creative Commons license and has been edited for style and length. The full article with citations can be read on Science Direct