At the same time, those counties have increased the rate at which they sentence women to jail.

This association, according to new research from the 91Ě˝»¨, indicates that monetary sanctions, also known as legal financial obligations or LFOs, have far-reaching social, economic and punitive effects. In other words, what may seem like a system of low-level penalties aimed at individuals actually affects whole communities.

“Here in Washington state, men’s incarceration rates have been trending downward for over a decade whereas women’s incarceration rates have continued to increase,” said lead author , a 91Ě˝»¨postdoctoral researcher in sociology. “This paper suggests this is because women have not benefitted from the legal system’s shift away from carceral sentencing toward monetary sanction sentencing in the same ways men have benefitted.”

The is part of a volume of research on legal financial obligations, published online in January in The Russell Sage Foundation Journal of the Social Sciences. , professor of sociology at the UW, spearheaded the multi-institutional study of trends and practices in eight states over five years relating to legal fines and fees.

More about the larger eight-state study, and Harris’ role in that research, is here.

The UW-led study focused on Washington state, and looked at the connection between a defendant’s gender and fines and fees.

Over the past two decades, women’s incarceration rates have remained steady or increased nationwide. The 91Ě˝»¨researchers wanted to use Washington state data to explore what factors may contribute to this trend, and specifically, whether the expanding system of monetary sanctions could be to blame.

The authors, including sociology graduate students and , used county-level statistics from the Washington Administrative Office of the Courts from 2007-2012, as well as county budget information from the Washington State Auditor and demographic data from the U.S. Census Bureau.

In the United States, , and people who are poor are disproportionately impacted not only by the legal system generally, but by monetary sanctions in particular.

Monetary sanctions, meanwhile, have contributed in communities around the country. The 91Ě˝»¨study found that, among Washington’s 39 counties, a 1% increase in county revenue from monetary sanctions was, on average, associated with a 23% increase in the incarceration rate of women. That may be due, researchers said, to increased law enforcement around the types of lower-level offenses women are more likely to engage in than men, and to the possibility that women, due to financial precarity, are forced to opt for incarceration because of their inability to pay fines and fees.

“Not only are women are going to find financial sentences more burdensome than men, they are also more likely to commit the types of crime that make them eligible for monetary sanctions instead of incarceration,” O’Neill said. “Among people who commit crime, women are disproportionately represented among misdemeanor offenders, whereas men commit more felonies. So, women who offend may be more likely to be sentenced to monetary sanctions and less likely to be able to pay than men who offend.”

Given the association the study found between revenues from monetary sanctions and the sentencing of women to incarceration in Washington counties, the researchers suggest that governments look for other revenue sources, and for ways to reduce the costs of their justice systems. The authors point to a 2019 on the spatial distribution of monetary sanction debt and revenue that suggests these findings could apply outside of Washington state as well.

“Heavy reliance on monetary sanctions as a source of revenue creates an obvious conflict of interest for local governments: They need people to violate the law in order to keep themselves out of the red,” O’Neill said. “Capping the annual proportion of local expenditures derived from monetary sanctions, or earmarking the monetary sanction revenue for community programs that address the root causes of crime, would go a long way in alleviating the social problems associated with the system. Or — even better — get rid of the system of monetary sanctions altogether.”

This study was funded by a grant from the National Institute for Child Health and Human Development to the 91Ě˝»¨Center for Studies in Demography & Ecology, and as part of the eight-state study by Arnold Ventures.

###

For more information, contact O’Neill at oneillkk@uw.edu.

A new 91Ě˝»¨ study of thousands of local rental ads finds a pattern of “racialized language” that can perpetuate neighborhood segregation, using specific terms to describe apartments in different areas of town.

Terms like “convenient” and “safe and secure” are more common in neighborhoods with a greater proportion of people of color, while “vintage” and “classic” are more popular in predominantly white neighborhoods.

“When you’re looking at racial segregation, we all make housing choices, and those choices we make affect segregation. We should know if we’re making choices based on racialized discourse,” said , a graduate student at the 91Ě˝»¨and lead author of the study. “A racialized society can be perpetuated through means that aren’t clearly conscious.”

The findings don’t mean the ads are overtly, or even intentionally, racist, Kennedy said. Rather, words and phrases – certain terms common to some neighborhoods, and certain terms for others – can reinforce perceptions of neighborhoods, influence where people choose to live, and ultimately, create areas of the city where some racial and ethnic groups are more prevalent than others.

The published Aug. 3 in the journal Social Forces.

Past research has documented segregation in Seattle, and the legacy of redlining in some neighborhoods. Through the mid-20^th century, real estate and rental ads identified properties in “restricted” areas – those with covenants designed to keep out people of color. The Fair Housing Act of 1968 prevented such discrimination in housing, and by 1970, overtly race-related language in local housing ads had essentially disappeared. But by then, as is the case , , and less-explicit forms of discrimination continued.

Given Seattle’s economic and population growth in recent years, Kennedy wanted to examine the factors that could sustain some of the de facto segregation that exists today. Seattle has grown by , and people new to the area may have little information about specific neighborhoods.

91Ě˝»¨sociologist Kyle Crowder has written about how people tend to move to neighborhoods where there are others “like” them, often because others in their social networks live there or recommend them. Combined with the legacy of racial segregation, this perpetuates neighborhoods where predominantly white people live and shop, neighborhoods where Black people tend to live and shop, and so on.

For this study, Kennedy and the research team started with more than 400,000 Craigslist ads for the Seattle-Tacoma-Bellevue area between March 2017 and September 2018. Removing duplicate ads cut the database to about 45,000 from nearly 850 census tracts; data from the Census Bureau’s American Community Survey established the racial and ethnic breakdown of each tract.

Kennedy used an approach called topic modeling to recognize groups of words appearing together, and categorized those groups into 40 topics. For example, the topic “vintage charm” typically included words like “vintage,” “classic” and brick.” The topic of “convenience and ease” included terms such as “easy,” “convenient,” “location” and “open.”

Kennedy was then able to spot patterns between the terms in the ads, and the neighborhoods the ads were tied to.

​

Topics such as “vintage charm” and those related to walkability and surrounding amenities were associated more frequently with predominantly white neighborhoods in Seattle, such as Wallingford and Queen Anne. In neighborhoods with a greater proportion of people of color, such as Seattle’s Northgate and in Kent, topics like “safe and friendly” and those pertaining to drive times and bus access were common. In particular, terms related to security – “safety,” “secure,” “controlled,” “courtesy patrol” – were associated more frequently with neighborhoods with a higher proportion of Black residents.

“These associations are sadly aligned with what we know about racial stereotypes in the United States,” Kennedy said. “We’re worried that, in addition to influencing housing patterns, that these ads could also be a site for the reproduction of racial stereotypes.”

A general theme emerged, Kennedy said: Listings in predominantly white neighborhoods highlighted history, culture and community. In neighborhoods that had a greater proportion of people of color, listings focused more often on features that separate the property from its surroundings, or simply on transportation out of the area.

The goal of rental ads, of course, is to occupy the unit, Kennedy said, so listings try to highlight what will draw a tenant; reversing perceptions and patterns requires a more systemic effort to discourage segregation. The study notes how the Chicago suburb of Oak Park to integrating its community by promoting a variety of neighborhoods and working with real estate agents, landlords and prospective tenants on changing perceptions.

The study was funded by the National Science Foundation. Co-authors were , a doctoral student in linguistics at the UW; Chris Hess of Cornell University; and Sarah Chasins of the University of California, Berkeley.

For more information, contact Kennedy at ikennedy@uw.edu.

A new 91Ě˝»¨ study of evictions around the state shows that more women are evicted than men, and in the state’s two most populous counties, eviction rates among black and Latinx adults are almost seven times higher than for white adults.

The disparities are among the findings in the , created by a team of 91Ě˝»¨researchers and led by , a former 91Ě˝»¨postdoctoral researcher now at the University of California, Berkeley. Based on eviction filings from each of Washington’s 39 counties, the report and its series of interactive maps illustrate where, and to whom, evictions hit hardest.

The surrounding narrative addresses why, and what may be done about it.

“By collecting all this data, we wanted to provide the public with an in-depth look at eviction trends at a geographical level in the region,” said , a data scientist at the 91Ě˝»¨eScience Institute and the university’s lead on the Evictions Study. “The hope is that the information sheds light on a problem happening in people’s backyards, and that folks begin having a discussion about solutions while not ignoring evictions and their intersection with race and social class.”

Eviction — the forcible removal of a tenant, usually due to unpaid rent — starts with a filing in court by a landlord. The process of notification, response and hearing before a judge is generally the same across the country, but how much time tenants are given to respond, for example, can .

Eviction data are not easily accessible. A few localized studies in other communities are tackling eviction trends, and a national effort, the at Princeton University, has attempted to assemble data from 48 states. However, several states, such as Washington, are undercounted in these efforts. The Evictions Study fills in the gaps using data science to mine court records, demographic estimations to enumerate individuals, and sociological theory to explain trends.

The 91Ě˝»¨effort involved drilling down on evictions in specific areas through multiple sources of data, starting with Washington state. The city of Baltimore is next, and researchers have begun work with advocacy organizations in Connecticut, Cincinnati and Chicago.

The team first turned to eviction filings in Washington State Superior Court to obtain eviction counts by county. To put the numbers in context and analyze other potentially contributing factors, researchers used population, rental, income and housing market data from the U.S. Census Bureau’s American Community Survey, as well as information from the U.S. Department of Housing and Urban Development and the annual regional Point in Time counts of homelessness.

Armed with the most detail from four counties — King, Pierce, Snohomish and Whatcom — the research team released for those counties and hopes to produce more as additional data becomes available. Users can filter by geography (census tract, municipality or county) and race to see a variety of data sets, including the percentage of renters, median household income and eviction counts.

For example, a look at the municipality view of King County and the eviction rate by race reveals higher rates of eviction among people of color in South King County.

“Our research shows that evictions are pervasive, where between 2013 and 2017, 1 in 55 Washington adults were named in an eviction filing — over 400,000 adults between 2004 and 2017,” said Thomas. “The most concerning finding is the severe over-representation of black adults in the Western Washington eviction filing process. In Pierce County, 1 in 6 black adults were named in a filing between 2013 and 2017, and 1 in 11 in King County during that same time. For whites, it’s 1 in 55 and 1 in 100, respectively. This severe racial disparity makes evictions a civil rights issue, requiring new laws to intervene.”

The growing homelessness crisis in Seattle has focused attention on where and how people are living on the street, what services and programs can provide permanent solutions, and who will pay to fix the situation.

But what’s often missing in the discussion, Thomas and the research team agree, are the factors that contribute to homelessness in the first place. Studies have shown that a person who is evicted has a 1 in 5 chance of ending up homeless; the 91Ě˝»¨study points to the connections among housing affordability, evictions and homelessness.

According to the study, in 2017, 46% of Washington households were “rent burdened,” meaning that almost a one-third of household income was spent on rent. In 2019, the average monthly rent statewide reached $1,000, so a tenant would need to make $40,000 to pay that bill. In King County, however, the average rent is often more than twice the statewide average: To avoid rent burden, then, a tenant would need to make $90,000 to pay a $2,250 bill.

But evictions tend to occur in neighborhoods that are the most racially diverse and where rents are lower, especially if those neighborhoods are near higher-income areas, Thomas said. There, landlords may be willing to get rid of tenants who fall behind in rent, because there’s likely to be another tenant ready to take their place.

Among the Evictions Study’s findings:

• Across Washington state, 1.8% of adults were evicted between 2013 and 2017. That compares to , according to the Eviction Lab.
• The relative risk of eviction — the probability of an eviction in a given county versus the rest of the state — is higher in Pierce, Cowlitz and Benton counties.
• In counties with fewer than 700 evictions annually — the majority of counties in Washington — women are evicted 11% more than men.
• Black adults in King and Pierce County make up a disproportionate number of the eviction filings relative to their population. Between 2013 and 2017, 9% of King County’s black adults faced an eviction, but made up 5.5% of the county’s total population; in Pierce, 18% of black adults faced an eviction, while they made up 6.6% of the population.
• In both counties, when eviction data is broken down by race and ethnicity, Asian adults had the fewest evictions.

The disparities are significant, Thomas said, because they relate to historic patterns of discrimination.

“Most of King County’s eviction filings occur in South King County, where households of color have been displaced from Seattle. In Pierce County, some of the highest risks of eviction occur in formerly redlined neighborhoods,” Thomas said. “This link between evictions, the legacy of segregation, and gentrification can’t be ignored and requires further investigation.”

As the study was in the works last year, Thomas worked with regional housing advocates and Democratic legislators to successfully promote eviction reform in Olympia. Senate Bill 5600 extended the “pay-or-vacate” deadline for tenants to pay rent, from three to 14 days.

In recent weeks, Evictions Study data was part of a regional report, the , produced by the King County Bar Association’s Housing Justice Project and the Legal Counsel for Youth and Children. 91Ě˝»¨research found that 4% of King County renters between the ages of 18 and 24 have faced an eviction; among the report’s recommendations are improved access to legal services for young people, and more resources to educate them on their rights as tenants.

Co-authors on the Evictions Study are , a lecturer in the 91Ě˝»¨Information School; , a graduate student in sociology, and , a graduate student in geography, both at the UW; and , a data scientist at Octave Bioscience who worked on the study while a postdoctoral fellow at the UW.

The study was funded by Cascade Urban Analytic Cooperative, the eScience Institute, the Bill and Melinda Gates Foundation, Microsoft Research, and Enterprise Community Partners, with technological support from the 91Ě˝»¨Center for Studies in Demography and Ecology.

For more information, contact Hernandez at joseh@uw.edu or Thomas at timthomas@berkeley.edu.