Markets Where Land is Expensive Must Densify. Everywhere Else Must Bring Down Construction Costs.

This article is part of The Rooftop, a blog and multimedia series from New America’s Future of Land and Housing program. Featuring insights from experts across diverse fields, the series is a home for bold ideas to improve housing in the United States and globally.

________________________________________________________________________________________________

There is broad agreement that the cost of housing is one of the most urgent policy challenges in the U.S. However, housing markets differ significantly across the country and therefore the solutions that work in one housing market may not work in others.

In this article, I propose an organizing principle by which to determine which solutions are most effective in which market:

• In housing markets with expensive land, the most impactful way to bring down the cost of housing is to densify by building more units per acre.
• In the rest of the country, the most impactful way to bring down the cost of housing is to reduce construction costs.

This article outlines the rationale behind this organizing principle, and uses it to assess five popular housing affordability solutions.

Breaking Down the Price of a Home

The price of a home comprises the price of the land and the price of the physical structure. In a few (mostly-Coastal large) Metropolitan Statistical Areas (MSAs), the biggest price challenge is that land is expensive. Meanwhile, in most of the country, where land is relatively less expensive, it is escalating construction costs that result in expensive to build physical structures.

The price of the physical structure comprises the price of a new structure less the wear and tear (depreciation). Since buildings depreciate slowly, a well-kept older home won’t be dramatically cheaper than a new one in the same location, and the cost of building a new home acts as a market ceiling when land is cheap.

Consider a median-sized US house (about 1,800 square feet), not including the lot (land), with construction costs of $350,000. Consider places where land is relatively cheap, say, a lot costing $30,000. This results in a home (including the land and physical structure) of $380,000. In such markets, fitting two housing units on the lot instead of one (an enormous increase in density) leads to relatively low savings. The land, now $15,000 per unit, takes the home price from $380,000 to $365,000. By contrast, cutting construction costs by even 10 percent would decrease the house cost significantly more, to $345,000.

Contrast to places where land is expensive. Let’s say a lot costs $600,000, bringing the combined price of the same house in the previous example and land to about $950,000. In markets where land is expensive, splitting the cost of land by doubling the density reduces the cost by $300,000, while similar-sized construction cost improvements are considerably less valuable to the overall cost of the home.

These scenarios illustrate the need to consider housing proposals in a more disciplined and nuanced way. In places where land is expensive, we should look to bring down the cost of housing by building more units per acre. In most of the country, where land is inexpensive, we should bring down the cost of housing by lowering construction costs.

Where are places with expensive land located?

According to the U.S. Census, across the U.S., the average derived price of land under a house increased from about $84,000 in 2014 to about $109,000 in 2024 – a $25,000 increase. This suggests that most of the country is closer to the $30,000 lot, as opposed to the $600,000 lot.

In the Table below, I use data from Realtor.com to highlight metro areas in the U.S. where land is expensive. These are metro areas where the median price of a home less the imputed cost of new housing construction (not including land) is greater than $150,000.

Evaluating current housing proposals

These organizing principles should be used in assessing federal, state and local housing proposals to determine how much a proposal to lower house prices might help. In this section, I provide a high-level overview of five often-discussed proposals.

Cutting zoning and local regulations (often referred to as “pro-growth” or YIMBY proposals) – great for expensive cities, but might not move the needle in most of the country.

Reducing various zoning requirements is a crucial tool against the not-in-my-backyard (NIMBY) status quo; this includes minimum lot requirements, maximum lot density requirements, maximum building height, offsets between the building and the road, floor area ratio (FAR) requirements, parking lot requirements, environmental review requirements, and the required number of staircases in a multifamily building should have. Proposals floating around include both carrots and sticks to offer to municipalities and states.

Analyzing the data on housing unit density across the U.S. shows that even expensive cities (as defined in the table above) are not nearly dense enough, and that NIMBY resistance to densification imposes enormous costs. For example, Ventura and Snohomish Counties – suburbs of expensive Los Angeles and Seattle – are at about 0.25 housing units per acre, which is one-third of the density of even the U.S. overall larger-county average, or 0.75 housing units per acre. There is plenty of room to densify – Manhattan and Brooklyn lead the way with over 20 housing units per acre. However, easing zoning regulations and making multi-story, multi-family housing easier to build will make less difference where cheap land comprises only a small fraction of the home’s cost.

Accordingly, to bring costs down, federal carrots to reduce many of the local regulatory and zoning barriers should be offered only to cities where land is expensive, primarily large, coastal cities. That might not be feasible especially in the current political climate, suggesting that federal sticks might be more pragmatic.

Allowing Manufactured Housing (MH) and Accessory Dwelling Units (ADUs) by right could dramatically lower prices in much of the country, but the impact might be muted in expensive cities.

We keep looking for a breakthrough in lowering housing construction costs, but it’s been here for decades. Between lower square footage and lower per square feet construction cost, MH is that mythical cheap new entry-level home, with construction cost of $120,000 (a third of the price of site-built). In contrast to most zoning changes, allowing MH by right (allowing it as long as it complies with existing zoning and code regulations that do not discriminate against MH) could be a game-changer in lowering the cost of construction for much of the country.

Converting garages, basements, and sunrooms into ADUs is also considerably cheaper than new construction. However, financing for both MH and ADUs remains a problem (in addition to zoning). MHs are often confined to MH parks (often due to zoning), forcing the residents to rent the land underneath their home. Financing for such home-only loans is even more complicated. In addition, the homeowner not owning the land, combined with a very high cost of moving MH from one place to another (they are not particularly mobile after all) also gives the MH park owners (landowners) an incentive to increase land rents.

However, there is only a limited amount of single-family homes that might add an ADU, and this policy change will not make a large difference in densifying the large expensive coastal cities. In Los Angeles, even with the pent-up demand, the fewer than 50,000 ADUs permitted over the last eight years only takes the county’s density from about 1.415 to about 1.434 housing units per acre.

Low Income Housing Tax Credits (LIHTC) are an inefficient developer subsidy that is unlikely to help in most of the country, but might (inefficiently) counteract other policy failures in expensive cities.

LIHTC is a Federal program that subsidizes most or all construction cost (but not land) through tax credits that developers resell to investors, and is one of the largest Federal programs supporting affordable housing, costing taxpayers over $10 billion each year. LIHTC requirements often add to the cost of construction via prevailing wage, hiring consultants, and excess time. Because these costs are paid for by taxpayers, and not developers, it incentivizes further cost overruns. In most of the country, LIHTC is an expensive inefficient subsidy for renters – while some of the forgone tax dollars lowers rents, it is at the expense of a significant fraction of these tax dollars staying with investors and developers, and paying for cost overruns. Moreover, research suggests that LIHTC construction often cannibalizes construction that would have happened even without LIHTC.

It’s possible that LIHTC results in more units in expensive cities, due to zoning relaxations that often accompany LIHTC, and the cost of construction being covered by taxpayers. Developers (and taxpayers) should not disregard that taller buildings cost more or that larger units are more valuable – building in a more expensive way unnecessarily increases home prices. But when most buildings are forced to be less dense than efficient by regulations, the LIHTC inefficiency going the other way might cancel some of these broader issues out. In other words, when zoning constraints force too few units per acre, while occasionally LIHTC results in too many units per acre, then LIHTC might help to bring the average closer to the efficient outcome, but at a great expense to the taxpayers. A much more effective solution is lowering zoning and regulatory barriers for all developers and construction.

Affordability and inclusionary zoning (IZ) requirements are unlikely to be effective anywhere.

Affordability and IZ-style requirements (for example, some fraction of the units in a new building has to be affordable to lower-income renters) do not immediately add to or lower the construction cost, nor do they result in more housing units (unless paired with relaxing zoning). Accordingly, they are not particularly helpful in either kind of housing market, but they are also not dramatically harmful – this transfer to lower-income renters is not nearly as inefficient as LIHTC. But since landlords are paying for it (through forgone rent), that might change what kinds of projects pencil out for developers, resulting in fewer units built.

Expensive cities should relax zoning restrictions instead for all construction. But even simple sale of perks might be more efficient than IZ – for example, the city allowing developers to create fewer parking spots in exchange for cash that the city can use for housing vouchers.

Banning large investors from owning single-family homes does not densify and does not decrease the cost of construction.

Large corporations that buy single family homes to rent out – these homes are still an important source of housing supply for tenants, and account for a vanishing fraction of the U.S.’s housing stock. . Further, large (“mega”) investors are more active in the Southeast and Midwest – where land is cheap, and forcing investors to finance condo units is likely counterproductive.

A more apt question for policymakers is why aren’t households who could otherwise afford to buy single-family homes instead renting? There are millions of renters who should be able to qualify for a mortgage, want to own, but are still renting, raising questions about whether our mortgage standards are too tight or are perceived as being too tight.

_____________________________________________________________________________________________

Editor’s note: The views expressed in the articles on The Rooftop are those of the authors alone and do not necessarily reflect the opinions or policy positions of New America.