October 25, 2022
Unsheltered homelessness is on the rise amid a systemic and widespread lack of affordable housing, supportive services, and livable wages. As the housing crisis worsens, homelessness has become increasingly visible and, as a result, increasingly dominant as a public concern.
Instead of addressing the issue’s root causes—a lack of housing and supportive services—many cities have leaned into punitive responses that criminalize homelessness, such as arresting people for sitting or sleeping in certain public places. But this approach is costly and ineffective. Police don’t solve homelessness, they only move it around—to other neighborhoods, jails, and emergency rooms—rather than connecting people with the housing and services they need.
What would it take to actually end homelessness for people living on the street? And how would that affect the time and resources police spend managing the problem without solving it? New data from a supportive housing program in Denver show what could happen when communities address the underlying causes of homelessness rather than continuing the status quo.
People experiencing chronic and unsheltered homelessness are more likely than people who have a house to interact with the police. That’s because being forced to live outside can lead to citations and arrests for low-level offenses (often called petty crimes), such as public drinking, that wouldn’t be crimes if the person was inside a home. As a result, living in plain sight of police and others who may report behaviors leads to the criminalization of poverty and homelessness.
To reach and stabilize the people most at risk of cycling in and out of jail, homelessness, and emergency systems, Denver created the Denver Supportive Housing Social Impact Bond Initiative (Denver SIB). The program, which started in 2016, focused on people who had eight or more arrests over three consecutive years, with at least three of the arrests marked as “transient,” meaning the person had no address or gave a shelter address. Understanding this Denver SIB target population is crucial to assessing the impacts of supportive housing programs on the people they serve, the police who interact with this group, and the broader community.
A recent Urban Institute analysis of Denver Police Department data found that among this SIB target population of 2,181 people in 2018, arrests were disproportionately for offenses associated with unsheltered homelessness. (For the analysis below, unless otherwise noted, the number of offenses does not include those among people who were in the SIB supportive housing treatment group.)
Take arrests for criminal trespassing, for instance. When a resident calls the police on someone experiencing homelessness outside of a business, police may issue a citation and ask that person to move. Because the person doesn’t have a home, they have limited options to relocate.
In 2018, criminal trespassing offenses accounted for 21 percent of cited offenses for people in the Denver SIB target population but only 3 percent of cited offenses among all people who were arrested in Denver that year (who we call the general population).
This overrepresentation applies to all offenses associated with experiencing homelessness (such as trespassing, public-order crimes, and liquor possession) among the target population.
But this overrepresentation does not apply to other types of offenses—that is, besides those associated with experiencing homelessness—such as crimes against people...
...or crimes related to theft, burglary, or robbery.
All other types of offenses make up a much smaller share of total cited offenses among the target population than they do for the general population.
Share of all cited offenses among the target population
Share of all cited offenses among the general population
Police spend significant time and resources citing and arresting people experiencing homelessness for these minor offenses. Of the 92,550 total cited offenses in Denver in 2018, 1,576 were for offenses associated with experiencing homelessness among the SIB target population.
Urban’s analysis found that in 2018, Denver police spent an estimated 2,789 hours arresting people in the SIB target population for these types of offenses. That figure doesn’t even include the time police spend on warrants, which are often the result of people missing their court dates (which is more common for people experiencing homelessness, who face a variety of challenges to showing up in court). In 2018, Denver police spent an additional 3,627 estimated hours on warrants among the SIB target population, many of which were for offenses associated with experiencing homelessness.
Addressing homelessness through policing is not only time consuming and expensive: it also doesn’t solve the problem. Arresting people for these types of offenses doesn’t prevent offenses from happening again if the arrest doesn’t also involve follow-up services or solutions to the underlying challenges.
In fact, many people are arrested over and over again for the same types of offenses. Among the 535 people in the SIB target population who were arrested for an offense associated with experiencing homelessness in 2018, 70 percent also had at least one arrest for these types of offenses in 2017.
Arrested for an offense related to experiencing homelessness in 2017 and 2018
Arrested for an offense related to experiencing homelessness only in 2018
Font source: Wee People font, by ProPublica and Alberto Cairo.
With the SIB supportive housing program, Denver aimed to see what could happen when it addressed the underlying cause of homelessness rather than maintaining the status quo. The Denver SIB provided supportive housing (a permanent housing subsidy and intensive wraparound services) to help participants stay housed, and it used a Housing First approach, which is built on the idea that secure, affordable, and permanent housing must be available before people can work on other challenges, such as mental health or substance use disorders.
To understand the effects of supportive housing through the Housing First approach, Denver partnered with Urban to evaluate the program. The study randomly assigned people in the target population to two groups: treatment and control. The treatment group (363 people in January 2018) received housing and supportive services, such as case management and mental health supports; the control group (361 people) received usual care in the community.
The evaluation found that supportive housing significantly reduced people’s interactions with police and reduced the number of times they were arrested. Arrests related to homelessness among SIB participants dropped by more than half: In the two years after being randomized into the program, people in the treatment group were arrested for 52 percent fewer offenses associated with experiencing homelessness than people in the control group.
These results show that because most of these SIB participants were in supportive housing and were connected with the services they needed (including treatment and harm-reduction programs), police weren’t arresting as many of them for the same offenses over and over again.
Note: Because study participants were chosen based on having a police contact for an offense the day before randomization, everyone in the treatment and control groups had a police contact for an offense at month zero. The contact did not have to be for an offense associated with experiencing homelessness, which is why this number does not hit the 1 mark.
Expanding the supportive housing program to serve everyone in this target population in Denver—2,181 people in 2018, not including those already referred to the SIB program—would take unprecedented new investment in housing and services at all levels of government. But the SIB evaluation shows that if everyone in this target population were offered a stable place to live and access to services to support them, there would have been 888 fewer cited offenses associated with experiencing homelessness among the target population in 2018.
These reductions among the SIB target population could reduce overall cited offenses associated with experiencing homelessness in Denver by 7 percent and reduce the time police spend arresting people for these types of offenses each year by an estimated 1,450 hours.
Even if the SIB program weren’t expanded to everyone in the target population, offering supportive housing to an even narrower group of people arrested for repeated offenses would still have significant effects. In fact, just expanding the program to serve the 305 people in the SIB target population arrested for three or more offenses associated with experiencing homelessness in 2017 could have reduced the time Denver police spent arresting people for these types of offenses the following year by an estimated 712 hours.
This analysis, along with other results from the Denver SIB evaluation, show that supportive housing is more than just a housing solution; it’s also a policing solution, a health care solution, and a public budget solution. Denver officials, including the city’s public safety director, have recognized the need to focus on and invest in alternatives like scaling supportive housing.
Police are spending significant time and resources on managing but never solving homelessness, even though these arrests are not related to serious crimes that threaten public safety. Investing in supportive housing can stop this cycle by addressing the root causes of homelessness, helping people find stability, and creating space for police to focus on violent crimes and better serve the needs of the entire community.