The effects of an Ellis Act eviction on neighborhood socioeconomic status

Rent-control advocates argue that its strongest feature is offering tenants strong protections from economic displacement. Nonetheless, rent control may have negative effects on tenants, as previous research has shown that these tenants have longer commutes and higher unemployment rates because they are incentivized to stay in place even after their location is no longer optimal. I study what happens to tenants when they are displaced from their rent-controlled apartments by exploiting a California law called the Ellis Act that allows landlords in Los Angeles and San Francisco to evict tenants even if they are lease-compliant, under the condition that all the tenants in the building must be evicted at once and are compensated by the landlord with substantial relocation payments. In large apartment buildings (five units or more), these Ellis Act evictions act as an exogenous shock because these landlords are unlikely to be evicting all their tenants just to target an individual household. Using Infutor data, I identify over 900,000 people who lived in a five-plus unit rent-controlled apartment in either San Francisco or Los Angeles in 1999, 11,470 of whom were evicted between 2000 and 2007. I find that evicted tenants were less likely to stay in their original city and more likely to live in lower-income and lowerintergenerational-mobility neighborhoods than control tenants. The negative effects of these evictions appear to be highly persistent: neighborhood socioeconomic status is lower for the evicted group than the control group at least 12 years ex post. These findings support that the Ellis Act imposes steep costs on tenants and may be partially undermining California's recent attempts to improve housing affordability and stability.