You say that "Millennials have never experienced a single time when the supply of new homes had shifted enough to change anyone’s expectation of rent trends at any scale that any developer would bother to add to a scenario analysis" but you also describe how rents have fallen in Austin and how the situation will not automatically reverse. So, wouldn't a developer we wise to add this supply scenario analysis to their underwriting?
Ah. Good point. I might have gotten a bit over my skis there. I was thinking in terms of national statistics. But, you're right, in a few cases like Austin, local supply trends were probably of a large enough scale to factor into builder scenario analysis. Austin is an outlier. Florida is the counterexample. There have been price fluctuations there similar to Austin, but supply hasn't recovered nearly as much there. Yet, the fluctuations in post-Covid migration and population growth created big shifts in housing economics in Florida. In Austin, developers might have had an eye on the effect of supply. In Florida, it is more dwarfed by the other factors, which is the more typical context.
San Francisco is an even better counterexample than Florida. Clearly, there, supply fluctuations never get close to being an important source of changing rent trends, but rents and prices have been volatile during the Covid aftermath.
I think San Francisco may be the more appropriate comparison for the Midwest. There has been very little supply response in the Midwest to a decade of rising rents. One reason for that is that anti-apartment zoning appears to be binding, as it is in San Francisco. Since 2008, both areas have permitted about 0.5% annual growth in the housing stock.
In San Francisco, single-family production is municipally limited, and so the market clears through elevated land prices. In the Midwest, since 2008, I would say that single-family production was limited because the price of existing homes was pushed too low, so new homes didn't pencil. But, increasingly, Midwest markets are clearing at low production levels because the prices of inputs are being bid up by builders in regions with more demand. Eventually, as capacity grows, input costs will normalize to allow Midwest neighborhoods to pencil again.
In San Francisco, significant regional displacement and outmigration are the main source of supply for new households, and so I think San Francisco has a lot more capacity to grow, because the displaced families have very inelastic demand to remain there if they can. But, eventually, I think supply in the Midwest will recover to something near the pre-2008 norm, and that will be associated with an end to the perpetual rent inflation. I think building will probably continue to grow beyond that when there is capacity for it, and eventually rents will correct back down a bit, but that will be after years of construction far above current levels. Eventually, my guess is that the Midwest will have some household formation recovery, though it appears that the number of adult children living with parents tends to be lower in the Midwest than it is in other regions, so there may be less household recovery potential there than in other regions (probably more in Columbus but less in Benton Harbor, etc.) But, I think in the more robust Midwest markets, supply will eventually correct to the upside to make up for 20 years of underbuilding.
Long story short, yes, I think you could say that supply matters more where growth is slow. If Cincinnati suddenly started building at Austin rates, it would probably be getting ahead of itself. On the other hand, in relative terms, if it takes 1% more supply to lower rents 2%, then tripling the rate of building in the Midwest would still only lead to minor shifts in the trajectory of rents.
You say that "Millennials have never experienced a single time when the supply of new homes had shifted enough to change anyone’s expectation of rent trends at any scale that any developer would bother to add to a scenario analysis" but you also describe how rents have fallen in Austin and how the situation will not automatically reverse. So, wouldn't a developer we wise to add this supply scenario analysis to their underwriting?
Ah. Good point. I might have gotten a bit over my skis there. I was thinking in terms of national statistics. But, you're right, in a few cases like Austin, local supply trends were probably of a large enough scale to factor into builder scenario analysis. Austin is an outlier. Florida is the counterexample. There have been price fluctuations there similar to Austin, but supply hasn't recovered nearly as much there. Yet, the fluctuations in post-Covid migration and population growth created big shifts in housing economics in Florida. In Austin, developers might have had an eye on the effect of supply. In Florida, it is more dwarfed by the other factors, which is the more typical context.
San Francisco is an even better counterexample than Florida. Clearly, there, supply fluctuations never get close to being an important source of changing rent trends, but rents and prices have been volatile during the Covid aftermath.
Thank you , this makes sense. Do you think supply trends also matter more in a region like the Midwest that has less organic growth?
I think San Francisco may be the more appropriate comparison for the Midwest. There has been very little supply response in the Midwest to a decade of rising rents. One reason for that is that anti-apartment zoning appears to be binding, as it is in San Francisco. Since 2008, both areas have permitted about 0.5% annual growth in the housing stock.
https://fred.stlouisfed.org/graph/?g=1UAC9
And, rent inflation has been similar in both areas since 2008.
https://fred.stlouisfed.org/graph/?g=1UACo
In San Francisco, single-family production is municipally limited, and so the market clears through elevated land prices. In the Midwest, since 2008, I would say that single-family production was limited because the price of existing homes was pushed too low, so new homes didn't pencil. But, increasingly, Midwest markets are clearing at low production levels because the prices of inputs are being bid up by builders in regions with more demand. Eventually, as capacity grows, input costs will normalize to allow Midwest neighborhoods to pencil again.
In San Francisco, significant regional displacement and outmigration are the main source of supply for new households, and so I think San Francisco has a lot more capacity to grow, because the displaced families have very inelastic demand to remain there if they can. But, eventually, I think supply in the Midwest will recover to something near the pre-2008 norm, and that will be associated with an end to the perpetual rent inflation. I think building will probably continue to grow beyond that when there is capacity for it, and eventually rents will correct back down a bit, but that will be after years of construction far above current levels. Eventually, my guess is that the Midwest will have some household formation recovery, though it appears that the number of adult children living with parents tends to be lower in the Midwest than it is in other regions, so there may be less household recovery potential there than in other regions (probably more in Columbus but less in Benton Harbor, etc.) But, I think in the more robust Midwest markets, supply will eventually correct to the upside to make up for 20 years of underbuilding.
Long story short, yes, I think you could say that supply matters more where growth is slow. If Cincinnati suddenly started building at Austin rates, it would probably be getting ahead of itself. On the other hand, in relative terms, if it takes 1% more supply to lower rents 2%, then tripling the rate of building in the Midwest would still only lead to minor shifts in the trajectory of rents.