This chart is a big freakin’ deal
Why zoning reform is not (on its own) going to make housing affordable
What if Zoning Reform Isn’t Enough?
In the past few years especially, cities across the country have seen surging housing prices–and not just in San Francisco, LA, Boston, and New York–but in previously affordable places like Spokane, WA, Providence, RI, and San Antonio, TX. Land use regulations like zoning have played a big roll in driving up costs in many cities and regions by sharply limiting the availability of land for new development. The growing Yes In My Back Yard (YIMBY) movement is bent on reforming the zoning regulations to allow for more building, with the explicit goal of lowering housing costs.
But what if zoning reform isn’t enough? A recent paper by housing researcher Yonah Freemark took a large sample of zoning changes and tried to find a relationship with changes in the housing market.1 From the abstract:
“Evidence indicates that upzonings offer mixed success in terms of housing production, reduced costs, and social integration in impacted neighborhoods; outcomes depend on market demand, local context, housing types, and timing. Research on regional upzoning impacts is nascent but outcomes appear positive. Downzonings limit construction and worsen affordability.”
This finding became an important touchstone in a recent piece by Daniel Herriges at Strong Towns, Upzonings Might Not Lower Housing Costs. Do it Anyway. Daniel’s piece is great, go read it now. I want to build on it, by talking more specifically about the many constraints on lowering the cost of housing.
Binding Constraints
A ‘binding constraint’ is a concept from economics. Loosely put: there are many constraints on what you can do. In real estate some examples include: land, access to capital, construction costs, firm capacity, operating costs, etc. At any given time, one of those will be “binding” meaning that it’s your biggest problem, the thing that’s limiting your ability to do more.
The San Francisco Bay Area is probably a good example of a place where zoning is the binding constraint. There are vast parts of the metro region that have been off limits for anything more than a single-unit detached home.2 You might have capital to build more, or a great design idea–but zoning is the binding constraint on what you can build.
But what would happen if the zoning constraint were removed? Suddenly more housing would be possible on many more lots.3 However, having cleared away one binding constraint, we will find ourselves facing a new binding constraint that had been, as it were, ‘lurking in the wings.’
Construction Costs
Another binding constraint you could run into is construction costs. Indeed, today in many parts of the country where construction costs are high and/or incomes are low, it can be hard to make new construction pencil out: the finished project costs more than buyers or tenants can afford to or are willing to pay.
‘Hard costs’ is the real estate term for the actual construction part of the budget–materials, labor, etc. In a typical new-construction project, hard costs will be between 70-80% of the total development cost. Everything else (land, permitting, architecture, carrying costs, financing, and the developer’s fee) fits in that other 20-30%. For this reason, construction costs are big constraint in the feasibility of new development.
In the past few years, since the initial disruptions of the construction industry at the start of the Covid-19 Pandemic hard costs have increased a lot, in some cases as much as 50%--and some materials or trades even more than that. As a result, overall construction costs are much higher. A recent estimate from our typical lowest cost builder was a minimum of $325 per square foot for a standard wood-frame, 4-story apartment building with elevator, sprinklers, etc. – no frills.
That $325 per square foot means that if everything else were free: free land, no permits, no architecture, no utilities, insurance, etc. etc. then if you build a 1,000 square foot 2-bedroom apartment, it would have to be priced at at least $325,000. If you add in all those other costs, it’s likely to be $375-425k for a new home–something that would be “affordable” only for households making well over $100,000 per year, or more depending on interest rates and other costs.
In the 1950’s, Levittown NY was selling houses (at a profit) for $100,000 (in 2022 dollars), or roughly “affordable” to a household making about $40,000 in today’s dollars. (Why is New Housing So Expensive?) Things have changed quite a bit!
Does Construction Ever Get Cheaper?
Enter Brian Potter and his excellent substack Construction Physics. A few months ago, Brian set out to answer the question: Does construction ever get cheaper?
The short answer: Nope.
After reviewing multiple indices of construction costs, across several industries, Brian generates the following chart.
The post goes through several different analyses and ways of looking at the data and arrives at the long answer: “So going back to 1915, the cost of constructing buildings, as measured by various cost indexes, has almost always risen faster than inflation.” Construction cost indices really start to diverge from CPI (measuring general inflation) in the 1960’s and 70’s and have shot up since the 2000’s. Construction costs had a valley after homebuilding crased in the great recession of 2007-2011, but then shot right back up. Even the great depression only saw costs dip and plateau, keeping pace with CPI.
“Building material cost indexes mostly seem to rise at or below the level of overall inflation, suggesting the problem isn’t that physical materials are getting more expensive. Increases in labor costs could plausibly account for it in most periods, consistent with construction labor productivity failing to improve.
Troublingly, over the past several decades years or so, cost trends seem to be headed in an even worse direction. Construction costs now seem to be rising well above the rate of wage inflation, and building material costs are rising faster than inflation (and this is especially true of the last 2 years.)”
What If Construction Never Gets Cheaper?
The price of a new home matters for more than just the immediate buyer of that new home. Each new home built is filled with a household that is moving from somewhere, leaving behind a vacant unit or a less-crowded household. The newly vacated unit prompts another household to move in, and so forth. This “chain of moves” can connect even upscale buildings with low-income neighborhoods in even just a few steps (Mast 2019).
But the scale of new construction is limited by the number of households who can make enough money to move in, and in housing markets where the market-price of homes is lower than the price of newly built homes new construction slows to a trickle. The competitor for the sellers of existing homes… is new homes being built to satisfy a demand. Without new homes, owners of the existing homes gain power in the market for housing.
The higher construction costs go, relative to income gains, the fewer families that will be able to afford new construction homes. As the market size for new construction homes shrinks, the market will be able to build fewer homes. Essentially building to the still profitable luxury end of the spectrum, while unable to provide economy homes–not just for poor households, but even upper middle class.
If we want to build abundant housing for all its manifest advantages, we need to figure out how to do something never achieved in the 20th century–bring down the cost of construction sustainable. As to how to do that, I don’t think anyone has a sure ensure. But I have some ideas I’ll explore in a future post.
Please share your ideas in the comments or on social media!
Seth
There are many caveats with this study, including that most zoning changes are small, and that the study couldn’t quantify the relative ‘scale’ of a zoning change.
This is changing because of state preemption!
Probably increasing land prices, but lowering per-unit land prices
Cool substack!
A lot of folks are talking about allowing for single-stair layouts, these days! This is more of a building code reform, but it could serve to bring down construction costs.
To Sukanya's point, modular can be helpful, less to bring down costs per se (my experience is that savings are minimal) but to reduce construction time, which ends up translating into savings.
Curious about the scale of opportunity that modular housing and off-site manufacturing presents to reduce construction costs in atleast large metropolitan areas.