Who benefits from rent control? Evidence from San Francisco

Main article

Steadily rising housing rents in many large and productive US cities have brought the issue of affordable housing to the forefront of policy debate and reignited discussions about expanding or enacting rent control provisions. While the details of rent control regulations vary across places, they generally regulate rent increases and place restrictions on evictions.

In 2019, Oregon and California became the first states to pass statewide rent control. Lawmakers in other states, including Colorado and Illinois, are considering repealing laws that limit cities’ abilities to pass or expand rent control. Rent control is already extremely popular around the San Francisco Bay Area: nine cities already impose rent control regulations, two of which recently passed rent control laws through majority votes on the 2016 ballot.

Potential consequences of rent control

A substantial body of economic research has warned about the potential negative consequences of limiting rent increases below market rates. For example, rent control could lead to excessive consumption of housing by tenants because rents are so low.

It could also lead to misallocation of housing units to households, with aging ‘empty nesters’ living in large apartments and growing families living in studios since moving away from a rent-controlled unit forces tenants to give up their below market rents. Other consequences of rent control include negative spillovers onto neighboring housing and neglect of required maintenance.

On the other hand, in the absence of rent control, many tenants are unable to insure themselves against rent increases. Of course, individuals who have little connection to any specific area may be able to insure themselves easily against local rental price appreciation by simply moving to a cheaper location.

But if long-term tenants have developed attachments to specific neighborhoods, such as networks of friends and family, proximity to their jobs, or proximity to the schools of their children, then they face large risks from rent appreciation. A variety of affordable housing advocates have argued that many tenants greatly value such insurance and that rent control can effectively provide it.

An experiment in rent control expansion

Despite the policy interest, due to a lack of detailed data and natural experiments, there is little well identified empirical evidence evaluating how introducing local rent controls affects tenants, landlords, and the broader housing market. Our research brings to bear new micro data and uses quasi-experimental variation in the assignment of rent control to fill this gap.

We make use of an unexpected 1994 law change that suddenly applied rent control to a subset of San Francisco buildings and their tenants, based on the year in which each building was built. The law left very similar buildings and tenants without rent control.

We find that tenants covered by rent control place a substantial value on the benefit, as revealed by their choice to remain in their apartments longer than those without rent control. Indeed, we find that the vast majority of those with the incentive to remain in their rent-controlled apartment would have been displaced from San Francisco had they not been covered.

But landlords of properties affected by the law change respond over the longer term by substituting to other types of real estate, in particular by converting to condos and redeveloping buildings so as to exempt them from rent control.

In the long run, landlords’ substitution toward owner-occupied and newly constructed rental housing not only lowered the supply of rental housing in the city, but also shifted the city’s housing supply towards less affordable types of housing that are likely to cater to the tastes of higher income individuals. Ultimately, these endogenous shifts in the housing supply seem likely to have driven up citywide rents, damaging housing affordability for future tenants and counteracting the stated claims of the law.

Rent control in San Francisco

In 1979, San Francisco imposed rent control on all standing buildings with five or more apartments. While all large buildings built as of 1979 would now be rent-controlled, new construction was exempt from the law since legislators did not want to discourage new development.

In addition, smaller multi-family buildings were exempt from rent control since they were viewed as more ‘mom and pop’ ventures, and did not have market power over rents. But this small multi-family exemption was lifted through a 1994 San Francisco ballot initiative. Proponents of this change in the law argued that small multi-family housing was now primarily owned by large businesses and should face the same rent control restrictions of large multi-family housing.

Since the initial 1979 rent control law only affected properties built in 1979 and earlier, the removal of the small multi-family exemption also only affected properties built in 1979 and earlier. This led to a quasi-experimental rent control expansion in 1994 based on whether the small multi-family housing was built before or after 1980.

To examine the effects of rent control on tenant migration and neighborhood choices, we make use of new panel data that provides address-level migration decisions and housing characteristics for the majority of adults living in San Francisco in the early 1990s.

This allows us to define our ‘treatment group’ as tenants who lived in small multi-family apartment buildings built before 1980 and our ‘control group’ as tenants living in small multi-family housing built between 1980 and 1990. Using our data, we can follow each of these groups over time up to the present, regardless of where they migrate.

Beneficiaries of rent control

We find that between five and ten years after the law change, the beneficiaries of rent control are, on average, 3.5 percentage points more likely to remain at their 1994 address relative to the control group. Since only 18% of the control group still remained at their 1994 address for that long, this estimate represents a 19.4% increase in not moving (3.5/18) relative to the control group.

We further find that the beneficiaries are 4.5 percentage points more likely to remain in San Francisco relative to the control group, indicating that a large share of the tenants who remained at their 1994 address due to rent control would have left the city had they not been covered by rent control. This would be likely to be viewed as a desirable outcome by rent control advocates.

We next analyze treatment effect heterogeneity along a number of dimensions. We first find that our estimated effects are significantly stronger among older households and households that have already spent a number of years at their address prior to treatment. This is consistent with the idea that both of these populations are less likely to experience personal shocks requiring them to change residence and thus, are better able to take advantage of the potential savings offered by rent control.

We then examine whether the effects that we estimate vary across racial groups. We do not directly observe race in our data, so we use an imputation procedure based on tenants’ names and addresses. We find rent control has an especially large impact on preventing the displacement of racial minorities from San Francisco, suggesting that rent control helps to foster the racial diversity of the city, at least among the initial cohort of tenants covered by the law.

Finally, we analyze whether rent control enables tenants to live in neighborhoods with better amenities. One might expect neighborhoods with the largest increases in market prices and amenities to be ones where tenants would remain in their rent-controlled apartments the longest, since their outside options in the neighborhood would be especially expensive.

But for these same reasons, landlords in these high-rent, high-amenity neighborhoods would have large incentives to remove tenants. In practice, landlords have a number of legal means to remove their tenants, including owner move-in eviction, Ellis Act eviction, or monetary compensation.

Landlords may also engage in various tactics, such as tardy maintenance, to put pressure on tenants to leave. They could then either reset rents to market rates with a new tenant or redevelop the building as condos or new construction, both of which are exempt from rent control. These landlord incentives would push rent control tenants out of the nicest neighborhoods.

In fact, we find that the landlords’ incentives appear to dominate. The average tenant treated by rent control lives in a census tract with worse observable amenities, as measured by the tract’s median household income, share of the population with a college degree, median house value, and share of unemployed people. Thus, while rent control does prevent displacement from San Francisco, it does not provide access to the best neighborhoods in the city.

Rent control expansion and rental supply

The evidence above suggests that landlords do not passively accept the burdens of the law. To examine landlords’ response to the rent control expansion in more detail, and to understand the impact of rent control on rental supply, we merge in historical parcel history data from the San Francisco Assessor’s Office, which allows us to observe parcel splits and condo conversions.

We find that rent-controlled buildings were 8 percentage points more likely to convert to a condo or a ‘tenancy in common’ than buildings in the control group. Consistent with these findings, we find that rent control led to a 15 percentage point decline in the number of tenants living in treated buildings and a 25 percentage point reduction in the number of tenants living in rent-controlled units, relative to 1994 levels.

This large reduction in rental housing supply was driven by both converting existing structures to owner-occupied condominium housing and by replacing existing structures with new construction.

The 15 percentage point reduction in the rental supply of small multi-family housing is likely to have led to rent increases in the long run, consistent with standard economic theory. In this sense, rent control operated as a transfer between the future tenants of San Francisco (who would pay these higher rents due to lower supply) to the tenants living in San Francisco in 1994 (who benefited directly from lower rents).

Furthermore, since many of the existing rental properties were converted to higher-end, owner-occupied condominium housing and new construction rentals, the passage of rent control ultimately led to a housing stock that caters to higher income individuals.

We directly test whether rent control led to in-migration of higher income residents by imputing household income as the per capita income of the census block groups in which the building occupants resided in five years prior. We find that this high-end housing, developed in response to rent control, attracted residents with at least 18% higher income, relative to control group buildings in the same zipcode.

The gentrification of San Francisco

Taking all of these points together, it appears that rent control has actually contributed to the gentrification of San Francisco, the exact opposite of the policy’s intended goal. Indeed, by simultaneously bringing in higher income residents and preventing displacement of minorities, rent control has contributed to widening income inequality in the city.

This article summarizes ‘The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco’ by Rebecca Diamond, Tim McQuade, and Franklin Qian, published in the American Economic Review in September 2019.

Rebecca Diamond is at Stanford University and NBER. Tim McQuade and Franklin Qian are at Stanford University.