Our new report, "Evictions in Boston: The Disproportionate Effects of Forced Moves on Communities of Color"

For over 18 months we partnered with researchers at Massachusetts Institute of Technology (MIT) to look at evictions filed in Boston Housing Court. We zeroed in especially on evictions in private-market (unsubsidized) housing. We also looked at Boston eviction filings during the COVID-19 pandemic (before they were banned in April 2020). What neighborhoods are most impacted? Who lives in these frontline communities?

DOWNLOAD the full report here: bostonevictions.org.

Here are our top findings:

  • Eviction filings in market-rate rental housing are disproportionately occuring in Boston’s communities of color. Over 2/3 (70%) of market-rate eviction filings are in census tracts where the majority of residents are people of color (even though only about half of the city’s rental housing is in these areas).

  • These disparate effects are magnified in predominantly Black communities, where market-rate eviction filings are two times more common than if these filings were equally distributed across the city. Over 1/3 (37%) of market-rate eviction filings occur in neighborhoods in which a majority of residents are Black (though only 18% of rental housing is in these neighborhoods).

  • We see similar racially disproportionate trends in evictions today, in a context where many more families are at risk of losing their home. During the COVID-19 pandemic, over 3/4 (78%) of all evictions filed in Boston during the pandemic (from the first outbreak to the moment evictions were banned state-wide, a 7-week period) are in census tracts where the majority of residents are people of color.

  • These trends are related to neighborhood poverty and income levels, but are more closely correlated with the racial composition of a neighborhood than other socioeconomic characteristics. Market-rate eviction filings are more likely to occur in census tracts where there’s a larger share of Black renters, controlling for other variables in predictive statistical models, including rent burden, median household income, and poverty rate.


Read the story in The Boston Globe and The Associated Press