Follow-Up to My Investor Ban Rant

Recently, I posted an exasperated rant about the rash of state-level bills that are being proposed that would ban large investors from buying homes.

I had written about this before in various outlets, predicting that this day would come and warning that, as the headline in my Discourse Magazine article said, this would be the “last no in housing”.

By that, I mean that a century of this sort of crap has literally left us with a single version of housing, on the margin, that is capable of growing - single family homes built for rent. Cities block apartments. Now federal mortgage regulators limit single-family homeowners. Those 2 markets are basically growing as fast as they are legally allowed to. That isn’t enough. We are short 15 or 20 million homes now, and in order to bridge the gap, they will have to be single-family rentals.

Rentals are owned by investors. If we need 15 or 20 million of them, they will have to be owned by investors that can scale. If we make this final source of housing illegal, we are directly, explicitly, legislating homelessness.

One of the reactions that I commonly get is that we could allow investors to build new homes, and we could just ban them from buying existing homes. The idea is that when they buy existing homes, they drive up the price for families who want to be homeowners.

There are a number of reasons why that’s not a workable compromise. But, here I just want to make one point of analysis.

Figure 1 shows 3 measures. The black line is the average price/income level of homes in ZIP codes with the highest quintile of incomes. The blue line is the average price/income level of homes in ZIP codes with the lowest quintile of incomes. The red line is the percentage of detached single-family homes that are owner-occupied.¹

I haven’t controlled for regional differences or anything, here, on the price/income measures, so there is a lot to unpack there that I’m not going to get into here. In 2007, for instance, the average price/income ratio of homes in the lowest income quintile was a combination of a lot of different contexts. There were a lot of poor ZIP codes across the country that remained inexpensive, while the poorest ZIP codes of the most expensive cities that have most thoroughly banned housing production became wildly more expensive.

After the 2008 crisis, trends have been more uniform across the country, because we engineered a housing shortage across the country.

Anyway, the main point of this chart, which I touched on in my Mercatus brief on this topic, is that the broad increase in low-tier home prices since 2014 has been associated with an increase in owner-occupier ownership of single family homes. In other words, a decline in investor ownership.

There are areas where investor owners have concentrated, and some analysis has found that rents are going up in those areas. The reason rents are going up in those areas, and the reason investors are concentrating, is because owner-occupiers are claiming more of the existing stock of homes. The set of homes that are available to be rented is shrinking. That’s not the fault of large-scale investors. That’s the fault of homeowners.

The reason this is happening is that, at all times, and in all places, homeowners with funding always outbid investors because homeownership has value. Homeowners are generally willing to pay more than investors for any given house.

That is why large-scale institutional ownership of single-family homes has never been a market segment before, beyond a few niches.

By the way, multi-family housing that is rented accounts for about 24% to 25% of the total stock of housing over the whole period of time since 2010. Investors would love to increase the market share of multi-family housing for rent. They can’t. Cities won’t let them.

So, that outlet for rental homes can’t grow, and the stock of single-family homes that could be rented keeps shrinking. There are fewer rentals available, and household formation of renters is being squeezed. Basically, at this point, owner-occupiers buy up more of the existing homes, and rents inflate further until the necessary number of renters gives up on forming a new household.

Finally, the demand from homeowners has driven rents up far enough that builders are starting to construct new single-family homes for large-scale investors. That might finally release the pressure and grow the stock of new homes enough to finally lead to moderation in rents.

But, back to the motivating question for this post. What is the optimal outcome here? If investors shouldn’t be buying single-family homes, what percentage of homes should they currently own? It declined from 17% in 2016 to 13% in 2023. How low is enough to satisfy you? 10%? 5%? At what point do they cease to be a problem? And, if single-family homes were 95% owned by families and only 5% owned by investors, what condition do you think renters would be in? Would rents for them be higher or lower?

Ironically, we are at this strange moment where price/income ratios of homes in the poorest ZIP codes are significantly higher than price/income ratios of homes in the richest ZIP codes, because low-tier rents have increased so much since 2014. And, they have increased so much because we have a shortage that we extended nationwide after 2008 by limiting who can buy homes.

If you are concerned enough about the price of homes to ban investors from single-family homeownership, you might want to think long and hard about exactly what the distributional consequences of your demands would be if another 3% or 5% or 8% of the housing stock was shifted from renter-occupied to owner-occupied.

At this late date, I would hope that everyone starts taking a position of humility before they start supporting laws that come up with new reasons why people or (gasp) corporations can't own, buy, or build housing.

1. I have only been able to find annual data for the percentage of single-family homes owned vs rented dating from 2010. The 1990 Census puts the owner-occupied portion at about 85.6%. I expect that it increased a bit between 1990 and 2003. I’m not sure at what percentage level it would have peaked. Then, it declined from about 2004 to 2014 before starting to rise again.