There's a strange academic debate about whether the market price of homes should equal their marginal construction cost. Here's why that idea doesn't make sense.
Sounds like you think it's impossible to operationalise the concept of marginal cost of supply in any diagnostic test of the competitiveness of housing markets, and how regulations vis-a-vis the natural monopoly in land contribute to that.
That seems quite significant.
To paraphrase you, cost data can be used at best to measure the marginal cost of production, not supply, because the economic (i.e. opportunity) cost of supply includes not just production costs but the value of foregone options. (For infill I'd note this includes not only the pure value of delay but the value of extant capital lost to premature redevelopment). And even then, how you measure the marginal cost of production changes the result.
And as a practical matter, even if these problems were avoidable, any test could only apply to brand new buildings, since the optimal density evolves.
"cost data can be used at best to measure the marginal cost of production, not supply, because the economic (i.e. opportunity) cost of supply includes not just production costs but the value of foregone options. (For infill I'd note this includes not only the pure value of delay but the value of extant capital lost to premature redevelopment). And even then, how you measure the marginal cost of production changes the result."
So yeah, that seems a reasonable summary of the third point of the article - the "true marginal cost" problem. However, it is tricky to explain without using the word "supply"! One way of explaining the point is that we are measuring the margin construction cost in the dimension of space, but for property owners their marginal cost is in the dimension of time.
But even as an argument about density it faces problems with interpretation, as no one know which margin they expect to be equated with price (problem 1) or at which density the marginal cost should apply (problem 2).
When someone wants to buy a house, they calculate the amount of money they need and then seek out financing to find a home that fits their budget. In essence, the housing market is not really about housing, but rather about money. In the realm of money markets, the cost of obtaining financing has no direct correlation to the cost of a house let alone the marginal cost (whatever that might be).
Is your basic point that academic analysis fails to explain what developers do because it uses the a wrong version of marginal cost? Marginal in space v marginal in time. Does this criticism get combined with the general notion that house prices are being analyzed as consumer goods rather than portfolio assets? Or it is a facet of that point?
In California there has been a steady stream of deregulation from the State level, say since 2017, and even newspapers are noticing it has done very little to "move the needle" on housing production. Would California's deregulatory failure begin to disprove the theory?
What about externalities? The whole debate over "costs" reduces it to construction costs. I was a decision maker keenly aware of external costs from new development on incumbents, particularly as they impacted city service levels and other hard and soft measures of livability. Some regulatory costs are imposed to recover those externalized costs needed to restore service levels and other real metrics of livability. (e.g. schools, public safety, climate change, congestion, open space) The real costs of congestion externalities might increase geometrically. Are any of these costs included in this kind of analysis?
"Is your basic point that academic analysis fails to explain what developers do because it uses the a wrong version of marginal cost? Marginal in space v marginal in time"
Yes. This is the third point. The first two are criticisms that apply even if you assume that the marginal cost of density is the "right" way to think about it.
"Does this criticism get combined with the general notion that house prices are being analyzed as consumer goods rather than portfolio assets? Or it is a facet of that point?"
Yes. Building new housing is an asset swap - a portfolio change. The appropriate way to analyse such a choice is about relative rates of return, not marginal production cost of density. These points are related.
Yes, I also think California's attempts to deregulate that don't seem to be changing housing production demonstrate that density regulations are not constraining the rate of new housing production. I wrote about that here:
Externalities I haven't considered, but are a major justification for planning/coordinating city growth. Having new housing happen at predictable locations helps know where to build public services and infrastructure.
I would say this rings true, but it also seems more complicated than it needs to be. The detailed analysis is definitely interesting at the least. I think the "portfolio" framing is the correct way to look at this.
The frustrating part of housing prices is you don't see much potential for mass production and marketing. Walmart doesn't build houses. There is little ability for a corporation to dominate the market by offering lower prices and cutting costs, the way that would happen for consumer goods.
While marginal costs may not determine the asking price, I definitely think much of the burden of housing affordability would be alleviated by making sure low cost options exist. The housing market is not homogenous and it creates significant problems when new construction only targets expensive upscale options, even if that eases price pressure in older developed areas. Definitely some important issues here with good points as well.
"you don't see much potential for mass production and marketing"
Sure, there is scope for.this, but who has the incentive to do it? What's stopping them now?
And in terms of making homes cheaper via changes in production technology, no, I don't see a big effect here. Sure, encourage it for overall efficiency reasons. But the market price of living at a location is not determined by the cost of building at that location (see second point in this post).
I'm not sure I agree with this either, as prices are the summation of costs.
"While marginal costs may not determine the asking price, I definitely think much of the burden of housing affordability would be alleviated by making sure low cost options exist. "
You can build cheap low-spec dwellings anywhere you want. But doing this in a high value area gives away the opportunity for more gains to property owners by choosing a higher quality and more expensive options (essentially the second point in this article).
Sounds like you think it's impossible to operationalise the concept of marginal cost of supply in any diagnostic test of the competitiveness of housing markets, and how regulations vis-a-vis the natural monopoly in land contribute to that.
That seems quite significant.
To paraphrase you, cost data can be used at best to measure the marginal cost of production, not supply, because the economic (i.e. opportunity) cost of supply includes not just production costs but the value of foregone options. (For infill I'd note this includes not only the pure value of delay but the value of extant capital lost to premature redevelopment). And even then, how you measure the marginal cost of production changes the result.
And as a practical matter, even if these problems were avoidable, any test could only apply to brand new buildings, since the optimal density evolves.
Sound about right?
"cost data can be used at best to measure the marginal cost of production, not supply, because the economic (i.e. opportunity) cost of supply includes not just production costs but the value of foregone options. (For infill I'd note this includes not only the pure value of delay but the value of extant capital lost to premature redevelopment). And even then, how you measure the marginal cost of production changes the result."
So yeah, that seems a reasonable summary of the third point of the article - the "true marginal cost" problem. However, it is tricky to explain without using the word "supply"! One way of explaining the point is that we are measuring the margin construction cost in the dimension of space, but for property owners their marginal cost is in the dimension of time.
But even as an argument about density it faces problems with interpretation, as no one know which margin they expect to be equated with price (problem 1) or at which density the marginal cost should apply (problem 2).
When someone wants to buy a house, they calculate the amount of money they need and then seek out financing to find a home that fits their budget. In essence, the housing market is not really about housing, but rather about money. In the realm of money markets, the cost of obtaining financing has no direct correlation to the cost of a house let alone the marginal cost (whatever that might be).
Is your basic point that academic analysis fails to explain what developers do because it uses the a wrong version of marginal cost? Marginal in space v marginal in time. Does this criticism get combined with the general notion that house prices are being analyzed as consumer goods rather than portfolio assets? Or it is a facet of that point?
In California there has been a steady stream of deregulation from the State level, say since 2017, and even newspapers are noticing it has done very little to "move the needle" on housing production. Would California's deregulatory failure begin to disprove the theory?
What about externalities? The whole debate over "costs" reduces it to construction costs. I was a decision maker keenly aware of external costs from new development on incumbents, particularly as they impacted city service levels and other hard and soft measures of livability. Some regulatory costs are imposed to recover those externalized costs needed to restore service levels and other real metrics of livability. (e.g. schools, public safety, climate change, congestion, open space) The real costs of congestion externalities might increase geometrically. Are any of these costs included in this kind of analysis?
Continuing thanks for your work.
"Is your basic point that academic analysis fails to explain what developers do because it uses the a wrong version of marginal cost? Marginal in space v marginal in time"
Yes. This is the third point. The first two are criticisms that apply even if you assume that the marginal cost of density is the "right" way to think about it.
"Does this criticism get combined with the general notion that house prices are being analyzed as consumer goods rather than portfolio assets? Or it is a facet of that point?"
Yes. Building new housing is an asset swap - a portfolio change. The appropriate way to analyse such a choice is about relative rates of return, not marginal production cost of density. These points are related.
Yes, I also think California's attempts to deregulate that don't seem to be changing housing production demonstrate that density regulations are not constraining the rate of new housing production. I wrote about that here:
https://www.fresheconomicthinking.com/p/upzoning-predictions-versus-reality
Externalities I haven't considered, but are a major justification for planning/coordinating city growth. Having new housing happen at predictable locations helps know where to build public services and infrastructure.
I would say this rings true, but it also seems more complicated than it needs to be. The detailed analysis is definitely interesting at the least. I think the "portfolio" framing is the correct way to look at this.
The frustrating part of housing prices is you don't see much potential for mass production and marketing. Walmart doesn't build houses. There is little ability for a corporation to dominate the market by offering lower prices and cutting costs, the way that would happen for consumer goods.
While marginal costs may not determine the asking price, I definitely think much of the burden of housing affordability would be alleviated by making sure low cost options exist. The housing market is not homogenous and it creates significant problems when new construction only targets expensive upscale options, even if that eases price pressure in older developed areas. Definitely some important issues here with good points as well.
"you don't see much potential for mass production and marketing"
Sure, there is scope for.this, but who has the incentive to do it? What's stopping them now?
And in terms of making homes cheaper via changes in production technology, no, I don't see a big effect here. Sure, encourage it for overall efficiency reasons. But the market price of living at a location is not determined by the cost of building at that location (see second point in this post).
I'm not sure I agree with this either, as prices are the summation of costs.
"While marginal costs may not determine the asking price, I definitely think much of the burden of housing affordability would be alleviated by making sure low cost options exist. "
You can build cheap low-spec dwellings anywhere you want. But doing this in a high value area gives away the opportunity for more gains to property owners by choosing a higher quality and more expensive options (essentially the second point in this article).
Cameron, you will love this….
https://josephnoelwalker.com/bryan-caplan-155/
Not sure that Joe Walker is convinced either.
Definitely not. I’ve chatted to Joe before about this supply arguments on his podcast https://josephnoelwalker.com/96-the-housing-supply-myth-cameron-murray-ian-mulheirn/