Addressing Constrained Housing Supply: Evidence on New Market-Rate Housing

It’s no secret that the United States has a housing affordability problem. The high cost of living in major cities stems from a shortage of available units. High rents burden low- to moderate-income families. The larger percentage of a household’s income goes to rent, the less a household can afford to spend on basic and necessary goods and services like energy, food, and medicine.

The NIMBY (“Not in My Back Yard”) attitude toward land use governance has some culpability in this affordability crisis. Restrictive zoning codes often discourage or exclude the construction of multifamily housing in favor of single-family homes on large lots, reducing supply and driving up prices as a result. Zoning restrictions are not only limited to the number of units a developer can build on a property, but also include height limits, lot setback requirements, and parking spot requirements.

Conversely, the YIMBY (“Yes in My Back Yard”) approach values a variety of government policies to increase the supply of affordable housing, such as the construction of public housing, housing vouchers, and inclusionary zoning requirements (in which a developer must set aside a portion of affordable or below market-rate units). A market-based solution might be to remove or change local zoning laws to allow for denser housing through mechanisms such as multifamily buildings, smaller lot sizes, accessory dwelling units, and minimized parking requirements.

The way forward on housing affordability will likely be a mixed approach using a variety of these interventions. Skeptics of the market-based or upzoning approach are concerned with the possibility of low-income properties and renters being affected by market-rate housing. The lack of market-rate housing, however, is what might be driving up rents in the first place.

Consider this scenario: as high-income renters move into a city and that region has a shortage of market-rate housing, those same renters might outbid lower income renters for below-market rental properties. The demand for market-rate housing exists, but the supply does not. Rents increase across the city as a result of this demand-side pressure on a constrained housing supply.

A recent study by economist Evan Mast tests how additional market-rate multifamily buildings in a city affect the demand for different housing submarkets.

Mast’s approach uses a migration chain theory of housing, in which higher income households move into new market-rate housing, lessening the demand for their previous unit. This allows a lower income household to occupy the vacated property at a lower rent. The chain theory continues for the lower income household’s previous unit as well and may proceed for a few more iterations. As this happens, the city’s housing market adjusts and reaches a new equilibrium of supply and demand.

“Prior residence of tenants of new buildings in Chicago. Note: Solid red dots represent the location of market-rate multifamily buildings completed between 2010 and 2017. Hollow black dots represent the previous residences of the current tenants in those buildings. The base map polygons are zip codes in Chicago proper, colored according to median household income in the 2013–2017 American Community Survey. Only residents whose prior residence was within the city proper are included. Small amounts of noise are added to each marker to avoid precisely identifying addresses.” (Mast 2021)

Mast’s data comprise of 686 market-rate multifamily buildings constructed since 2009 in 12 major cities, totaling over 52,000 individuals. Individual address histories are compiled and verified from a variety of sources, such as ​​USPS change of address forms, county assessors, and magazine subscriptions.

Using this dataset, Mast analyzes the chain of migration resulting from market-rate housing construction in two ways:

First, the author identifies the tenants currently living in the multifamily buildings and matches them to their previous address, repeating this process for six rounds. Through this method, Mast finds that a large proportion of residents in the new buildings come from high-income areas. Only 20% of new residents moving from within the metropolitan area are from below-median income tracts, but this number rises to 40% by round six of the chain migration analysis. This method also demonstrates the geographic impact of adding new market-rate housing, as the affected households in the sixth round include similar shares of suburban and metropolitan residents.

Second, Mast runs a range of simulations with varying assumptions to estimate the unit-equivalent effect of new market-rate housing on low- and moderate-income markets. The baseline simulation assumes a marginal increase in supply and allows for vacancies which end the migration chain. These vacancies might be a result of matching frictions or landlord market power. The baseline model does not account for out-of-metro migration or new household formation. In this model, the building of 100 new market-rate units is equivalent to the creation of 70 depreciated units in lower income tracts.

To capture a larger supply increase, assumptions about out-of-metro migration and new household formation are added to the model. In this scenario, the building of 100 new units is equivalent to the creation of 45 depreciated units in lower income tracts. These effects occur within five years, with most happening within three years.

Together, Mast’s analyses of these chains of migration as a result of new multifamily housing suggest that dense, market-rate housing alleviates the pressure placed on low-income housing markets. These higher income households that can afford market-rate housing benefit from fully meeting their preferences by renting new housing, while the low-income households benefit from lower rents as a result of lessened demand at their level of the housing market.

“Percent of individuals originating in below-median income tracts versus tract income of new building. Note: This figure plots the share of origin sequence units in below-median income tracts against the characteristics of the new building’s tract. Each line represents a different round of the sequence. New buildings are separated based on the income decile of their tract.” (Mast 2021)

The method used by Mast to study migration chains could be applied in future research to a specific market-based policy change, such as upzoning. This would be one way of assessing the effects of a natural policy experiment on different housing submarkets.

The author notes a few limitations with their study. The paper focuses on migratory effects within the housing market between different income levels but does not measure exact price changes as a result of the new market-rate multifamily housing. Price changes are an additional way of measuring the effects of zoning changes and housing policies and could be explored in separate studies. However, as Mast points out, some neighborhoods may have rents that are close to or at operating costs, allowing little flexibility for reducing price levels as a result of lessened demand.

If the creation of multifamily housing results in the availability of a significant number of low-income units as this paper suggests, then policies that are friendly to the construction of dense, market-rate housing will help alleviate pressure on the housing market across the board. These market-based solutions include zoning changes that allow for multifamily buildings, smaller lot sizes, accessory dwelling units, and decreased parking requirements.

Although pushback from local homeowners and the resulting vetocracy from loud, concentrated interests may seem like a major obstacle to overcome for proponents of increased housing supply, including through market-based reform, there has been progress in select states and localities. California, for example, recently passed a bill that allows property owners in most parts of the state to develop duplexes in areas that were previously zoned for single-family homes. Connecticut, meanwhile, legalized accessory dwelling units and put limits on parking mandates. These changes, while not necessarily the scale of the multifamily buildings in Mast’s study, provide hope for the politics of supply-friendly zoning policies.

While it may be too early to assess fully the effects of these upzoning efforts in California, Connecticut, and elsewhere, the evidence presented by Mast’s research suggests that these policies have the potential to increase housing supply and, as a result, housing affordability.