I previously wrote in general about the recently posted NBER working paper titled, “Supply Constraints do not Explain House Price and Quantity Growth Across U.S.
>In terms of carrying capacity, you could say they have a bizarre downward sloping supply curve. As demand for housing increases, their populations decline. And they decline by displacing the poorest residents that can’t afford the rising prices.
I think we need a two-sector model here: demand for SFHs increases, so apartments are torn down to build mansions. So S_sfh increases and S_apt decreases, which pushes up P_apt.
>Supply elasticity after 2008 was largely driven by the mortgage crackdown, which pushed home prices below replacement value and destroyed the entry level new single-family home market across the country.
If the mortgage crackdown suppressed demand, in what does sense does it affect supply elasticities? Because builders moved away and the industry died, so there were no suppliers to respond to price changes?
The mortgage crackdown created price deflation highly correlated with neighborhood income. So, new homes in price points that were a majority of new home sales before 2008 fell well below the cost of construction. So, almost all new demand on the margin just pushed prices up among existing homes until prices recovered back to the level that induces new building. I think many markets have passed that margin in the post-Covid period, which is why we are now seeing, and will continue to see, most of the marginal growth in new single family construction happening in the build-to-rent segment (because of the mortgage crackdown).
Millions of families have suddenly been excluded from mortgage financing. They are all generally living in homes as renters that are relatively similar to the homes they would have lived in as owners. But millions of homes have not been constructed because they are unable to purchase one from a builder. I think it’s fundamentally mistaken to think of this as a change in demand instead of a change in supply.
On a supply and demand graph, I would model that as a shift left in demand, moving along the supply curve; the supply curve itself doesn't change, so there's no change in the elasticity.
The shift in demand is not from a change in 'raw' willingness-to-pay, but from how WTP is converted in mortgages.
I think the tricky part is being clear about what we’re measuring. The consumption of shelter or the ownership of homes. You are describing demand for ownership of homes. And I think that’s right. The mortgage crackdown drove demand left to the vertical portion of the supply curve (representing the existing stock). That was related to a compression of price/rent ratios. The demand for shelter didn’t decline.
Then as the demand for shelter rose naturally from there, the supply curve for new homes was relatively vertical because the demand for purchasing them had been moved left. That was associated with both rising rents and rising prices.
The decline in demand for purchasing led to a steep supply curve for consuming, just like a decline in demand for farmland from tighter financing would lead to a steep supply curve for grain.
>The decline in demand for purchasing led to a steep supply curve for consuming
I'm wary of saying that a change in demand can change the supply curve. Do you mean the shift in demand moved the equilibrium to the steeper section of the supply curve?
If they had put more thought into their research, and looked at urban development trends prior to 1980, they should have concluded the following:
"Supply inelasticity induced by regulations on land use and mortgages has destroyed filtering effects for housing in many urban areas. The worst effects are in prominent cities with wealthy residents surrounded by politically active suburbs that further constrain housing production."
I admire the patience and tenacity Kevin Erdmann has shown in trying to tell the tale of US housing markets.
I will go out on a limb, and say "Yes, supply matters." It is a new idea I have, called "supply and demand."
Solutions?
Perhaps the US government should offer "bribes" for un-zoning or radical up-zoning of property in densely populated neighborhoods. Decriminalizing housing construction is a good idea.
That is, a city would get a large grant for up-zoning property in dense ZIP codes.
Another tactic is to pay a city, say, $200,000 for every housing unit built, if in a crowded zip code. That would be $100 billion a year, for 500,000 "extra" units.
By way of comparison, just the interest on the national debt today is near $1 trillion a year.
Yes, for 10% of the cost of interest on the national debt we might be able to have another 500,000 units a year. Worth a try.
I challenge the macroeconomic community to come up with better ideas.
Figure 4 is really eye opening. It’s not showing a *shift* from affordable to higher-priced homes, it’s showing the complete evaporation of construction of homes affordable to first-time home buyers. Probably the most stark graphic representation that I’ve seen
>In terms of carrying capacity, you could say they have a bizarre downward sloping supply curve. As demand for housing increases, their populations decline. And they decline by displacing the poorest residents that can’t afford the rising prices.
I think we need a two-sector model here: demand for SFHs increases, so apartments are torn down to build mansions. So S_sfh increases and S_apt decreases, which pushes up P_apt.
(or using High and Low quality)
That's one margin. Also, things like smaller household size could be part of it.
>Supply elasticity after 2008 was largely driven by the mortgage crackdown, which pushed home prices below replacement value and destroyed the entry level new single-family home market across the country.
If the mortgage crackdown suppressed demand, in what does sense does it affect supply elasticities? Because builders moved away and the industry died, so there were no suppliers to respond to price changes?
The mortgage crackdown created price deflation highly correlated with neighborhood income. So, new homes in price points that were a majority of new home sales before 2008 fell well below the cost of construction. So, almost all new demand on the margin just pushed prices up among existing homes until prices recovered back to the level that induces new building. I think many markets have passed that margin in the post-Covid period, which is why we are now seeing, and will continue to see, most of the marginal growth in new single family construction happening in the build-to-rent segment (because of the mortgage crackdown).
My question is how a change in the demand curve affects the slope of the supply curve.
Millions of families have suddenly been excluded from mortgage financing. They are all generally living in homes as renters that are relatively similar to the homes they would have lived in as owners. But millions of homes have not been constructed because they are unable to purchase one from a builder. I think it’s fundamentally mistaken to think of this as a change in demand instead of a change in supply.
On a supply and demand graph, I would model that as a shift left in demand, moving along the supply curve; the supply curve itself doesn't change, so there's no change in the elasticity.
The shift in demand is not from a change in 'raw' willingness-to-pay, but from how WTP is converted in mortgages.
I think the tricky part is being clear about what we’re measuring. The consumption of shelter or the ownership of homes. You are describing demand for ownership of homes. And I think that’s right. The mortgage crackdown drove demand left to the vertical portion of the supply curve (representing the existing stock). That was related to a compression of price/rent ratios. The demand for shelter didn’t decline.
Then as the demand for shelter rose naturally from there, the supply curve for new homes was relatively vertical because the demand for purchasing them had been moved left. That was associated with both rising rents and rising prices.
The decline in demand for purchasing led to a steep supply curve for consuming, just like a decline in demand for farmland from tighter financing would lead to a steep supply curve for grain.
>The decline in demand for purchasing led to a steep supply curve for consuming
I'm wary of saying that a change in demand can change the supply curve. Do you mean the shift in demand moved the equilibrium to the steeper section of the supply curve?
If they had put more thought into their research, and looked at urban development trends prior to 1980, they should have concluded the following:
"Supply inelasticity induced by regulations on land use and mortgages has destroyed filtering effects for housing in many urban areas. The worst effects are in prominent cities with wealthy residents surrounded by politically active suburbs that further constrain housing production."
I admire the patience and tenacity Kevin Erdmann has shown in trying to tell the tale of US housing markets.
I will go out on a limb, and say "Yes, supply matters." It is a new idea I have, called "supply and demand."
Solutions?
Perhaps the US government should offer "bribes" for un-zoning or radical up-zoning of property in densely populated neighborhoods. Decriminalizing housing construction is a good idea.
That is, a city would get a large grant for up-zoning property in dense ZIP codes.
Another tactic is to pay a city, say, $200,000 for every housing unit built, if in a crowded zip code. That would be $100 billion a year, for 500,000 "extra" units.
By way of comparison, just the interest on the national debt today is near $1 trillion a year.
Yes, for 10% of the cost of interest on the national debt we might be able to have another 500,000 units a year. Worth a try.
I challenge the macroeconomic community to come up with better ideas.
The positive effect on real incomes and real income distribution would likely be high enough to justify such subsidies.
Figure 4 is really eye opening. It’s not showing a *shift* from affordable to higher-priced homes, it’s showing the complete evaporation of construction of homes affordable to first-time home buyers. Probably the most stark graphic representation that I’ve seen