Our housing system functions at various scales, but there’s often a significant mismatch between where housing policy is made and where housing markets function. After all, most people don’t choose where to work or live based on the boundaries of a local government, yet housing policy is so intertwined with local zoning and subsidies (including federal subsidies that flow through to local bodies), that much of the focus of housing policy, research, and advocacy is on local jurisdictions. And though central cities are often the focus of affordability policies and investments, suburbs are often larger and have more opportunities for economic mobility and so have a critical role to play in a regional market’s housing affordability.

A newly updated forecasting model created by a team at the Urban Institute uses the latest in data science to account for this mismatch. It was built for the metropolitan Washington, DC, region but also allows for analysis at the various jurisdictional-level governments and even the 40 subjurisdictional areas (called public use microdata areas, or PUMAs). The model forecasts the effects of policy interventions, such as vouchers, land-use reform, and production subsidies on housing availability, affordability, segregation, and sprawl at the PUMA level.

As former director of the DC Office of Planning, I have seen firsthand how much local policymaking on housing affordability is based on theoretical mental models and ideology. This model, which is based on existing zoning as well as demographic data and forecasts, represents a significant advancement for housing researchers, advocates, and policymakers by providing a regionally specific, forward-looking housing tool that allows for analysis within jurisdictions.

Why understanding prospective policy changes within a jurisdiction and across a region is so important

One question the analysis asks is, What if the federal government were to convert the housing voucher program to an entitlement from a program based on annual budgeting?

The model shows that all the households with the lowest incomes would then have housing. But in our existing housing market, in which housing growth is not abundant and strictly limited by zoning (PDF), the units voucher holders find are those that that would have been occupied by households with slightly higher incomes without the additional vouchers. As a result, those households would be outcompeted for the limited housing and would move much further out in the region.

Yet, if land-use regulations were simultaneously reformed to allow for more production, then many more of those households could be accommodated without causing the same level of displacement or sprawl. In this way, the model can help policymakers think through various intended and unintended consequences.

In addition to these regional outcomes of potential federal housing changes, the model can surface how one jurisdiction’s decision whether to invest in vouchers or production subsidies may affect its overall goals. Given that there are three state-level governments in the DC region, this model can also show how different state policies could affect regional housing.

Future applications for the model       

As the model is refined, it could be especially helpful in advancing housing discourse and policymaking around improving equitable outcomes. It can help not only policymakers and researchers but also housing advocates, activists, and residents seeking to ensure their communities help the region address its affordability challenges and become more equitable.

Based on my experiences working on these issues, I’d be interested in exploring the following potential scenarios based on future resources:

• What would happen to affordability and segregation if DC could produce a magnitude more housing and affordable housing in Upper Northwest, for example by legalizing buildings in Friendship Heights that are as tall as those are across the street in Maryland?
• What would happen if Maryland implemented a minimum zoning requirement of at least quadplexes in areas of high opportunity?
• How could a significant increase in inclusionary zoning requirements in Loudoun County affect affordability?
• How could a significant increase in state-level vouchers affect Virginia’s jurisdictions?

Housing policies discussions can sometimes revert to ideological or emotional debate because of a lack of rigorous quantitative modeling and shared understanding of historical inequities. This tool is exciting because its versatility and ability to work at different scales vastly improve the ability of evidence-based policymaking, in the DC region and potentially other regions across the country. I only wish it had been around when I was working on these issues in the region.