Resiliency for Renters

In the wake of the coronavirus pandemic, Massachusetts state and municipal policymakers have acted quickly. By March 10, a state of emergency  had been declared . On March 17,  public schools closed . On March 23, Governor Charlie Baker  issued the state's first stay-at-home advisory  and ordered the shutdown of all non-essential businesses.

As stay-at-home measures and job losses persist, however, housing insecurity will become a growing problem. It's clear that government officials recognize this. On April 20, Massachusetts finalized  a moratorium on evictions and foreclosures  which will last six months or 45 days after the current state of emergency ends, whichever comes first. In addition, the City of Boston's Department of Neighborhood Development created a  Rental Relief Fund , which has already distributed an initial round of aid to renters in the city.

These measures provide important safeguards for people impacted by COVID-19, yet there is no guarantee that they'll reach those who need help the most.

We've already seen that the virus has struck some populations, such as  Black and Latino communities in New York City , harder than others. In addition, the work of scholars such as Matthew Desmond–author of  Evicted –has shown that one forced displacement can set off a chain of events that leads to long-term housing and economic insecurity for families. In the time of coronavirus, becoming homeless could have  even greater repercussions  in terms of health risks.

With this in mind, Boston's government should consider targeting vulnerable renter households with outreach material about the various forms of aid and protections available to them. Otherwise, despite its efforts to contain and address the pandemic's fallout, the city might leave some of its most at-risk residents behind.

Many of the data sources needed for this type of assessment are already available. In 2016, the city's  Climate Ready Boston  initiative created a standard by which to assess social vulnerability. Similar to the U.S. Census' definition of  Environmental Justice Communities , this measure included the number of older adults, children, people of color, people with limited English-language proficiency, people with low or no income, people with disabilities, and cases of medical illness across the city.

Climate Ready Boston researchers used these factors to measure relative  susceptibility  to "chronic extreme heat," "frequent stormwater flooding," and "acute and chronic coastal and riverine flooding." Coincidentally (or not), many of these risk factors also apply to the coronavirus pandemic.

The  CDC  states that "Older adults and people who have severe underlying medical conditions... seem to be at higher risk for developing more serious complications from COVID-19 illness." The link between race and coronavirus fatality rate has already been pointed out above. And households with limited English-language proficiency, low or no income, members with disabilities, or children–at a time when child care services have been severely curtailed–may face economic barriers as well as lack of access to public resources.

For these reasons, I used Climate Ready Boston's index to measure the vulnerability of households in Boston. Below, I have replicated their risk assessment maps using data from U.S. Census surveys (2018) and the  CDC  (2016). Where possible, I tried to narrow the scope to renter households only. Out of a total seven factors, I created a social vulnerability index where five coronavirus-related measures–age, limited English proficiency, very low income, race, and poor physical health–are weighted more heavily than the other two–households with minors and people with disabilities. All were standardized based on the number of individuals or households in each given spatial unit.

To assess more renter-specific risk, I created another index that was mainly based on the percentage of households which are both low-income and cost-burdened. The federal Department of Housing and Urban Development (HUD)  defines the latter term as  those who spend over 30% of their income on housing; individuals or families who spend over 50% of their income are on rent are "severely" cost-burdened.

To supplement these figures, I included local eviction filing rates as well as the number of affordable rental units in each area. I drew this data from Matthew Desmond's  Eviction Lab  and the  City of Boston , respectively. Lastly, I referenced a list of the City's housing-related  public service requests  in 2020, which could indicate where residential buildings are in relatively poor condition. Where applicable, I standardized these factors based on the total number of households or housing units in each area.

I then combined these two maps–social vulnerability and housing insecurity–into a single index. To make up for missing data in some census tracts and block groups, I used areal interpolation to create new estimates for these places. My results are shown below.

Showing where renters will be hit the hardest is only the first step towards a solution. After singling out the most vulnerable areas from the composite map above, I checked which zip codes they belonged to (see below). This information could help the City of Boston organize targeted mass mailings to those most in need of aid. Rental assistance resources can be found online, but not everyone has access to technology at home or the ability to use it. In addition, at least one study  has shown  that increasing outreach to subsidy-eligible households using methods such as mailing information and providing application assistance increases enrollment in these programs.

"Snail mail" is one of many ways to spread the word about rental assistance. Another method is posting flyers, which community-organized groups like  mutual aid networks  have already been doing in Greater Boston. In addition, once public health conditions have improved, it might be feasible to mobilize volunteers for door-to-door visits. These options may help make up for what automated calls and online publicity cannot achieve alone.

Many improvements are possible for the mapping I've done thus far. As shown above, missing information for some areas left literal gaps in the data. In addition, the weighting system and factors are based on publicly accessible information sources, as well as my own perception of various measures' relative importance. These could be strengthened with other forms of data collection and rigorous research, respectively. Finally, the conversions and computations I used to create my maps incorporated uncertainty into multiple parts of the process.

Of course, uncertainty seems to be characteristic of this pandemic period. This highlights the vital importance of measuring health and economic impact during a rapidly-evolving crisis. In order to ensure that post-disaster aid reaches those who need it the most, we first need a better picture of what's happening around us.

This project would not have been possible without the help of various faculty and students of Tufts University's Urban and Environmental Policy and Planning department. Thanks to Sumeeta Srinivasan, Carolyn Talmadge, Megan Morrow, and Brian Froeb in particular.