I'm generally favorable to rent control (it's a socio-political tradeoff, like all public policies), but rent control on single-family homes or single units (as

I'm generally favorable to rent control (it's a socio-political tradeoff, like all public policies), but rent control on single-family homes or single units (as with renting an in-law), which is what that paper examines, is a horrible policy. The consequences for landlords are far more extreme than rent control on large, multi-unit buildings. This makes it's difficult to infer similar supply effects between rent control on single-family homes and on multi-unit buildings. Turnover in SF is actually quite robust. The vast majority of tenants don't stay in the same apartment for decades. This is especially true during boom periods. IOW, ebbs & flows in turnover actually tend to benefit landlords. But a small number do. The average turnover rate is ~20%, which means the average occupancy is ~5 years. For large buildings what matters is the average. The fewer the number of units (either in a building, or in your portfolio), the riskier things become. The riskiest of all is a single-family home, whether you rent out an in-law or the whole house. In the face of rent control you're not only risking being stuck with the same person indefinitely, paying an increasingly below market rent, but you're locking up for an indefinite period what is probably your largest and most important asset. It's a potential financial nightmare, and one with realistic odds (e.g. maybe 1 in 10 instead 1 in 10000) of befalling you. As the building grows larger, the more the risk is spread and you can rely on the average occupancy term. Moreover, the larger the building, the less likely one would want to tear it down to build something larger. IOW, the larger the building, the more closely your expected future investment returns approach non-rent control levels. Rent control policies should be accompanied by policies that promote densification. All forms of rent control will tend to reduce supply to some extent (at least, that should be the a priori expectation), but rent control also provides important social benefits. IMO, those lines cross, but I would never expect the benefits to be greater than the costs in the context of single-family homes.