Kevin, have you ever looked at permits versus population growth? I was hoping to produce a graphic that supports the lack of houses, but I do not believe it does. Since 1960, the median number of permits (FRED: New Privately-Owned Housing Units Authorized in Permit-Issuing Places: Total Units (PERMIT)) per person of population growth is .53 (or 1.88 people per permit). The median is skewed slightly above the average (permits / person). Any thoughts on this?
That is mostly due to declining family size. There is a relatively steep downward trend in children per home for decades. And, a shallower decline in adults per home which reversed in 2008 and has trended up since then.
Homes per capita includes children in the denominator, so it has more of an upward drift over time, although that is generally a steep rise in homes per capita until the 1990s that has generally flattened out since then.
That's why a more thorough explanation of whether there is a shortage or not depends more on rent inflation and migration trends, where rising housing costs affect decisions about moving or not moving for millions of American households in a way they did not used to.
Just to expand on this, it wouldn't be a shortage if we just decided that, as an economy, we opposed consumption of housing. We could tax it more, etc. So, it could be the case that even if incomes increase 60% over the next 30 years, Americans still consume the same amount of housing per capita as they do today. And it could be fine. We would just consume other things instead.
Now, I think that's a weak position to take. Skeptics of the housing crisis sometimes try to benchmark to earlier consumption levels, as if any increase in the consumption of housing since, say, 1970, is evidence that there isn't a problem. As if we shouldn't benchmark to any rate of economic progress.
But, more specifically, the forms of housing opposition and the intensity of it have had very asymmetrical consequences, so that many households are simply trying to live the lifestyles they have been leading, but housing costs are rising so much that they cannot. Those are the consequences I describe as a true shortage.
But, the long-term trends in things like units per capita are sort of coming at that problem sideways, and don't quite capture the full consequence of the problem.
Very interesting but I want to push back on the language of how many houses we "need". In my experience talking about how many houses are needed encourages a view amoung the general public that development plans should figure out how many houses are needed and only authorize development for that many rather than just zone to allow building and let the market determine if it makes economic sense to build more.
And yes, there is obviously going to be some socially optimal level of housing so I don't mean to suggest anything is wrong with what you said but I just think it would be better to talk about how many houses would be ideal as I think people react to that language differently.
Yes. That’s a good point. It also might encourage the ridiculous implication that seems to have even affected the academy in 2008 that if you build too many houses the market will collapse.
But there is a basis for writing that way. It’s the amount of housing that would reverse excess rents.
Sorry, but what exactly do you mean by excess rents? I mean the more housing that gets built the lower the rents.
Of course you can ask what the rents would be without some specific distortion. Are you suggesting that is the number of units which if built would eliminate any premium as a result of zoning restrictions or what?
The key is those scatterplots of LA and Phoenix. Blocking housing is a peculiar issue historically. When it happens it creates that peculiar regressive cost inflation. Rents and prices rise in the poorer neighborhoods. Every city used to look somewhat like Phoenix before 2008. Where the correlation is especially negative between price/income and income, that’s all land rents going to a land cartel. We will have enough housing when the cartel isn’t collecting rents.
Surely that can't be the definition of "enough." I mean one world in which you might not have enough housing is one in which the local municipalities simply pass laws which impose lots of unnecessary cost on developers (say tons of various studies and impact reports).
No cartel is collecting any rents in this situation though. Sure, the municipalities could have choosen to collect rents instead of simply imposing wasteful costs but they often don't because it prevents the excuse that the extra requirements are actually important.
High costs wouldn’t create the price/income pattern I highlight. High costs would lead to either smaller units or prices that were elevated across the board, depending on how they were imposed. This pattern only occurs where new housing is so obstructed that rising rents are broadly displacing residents who are simply trying to remain in place. It’s a very peculiar signal.
Kevin, have you ever looked at permits versus population growth? I was hoping to produce a graphic that supports the lack of houses, but I do not believe it does. Since 1960, the median number of permits (FRED: New Privately-Owned Housing Units Authorized in Permit-Issuing Places: Total Units (PERMIT)) per person of population growth is .53 (or 1.88 people per permit). The median is skewed slightly above the average (permits / person). Any thoughts on this?
That is mostly due to declining family size. There is a relatively steep downward trend in children per home for decades. And, a shallower decline in adults per home which reversed in 2008 and has trended up since then.
Homes per capita includes children in the denominator, so it has more of an upward drift over time, although that is generally a steep rise in homes per capita until the 1990s that has generally flattened out since then.
That's why a more thorough explanation of whether there is a shortage or not depends more on rent inflation and migration trends, where rising housing costs affect decisions about moving or not moving for millions of American households in a way they did not used to.
Just to expand on this, it wouldn't be a shortage if we just decided that, as an economy, we opposed consumption of housing. We could tax it more, etc. So, it could be the case that even if incomes increase 60% over the next 30 years, Americans still consume the same amount of housing per capita as they do today. And it could be fine. We would just consume other things instead.
Now, I think that's a weak position to take. Skeptics of the housing crisis sometimes try to benchmark to earlier consumption levels, as if any increase in the consumption of housing since, say, 1970, is evidence that there isn't a problem. As if we shouldn't benchmark to any rate of economic progress.
But, more specifically, the forms of housing opposition and the intensity of it have had very asymmetrical consequences, so that many households are simply trying to live the lifestyles they have been leading, but housing costs are rising so much that they cannot. Those are the consequences I describe as a true shortage.
But, the long-term trends in things like units per capita are sort of coming at that problem sideways, and don't quite capture the full consequence of the problem.
Very interesting but I want to push back on the language of how many houses we "need". In my experience talking about how many houses are needed encourages a view amoung the general public that development plans should figure out how many houses are needed and only authorize development for that many rather than just zone to allow building and let the market determine if it makes economic sense to build more.
And yes, there is obviously going to be some socially optimal level of housing so I don't mean to suggest anything is wrong with what you said but I just think it would be better to talk about how many houses would be ideal as I think people react to that language differently.
Yes. That’s a good point. It also might encourage the ridiculous implication that seems to have even affected the academy in 2008 that if you build too many houses the market will collapse.
But there is a basis for writing that way. It’s the amount of housing that would reverse excess rents.
Sorry, but what exactly do you mean by excess rents? I mean the more housing that gets built the lower the rents.
Of course you can ask what the rents would be without some specific distortion. Are you suggesting that is the number of units which if built would eliminate any premium as a result of zoning restrictions or what?
The key is those scatterplots of LA and Phoenix. Blocking housing is a peculiar issue historically. When it happens it creates that peculiar regressive cost inflation. Rents and prices rise in the poorer neighborhoods. Every city used to look somewhat like Phoenix before 2008. Where the correlation is especially negative between price/income and income, that’s all land rents going to a land cartel. We will have enough housing when the cartel isn’t collecting rents.
Surely that can't be the definition of "enough." I mean one world in which you might not have enough housing is one in which the local municipalities simply pass laws which impose lots of unnecessary cost on developers (say tons of various studies and impact reports).
No cartel is collecting any rents in this situation though. Sure, the municipalities could have choosen to collect rents instead of simply imposing wasteful costs but they often don't because it prevents the excuse that the extra requirements are actually important.
High costs wouldn’t create the price/income pattern I highlight. High costs would lead to either smaller units or prices that were elevated across the board, depending on how they were imposed. This pattern only occurs where new housing is so obstructed that rising rents are broadly displacing residents who are simply trying to remain in place. It’s a very peculiar signal.