Let’s Talk About Rent Control

I’ve been putting off writing a post on rent control for a while now. I knew I was going to need to start talking about it, with the Justice for Renters (Prop 33) ballot measure coming up, but Biden’s proposal last week for a national rent control, and the responses I’ve seen to it on Twitter, have finally pushed me over the edge.

I’ve avoided talking about rent control because it’s one of those “no win” topics. No matter what I say about it, I’m bound to piss somebody off. I have followers with a wide range of ideological backgrounds, and rent control is a particularly polarizing litmus test. If you ask a landlord or a libertarian-leaning supply-focused YIMBY, rent control is one of the most devastating housing policies you could implement, and economists apparently universally agree with this. If you ask a left-leaning housing advocate, rent control is a crucially important policy for preventing homelessness.

Before we can say whether rent control is “good” or “bad,” we first need to figure out what we’re talking about when we say “rent control” This is one of my biggest frustrations with discussions on this topic: no one ever specifies what they actually mean! This is particularly important because rent control is an area where subtle policy details make a massive difference in impact.

Policy Differences

There are three main areas where rent control policies can differ. The first is whether or not the rent control includes “vacancy control.” The second is what buildings rent control applies to. The third is how much rents in rent-controlled buildings are allowed to be raised each year.

Rent control with and without vacancy control are essentially two completely different policies. When rent control is implemented without vacancy control (sometimes referred to as “rent stabilization”), landlords are not allowed to raise the rent on an existing tenant (beyond the allowed amount), but when a new tenant signs a lease, the landlord is allowed to set their initial rent however they want. When rent control is implemented with vacancy control, a landlord must charge the same amount to a new tenant as the previous tenant was paying. In other words, rent stabilization-style rent control lasts for each tenancy, while vacancy control lasts for as long as the unit stands.

Rent control policies also often differ on what buildings they apply to. Most commonly, this is done by building age or construction date. Some rent control policies apply rent control universally to all buildings. Other policies apply rent control to all buildings built before a certain cut-off date. For example, Los Angeles’ Rent Stabilization Ordinance applies to all buildings built before October 1978. Finally, many rent control policies apply on a rolling basis to buildings beyond a certain age, typically in the range of 15-30 years old. In addition, rent control policies may exempt certain types of buildings or units, such as single-family homes, condos, or “luxury” units with a starting rent above a certain amount.

The last major difference between rent control policies is how much rents are allowed to rise. The most extreme rent control measures allow for no increases at all. More commonly, rent control policies allow annual rent increases to keep up with the increase in operating costs or inflation. On the opposite end of the spectrum are rent control measures that are merely “anti-price gouging” policies and only prohibit rent increases above what lawmakers consider an “obscene” amount, such as 10% of 30%. From what I’ve seen, most rent control policies seem to lie somewhere in between, allowing annual increases in the range of 3-6%.

Why I Support Rent Control

Many of you will be shocked to learn that, yes, I am in favor of some forms of rent control, particularly rent control without vacancy control. Let me explain why.

The clearest argument against all forms of rent control is that it’s economically inefficient. You’ll often hear opponents of rent control point out that the vast majority of economists oppose rent control. From an abstract point of view and on a long enough time horizon, they are right. Rent control is the classic example of a price ceiling that interferes with the functioning of a market and results in reduced goods allocation. As Khan Academy aptly puts it in their Micro Econ 101 lesson: “When a price ceiling is set below the equilibrium price, quantity demanded will exceed quantity supplied, and excess demand or shortages will result.”

While fundamentally correct, there are a lot of issues with this framing. First, housing markets are already highly regulated in the US, and these regulations may have a much greater impact on efficient market operation than rent control. Second, the housing market in a city is not actually a single market with a single market-clearing rent, but rather a complex system of interacting markets. Third, both supply and demand in a housing market is relatively inelastic in the short-term. And finally, although it’s a reasonable approximation of the impact in an abstract model, rent control without vacancy control is not actually a price ceiling^[1].

My primary reason for supporting rent control without vacancy control, however, is not related to the economic impact, but rather the real-life human impact.

I believe that housing stability is one of the most important prerequisites for both personal well-being and societal cohesion. When rents are quickly rising, the people who are most severely impacted are the people who are most vulnerable. I believe that it is worth sacrificing some economic efficiency to prevent the complete disruption of these people’s lives.

Being forced to move because your rent is being increased beyond what you can afford is incredibly disruptive and harmful. Even in the best-case scenario where there’s plenty of affordable options nearby, moving apartments is time consuming and costly. You have to spend time and effort to look for and tour apartments, time and effort going through the application process, and time and effort actually moving. This time and effort is a significant sacrifice, particularly for working parents, and it often requires taking time off work, resulting in reduced income. And moving is incredibly costly too! The working poor don’t have flexibility in their budgets for an extra couple hundred-dollar expense every year.

Chances are, however, that if your rent is rising, so is that of the other homes available in your neighborhood. Oftentimes, tenants who are priced out can’t find other affordable options nearby and are forced to move to another neighborhood, or even another city or state. Stating the obvious, being displaced from your neighborhood is a whole nother magnitude of disruptive and harmful.

Long-term residents with family and community ties may find themselves having to move somewhere where they are far from family and friends. Chances are, your new neighborhood is further from work and is going to require a longer commute. And that’s if you’re lucky. If you have to move far enough away, you might need to find a new job. Your kids will face even more disruption. They’ll have to start at a new school and make new friends. You may no longer have family nearby that can watch your kids after school, forcing you to work fewer hours or pay money for after-school care.

And if you’re uncharacteristically heartless about tenant well-being, perhaps you may be interested to hear that rent control without vacancy control is good for a city’s economic health. Ensuring working-class renters can remain in their neighborhoods and closer to their jobs reduces worker turnover and prevents a service-worker shortage. Small businesses, particularly local shops and restaurants, also benefit from a stable customer base.

Let’s shift back to talking about the economics, because understanding the dynamics of what’s actually going on in a housing market is helpful for weighing the benefits and tradeoffs of rent control.

As I mentioned earlier, supply in a housing market is relatively inelastic in the short term. Building more housing takes years and millions of dollars, and in a somewhat-functioning housing market there is rarely much inventory being mothballed that could quickly be brought online as prices increase. As a result, a demand shock will cause rents to skyrocket without significantly increasing the quantity supplied in the short term. When this happens, rent control without vacancy control protects current renters from the impact of that demand shock.

But here’s the key: rent control without vacancy control still allows rents to increase on the margins, which induces the production of new supply. In other words, when there’s a demand shock, while existing tenants won’t be impacted, new people looking to move into the market will see higher rents, and developers will be incentivized to build in that market to capture those higher rents.

In fact, when hit with a demand shock, a market with rent control may actually see marginal rents increase faster than one without rent control, incentivizing the needed production to happen sooner. This would occur because a market with rent control lacks the increased vacancies caused by long-term tenants getting priced out of the neighborhood. As a result, the same number of newcomers are vying for a smaller pool of available apartments and thus bidding up their pricing.

What happens in the aftermath of a demand shock? While housing supply is highly inelastic in the short term, as long as there is adequate zoning capacity to allow for more development^[2], housing supply should be long-run elastic, and new homes will get built to accommodate the increased demand. This in turn will cause a new market equilibrium, one that may have somewhat higher rents than prior to the shock, but with rents far below what they were at the height of the shortage. You can see a real-world example of this happening right now in Austin, Texas.

Let’s look at what occurs in the non-rent-controlled market. When prices settle into their new equilibrium, many of the tenants who were displaced may be able to afford their old neighborhood again. However, they’ve already uprooted their lives and would need to deal with all the same disruptions and costs all over again to move back. Furthermore, their community has already been destroyed, so moving back would not be the same as had the demand shock never occurred. Meanwhile, their landlord got a few years of additional income from elevated rents, but they now have to deal with the hassle of dealing with the vacancies from tenants moving into the newer, nicer buildings.

In the rent-controlled market, the tenant’s life was never disrupted, and the community has remained intact. Small businesses still have their local employees and old customers, along with new customers that can help them thrive. The landlord was not able to take advantage of the frenzy, but they are no worse off than they were before the demand shock, and they can raise rents to the new market equilibrium over time as tenants leave on their own accord.

That’s not to say that there aren’t some downsides of rent control without vacancy control. For landlords, there can be substantive operational and financial difficulties from having super-long-term tenants with significantly under-market rents. It is also undeniable that tenants who are looking to move, whether to a new neighborhood or just upsizing or downsizing, will have fewer options, and these options will be more expensive than in a non-rent-controlled market. And because tenants may want to “hold on to” their unit, rent control can act as a disincentive for households to upsize or downsize or move into a neighborhood or home that better fit their needs. However, I believe that these trade-offs are worth it for the benefit of preventing the pain and suffering from forced-displacement.

Why I Oppose Vacancy Control

What is the purpose of rent control? There are two possible answers here. The first answer, which I’ve explored above, is to protect current tenants against displacement. The second answer, which you might hear from someone who isn’t super in-the-weeds with housing or economic policy is to “keep rents low.”

In order for rent control to “keep rents low” you need to have vacancy control. On the surface, it sounds pretty straightforward. If landlords are not allowed to raise the rent, rent will stay affordable. However, reality is never that simple. I believe that rent control is NOT an effective policy to keep prices low, and that the negative consequences of vacancy control far outway the promise of positives.

Rent control with vacancy control is much closer to the economics textbook example of a price ceiling. While it is still not a pure price ceiling for the reasons discussed in the earlier footnote, as time goes on it acts more and more like a price ceiling and is more and more or a market-distorting factor. Just like overly-restrictive zoning, vacancy control serves as an entropy factor, leading to ever-growing dysfunction in a housing market.

The first obvious and much-discussed impact of vacancy control is that it significantly reduces the production of new rental housing. There is less capital willing to invest in a market with vacancy control, and to account for the greater risk vacancy control causes, buildings need to sell for a higher multiple of rent (i.e. cap rate). However, buildings cost the same to build whether or not there is vacancy control, so yield on cost is the same. Developers’ profits come from the spread between cap rates and yield on cost. As a result, at the same market rent levels, there are significantly fewer instances where development will be profitable in a city with vacancy control vs one without.

And, while the risk created by vacancy control impacts cap rates even when the policy is designed such that the rent control doesn’t kick in until the building is older, it has a second impact on development when vacancy control is applied to newly-constructed buildings. In this case, it makes development risky no matter what the cap-to-yield spread is. This is because you have only one chance to set rents that will impact the value of the building for all of eternity. If there is an economic crash or some other factor that hits while the building is under construction, you can’t set rents low and then raise them once the economy recovers. Instead, you have to either hold the building empty for years or take a massive loss on what otherwise should have been a good investment.

The next impact from vacancy control is that it lowers housing quality and leads to poor living conditions. Apartment buildings need constant maintenance, and this maintenance takes money. While some maintenance items, such as a leaky roof or a broken boiler, need to be fixed right away, others can be put off. Often, when mandatory expenses (such as taxes and insurance) rise, but rents can’t rise to compensate, landlords will make up for the missing revenue by skimping on non-urgent maintenance. While this can be indicative of a slumlord that is milking profits at the expense of their tenants, this also frequently occurs with well-meaning landlords who are in over their heads and need the revenue to cover debt payments.

Over time, this deferred maintenance adds up, and the housing quality deteriorates. At this point, the building needs an influx of capital to bring it up to habitability standards. Even buildings owned by responsible landlords who are diligent about doing the necessary maintenance run into instances where they need an influx of capital for repairs. Major building components and systems such as roofs, balconies, plumbing, HVAC, and electrical have a finite lifespan and need to be replaced every 30-70 years. Without these significant repairs, a building may become completely uninhabitable.

At this point, the building owner has three options for getting money to make the necessary repairs. If they were truly responsible, they should have been putting money aside to cover future maintenance expenses. But if they haven’t, or they already spent the savings on prior repairs, they will need to either take out debt by refinancing their building or get an outside equity investor. However, if existing rents aren’t high enough to cover interest payments or provide investment returns, and there are no prospects of ever raising rents in the future, the owner has no way of paying for fixes. So cities with vacancy control often end up with increasingly derelict housing conditions.

Which brings me to the final impact of vacancy control: supply reduction. When a tenant leaves an apartment, the landlord must “turn” the apartment. If the tenant was just there for a year or two, a turn may consist of just a deep cleaning and maybe a fresh coat of paint. On the other end of the spectrum, if a tenant leaves after fifty years, an apartment might need tens to hundred of thousands of dollars of work to bring it up to a condition where it can be rented. If the apartment’s rents are low enough, and vacancy control prevents rent from being raised for the next tenant, it might be impossible to break even on the cost of turning the apartment.

When this happens, it can make sense for the building owner to leave the apartment vacant with the plan of eventually abandoning the building as a total loss rather than throwing good money after bad. How common of an occurrence this is is hotly debated, and the impact can be negated by implementing policies that allow for rent increases specifically to cover these expenses.

Taken together, the end result of vacancy control is a worsening housing shortage and worsening housing conditions. A tenant who is remaining in their apartment sees no savings from vacancy control over rent control and may be negatively impacted by deferred maintenance. Although someone looking to move apartments or who is moving to a new city would benefit from lower rents, they can only benefit if they are able to find a place to move to. In reality, they’re more likely to find themselves competing with hundreds of other families whenever an apartment becomes available.

Advocates who are more familiar with the downsides of vacancy control will sometimes still advocate for it in order to prevent tenants from harassment. The reasoning here is that when rent control is implemented without vacancy control, landlords are incentivized to remove long-term tenants at all costs. This is a legitimate concern. There are many examples of landlords harassing rent controlled tenants in order to get them to leave, often in despicable ways. However, I think there are better, more targeted ways to stop tenant harassment that don’t have the significant consequences of vacancy control.

What About New Development?

While I am very confident in my views on rent control and vacancy control, I am less certain on whether or not it is necessary for rent control policies to exclude new development. As I discussed earlier, if vacancy control is implemented, not excluding new development makes an already-bad situation worse. However, it’s far from clear that this is a major issue for rent control without vacancy control.

On one hand, I think it’s better to be safe than sorry in terms of ensuring that rent control does not hinder much needed new development. Tenants of new buildings are generally higher-income and are less likely to stay long-term, so it’s less important to provide protections for them.

On the other hand, I don’t think the evidence is there to support the claim that applying rent control to newly-constructed buildings has a major impact on whether or not developers build in an area. Sure, all things being equal, a developer would prefer to build in a jurisdiction without rent control, and many developers will vocally complain about rent control, but unless they are projecting major short-term rent growth, rent control shouldn’t make a major difference in a developer’s pro-forma.

In 2017, Los Angeles implemented a policy that applies the Rent Stabilization Ordinance to any new building that replaced a demolished rent-controlled building, unless the new building contained 20% affordable units. Despite this, there does not seem to have been any subsequent shift in development to properties where this would not apply. (I will note that some people have suggested that most developers were not aware of the new rule and didn’t know that their buildings would be rent controlled)

Another example is Montreal. They have a new-construction exemption to rent control, but it only applies for five years, rather than the 15-30 years generally advocated for by developers. And despite this, Montreal continues to build more of its new housing as rental housing than any other major city in Canada.

Fundamentally, developers are going to decide whether or not to build based on if their pro-forma financials show an adequate spread between the yield on cost and the exit capitalization rate to provide their investors with adequate returns given a project’s risk. If they need to rely on future anticipated rent growth to reach the necessary yield, it’s going to be a much riskier project, and they probably shouldn’t proceed with it. Rent control could impact the exit capitalization rate, but I don’t think the cap rate difference between rent control that kicks in right away vs rent control that kicks in in 20 years is going to be that big.

All that being said, given the housing shortage, we need to be removing all possible barriers to new housing construction. There are certainly some developers who gamble on projecting future rent growth, and I might be underestimating the impact of rent control on building sales prices. As a result, I don’t think that the benefits from applying rent control to new construction are worth the risk that it will reduce development activity. I think a reasonable exclusion length would be around 20 years.

Why I Oppose Prop 33

The reason why it’s so important to be having the “rent control discussion” now is because California voters will have to make a choice on rent control in the November election. Proposition 33, marketed as the Justice for Renters Act, would overturn the 1995 Costa–Hawkins Rental Housing Act that limits rent control in California cities. Unfortunately, while I support the idea of overturning many parts of Costa-Hawkins, Proposition 33 is a bad ballot measure that should be defeated.

I have both objective and subjective reasons for opposing Prop 33, and I’ll start with the objective. The text of Prop 33 is deceivingly simple. It removes the entire text of Costa-Hawkins from the California civil code and replaces it with “The state may not limit the right of any city, county, or city and county to maintain, enact or expand residential rent control.” While simply-written laws can be appealing on the surface, if you start to think about what this does, it gets alarming. If Prop 33 passes, the legislature cannot, under any circumstances, regulate rent control in any way.

The first obvious result of this is that the state cannot stop cities from implementing vacancy control. In fact, some cities have vacancy control measures on the books that will kick in as soon as Prop 33 goes into effect. If you agree with my earlier argument that vacancy control is bad, this is reason enough to oppose Prop 33.

However, the problems with Prop 33 go beyond vacancy control. Due to the lack of specificity in the ballot measure, cities may be able to implement far more asinine forms of rental pricing controls, such as city-wide maximum rent. Another very real possibility is that some cities, who do not care at all about renters, will weaponize rent control to block the construction of new rental housing. Most notably, a right-wing anti-development city council member of Huntington Beach suggested that the city would be able to use the proposition’s language to stop the state’s efforts to force them to allow more housing.

If it’s puzzling to you why this ballot measure contains such obvious flaws, the answer is that it was written and funded by the AIDS Healthcare Foundation (AHF) and its president Michael Weinstein. If you haven’t heard of AHF, you’ve certainly seen their billboards telling you to wear a condom or get an STD check. What started as a hospice non-profit that did important work in the early days of the AIDS crisis, AIDS Healthcare Foundation has grown into a financial and political behemoth that is one of the largest distributors of AIDS medication in the world. Taking advantage of medication rebate programs from the Federal Government, AHF brings in a massive amount of cash. Unfortunately, rather than reinvesting that money into helping AIDS patients, AHF founder and president Michael Weinstein uses the foundation as his piggy bank for political whims.

Michael Weinstein’s political campaigns are all over the place. They are sometimes well-intentioned, often misguided, and typically failures. He pushed the City of Los Angeles to require porn actors to wear condoms, which resulted in the porn industry moving to Arizona. He somewhat perplexingly also opposed the approval of AIDS-prevention medication PrEP. Thankfully, that effort was ignored, and the advent of PrEP has significantly reduced the spread of AIDS in gay men.

Weinstein first got interested in the topic of land use and housing after a building was proposed that would block his office’s view of the Hollywood sign. In reaction, AHF sued to block the project and also sponsored a ballot measure that would have put a two-year moratorium on development in Los Angeles. The ballot measure lost in a landslide, and AHF lost the lawsuit, but they were successful in delaying the project enough that it still hasn’t been built.

Since then, AHF has poured millions of dollars into losing housing ballot measures, including two previous state-wide rent control measures. They’ve also gotten into the housing business themselves, buying up dozens of buildings on Skid Row. This too has gone badly. An LA Times investigation found dangerous conditions at their buildings and significant mismanagement. The report was so alarming that the state recently intervened to prevent AHF from buying the buildings the City of Los Angeles was forced to take over after Skid Row Housing Trust went defunct.

I’m not going to go so far as calling AHF slum lords, as many others have, because managing homeless housing in ancient buildings is incredibly difficult. But perhaps there’s a lesson here. AIDS Healthcare Foundation (and for that matter, Skid Row Housing Trust) isn’t able to properly maintain and repair their buildings, because the fixed rents from homeless tenants’ vouchers aren’t enough to cover the expenses. Maybe we should think before we impose this same problem on apartment buildings throughout the state.

But Michael Weinstein isn’t thinking of second order effects or unintended consequences. He never does. Instead, he brashley charges forward with big policy proposals based on an undeveloped understanding of how housing actually works.

The danger is: this time he might actually be successful. Since failing at his last rent control ballot measure in 2020, Weinstein has been using AHF’s deep pockets to gain influence among organizations in hopes of building grassroots support for Prop 33.

Most of the vocal opposition for Prop 33 is going to be coming from Landlord organizations, and I suspect their messaging is going to be against rent control of all forms. I too will be working to defeat Prop 33, but I don’t agree with the Landlords’ perspective. I believe that when well designed, rent control can serve an important role to protect tenants without impeding development or breaking the housing market. I look forward to a future where we can both build the housing we desperately need and ensure tenant stability. Unfortunately, Prop 33 takes us further from that future.