POLICY REVIEW: Portland Police Bureau should identify its role in responding to the City’s homeless crisis July 2019 Mary Hull Caballero, City Auditor Policy Review Staff Constantin Severe, Former Independent Police Review Director KC Jones, Analyst II Molly Christmann, Analyst I Cover Photo: iStock.com/Xiao Zhou Published July 2019 Table of Contents Summary 1 Background 2 People experiencing homelessness report increased scrutiny by other community members and police 3 Results 4 Data to analyze interaction with people experiencing homelessness is unreliable 4 Police contacts are initiated by officers and result from 911 calls 6 More than half of homeless arrests resulted from an open warrant 7 Police officers are unclear on expectations because of the Bureau’s evolving responsibilities 9 Recommendations 11 Methodology 12 July 2019 Homeless Policy Review Summary Portland’s growing population and increased development downtown has led people experiencing homelessness to feel increased scrutiny by other community members and police. This issue was highlighted in local media reports, which have estimated that over half of all arrests made by the Portland Police Bureau involve people experiencing homelessness. Following these reports, the Police Commissioner and Police Chief requested IPR conduct a deeper review of the Police Bureau’s interactions with people experiencing homelessness. IPR found that the Police Bureau currently does not collect sufficient data to effectively analyze its officers’ interactions with people experiencing homelessness. To better understand police contacts that result in arrest, IPR reviewed more than 700 arrest reports and found police contacts with people experiencing homelessness were initiated by officers or resulted from calls to 911. More than half of the arrests resulted from a warrant, but the criminal justice system needs efficient mechanisms for resolving those warrants so that people experiencing homelessness could avoid follow-up arrests. IPR also found that changes to enforcement tactics regarding homelessness has also caused confusion for police officers about their role in responding to it. We recommend that the Portland Police Bureau identify its role in addressing homelessness, implement consistent direction for officers, improve data collection, and work with criminal justice partners to minimize follow-up arrests. IPR has developed an interactive dashboard to enable readers to review the data used in this report. The dashboard can be found on IPR’s website: https://www.portlandoregon.gov/ipr/data 1 July 2019 Homeless Policy Review Background The Portland region’s most recent survey to estimate the number of people experiencing homelessness found there were 4,177 people who were homeless in Multnomah County in 2017. People were considered unsheltered if they were living in a place not meant for human habitation, such as a vehicle, tent, or abandoned building. The survey also counted people living in shelters and transitional housing as experiencing homelessness. This survey may have underestimated the homeless population because it was conducted in February, during the heart of Portland’s rainy season. The survey also did not include the estimated 9,522 people in Multnomah County who were “doubled up” in temporary or unstable living arrangements. A recent study found rising rental prices contributed to the growth in Portland’s homeless population more than other factors, such as addiction and mental illness. 2 July 2019 Homeless Policy Review People experiencing homelessness report increased scrutiny by other community members and police Many of Portland’s construction and redevelopment projects in Old Town and the Central Eastside are located where there were existing homeless communities. This redevelopment has led to increased potential for conflict between new residents and members of the homeless community. Some people experiencing homelessness have reported increased scrutiny in recent years. One person said in an interview he felt particularly targeted when people take pictures of a homeless person or their campsite so it can be reported to the City. People experiencing homelessness also said their community is vulnerable to having police interactions because their lives are conducted in public view, such as when camping or riding transit, and this causes them to believe they are viewed as a nuisance. Some said they need to be downtown to be close to social service resources, but they recognize this increased visibility puts them at risk for police contact. Unsheltered people face the unique risk of community members calling police based on their appearance or behaviors that may be beyond their control. As one social service provider said, “[i]t can be hard to have a mental crisis or even a bad day in public.” Homeless community members said it seems like police are less inclined to believe what they have to say during an interaction. One person gave the example of security guards calling the police and believing that the police will automatically take the guard’s word over that of a homeless person. This lack of trust in the police also leads people experiencing homelessness to be less inclined to contact police when they are victims of a crime. Homeless community members said they think police won’t do anything or will conduct an inadequate investigation because of their housing status. Recent media reports suggested that as many as 52 percent of all Portland Police’s arrests in 2017 involved a person experiencing homelessness. The Police Commissioner and Police Chief requested a deeper review of the questions raised by those reports. 3 July 2019 Homeless Policy Review Results Data to analyze interactions with people experiencing homelessness is unreliable The Police Bureau is unable to effectively evaluate officers’ interactions with people experiencing homelessness because it lacks relevant data. Officers are required to write police reports for certain types of interactions, such as arrests, but are not required to document exchanges that they call “mere conversation.” There are no written records or data to account for situations where officers asked someone to move off the sidewalk or referred someone to a shelter. For interactions that lead to arrest, improvements to data collection and better guidance for officers on data entry would enable the Police Bureau to analyze the circumstances of homeless community members’ arrests to examine trends and strategize solutions. Arrest data is entered into the Regional Justice Information Network (RegJIN), a multi-agency database managed by the Portland Police Bureau. The database has many fields that officers may use to enter information about an arrest, but none specifically related to housing status. The Police Bureau has no written guidance for report writing or data entry to record a person’s housing status. Officers said they received no verbal guidance either. Sometimes a suspect may not want to tell officers where they live. In the past officers wrote “refused,” but they were told not to do that and enter “transient” instead. Officers were also told not to use old addresses for those arrested. Some fields, such as the time, GPS location, and precinct of the arresting officer are more reliable. Using this information, we grouped arrests based on whether the person provided a street address, using that information or lack thereof, as a proxy for housing status. This grouping showed that about half of all arrests can be attributed to someone without a fixed street address in the Police Bureau’s data system. Arrests of those individuals without street addresses were clustered most densely in downtown Portland and the central Eastside, with other concentrations in North and East Portland. 4 July 2019 Homeless Policy Review Data quality issues made it difficult to analyze trends. Location and address elements are entered by officers in text fields, which can result in misspellings and a lack of standardization. Information about the arrested person is entered into a universal profile that may be updated or overwritten by officers from another agency. Officers do not use the system to denote various vulnerability factors. The arrested person’s street address is often used as a proxy for housing status with officers entering “transient” or “homeless” in lieu of a fixed address. It is unclear if officers must ask arrestees to confirm their profile information during an arrest. The database has a field related to mental health, but officers do not appear to use it. Without such designations, the Police Bureau is unable to use RegJIN data to do meaningful analysis of officers’ interactions with vulnerable populations. 5 July 2019 Homeless Policy Review Police contacts are initiated by officers and result from calls to 911 Because officers often do not have to write reports or otherwise enter data for encounters that do not result in arrest, there is no way to track how many police encounters with people experiencing homelessness occur overall, and what percentage of overall encounters lead to arrest. To better understand the interactions with police that resulted in arrest, we reviewed a representative sample of 727 arrest reports involving 843 homeless individuals from 2017 and 2018. Reviewing arrest reports allowed us to better understand how interactions originated and other aspects of the arrest, such as whether the arrested person was in mental health crisis or had an outstanding warrant. Arrests of homeless people started in various ways and often involved warrants  6 July 2019 Homeless Policy Review We found about 45 percent of the arrests followed a dispatched call, and 44 percent were self-initiated by officers, with smaller percentages originating from private security and traffic stops. Regardless of what prompted an encounter, most people were taken into custody (89 percent). Most of the arrests did not occur at homeless encampments. Arrest locations varied and occurred on streets (45 percent), TriMet vehicles or stations (14 percent), and at businesses (14 percent). Arrests in our sample mostly involved officers from Central (38 percent) and East Precinct (23 percent) with some arrests made by the Transit Division (18 percent) and North Precinct (17 percent). Over 85 percent of officers’ reports did not mention housing status in the arrest narrative. The reports commonly noted how the situation started, investigation details of any alleged crimes, a warrant check, and whether the arrested person was taken into custody. More than half of homeless arrests resulted from an open warrant Roughly 60 percent of the arrested persons in our sample that were homeless had an outstanding warrant for their arrest. Most of the warrants were for misdemeanor charges and about a quarter were for failing to appear in court for a prior offense. Officers have broad discretion in making arrests, but it is the Police Bureau’s practice to arrest people with an active warrant. For misdemeanor warrants that do not involve individual victims, officers said, they may forego an arrest and tell the person they have a warrant and need to turn themselves in at jail. Activities prohibited by City Code can disproportionately affect people experiencing homelessness because they are functions of living outside, such as: erecting structures or camping on public property, setting fires on public property, and drinking alcoholic beverages in public. Most officers say these violations are low priority offenses. Other types of minor crimes, however, such as disorderly conduct, criminal mischief, and criminal trespass, provide sufficient legal justification to stop someone. These stops can lead to a request for identification, a warrant check, and even arrest. 7 July 2019 Homeless Policy Review People experiencing homelessness said they perceive warrant checks as an excuse to remove them from the street. Homeless advocates pointed to warrant checks conducted by officers along the Springwater Corridor or I-205 Multiuse Trail as examples of this trend. Officers noted warrant checks in these locations were often done in response to the large number of complaints by users of the trail and neighboring residents. They described the warrant checks as a way to address people in camps engaged in criminal behavior without displacing uninvolved people. While our review of arrest reports showed that some officers spotted a “known individual” and ran a warrant check, this was not common. Some people with warrants are caught in a perpetual cycle of getting arrested, receiving a court date, failing to appear in court, and getting a new warrant. In Multnomah County, it is unlikely that a person with a lowlevel misdemeanor charge will be booked in jail, but officers still take them in to be processed. Those arrestees are then released based on their lowrisk status and given a court date. Court dates for the arrestees are often 30 days after the arrest date, which homeless advocates noted increases the likelihood of them not appearing in court. A month is a long time for a person to wait to address legal issues, the advocates said, particularly if the person is unsheltered and may have mental health or addiction issues. According to local prosecutors and other Multnomah County officials, previous attempts at addressing the high failure-to-appear rates among people experiencing homelessness proved to be unsuccessful. In 2012, a weekly community court location for low-level offenses was opened at Bud Clark Commons in Old Town. Another effort was to give a citation in lieu of arrest for certain misdemeanors. However, even with reduced lag time between arrests and court dates, both efforts had a high failure-toappear rate and were deemed ineffective, a Multnomah County Deputy District Attorney said. 8 July 2019 Homeless Policy Review Several officers and social service providers interviewed for this report provided possible solutions, such as an ability to reset court dates without turning yourself in at the jail and the ability to resolve charges at the initial appearance. Police officers are unclear on expectations because of the Bureau’s evolving responsibilities Police officers said the City needs a consistent approach to enforcing laws and other actions related to homeless community members. Officers previously were involved more heavily in camp clean-ups and have received conflicting messages on the types of tent camping the City would allow. As those responsibilities have changed, officers said they do not have enough guidance on their role in the City’s strategy to address homelessness. The Police Bureau’s current role in camp clean-ups is limited given that a separate City program was created to coordinate them using vendors under contract to the City. Officers do not clean up campsites or remove property from them. If asked, police officers may check on the A Police Bureau Neighborhood Response officer checks on someone sleeping in a tent on the Central Eastside. wellbeing of campers, Photo by Audit Services refer them to shelters or services, and provide security for the City’s private clean-up contractor. Officers generally agreed that the Police Bureau was not the appropriate agency to lead camp clean-ups and said using vendors for this purpose is an improvement. An officer described 2015 and 2016 as a time when the City’s approach was less effective because officers were more involved in clean-ups. 9 July 2019 Homeless Policy Review Officers pointed to the now defunct “Safe Sleep Guidelines” pilot in early 2016 as a particularly confusing time. The program allowed an individual or groups of six people or fewer to camp on sidewalks or other rights of way from 9 p.m. until 7 a.m. Tents were supposed to be dismantled each morning, but officers and community members said compliance with the guidelines was inconsistent. Officers said they were overwhelmed by the number of camps and the exemption made it hard to enforce laws against camping during daytime hours and other associated illegal activity. Community members blamed the Safe Sleep Guidelines for the large encampment that grew on the Springwater Trail. The City ended the program abruptly in August 2016. The Police Bureau’s role is further complicated because of its additional responsibility as a service provider. The Bureau’s Service Coordination Team offers housing and treatment to frequent drug and property crime offenders to address their addictions, mental health issues, and criminal behavior. Police Bureau staff serves as case managers for frequent arrestees and the Bureau contracts with Central City Concern for housing. Without a role in camp clean-ups, several Police Bureau members said they were unclear as to the City’s current expectations for the Bureau. Their role is limited to responding to allegations or observations of criminal conduct, whether dispatched as result of 911 calls or self-initiated while on patrol. Officers are given discretion in how they enforce low priority offenses but are not given guidance on how this discretion should be applied as part of an overall strategy in addressing homelessness. With possible changes to the emergency response system on the horizon, the Police Bureau should work with city leaders and community members to define its specific role in addressing homelessness. 10 July 2019 Homeless Policy Review Recommendations We recommend that the Portland Police Bureau: 1. Seek direction from the Mayor and City Council to identify its role in addressing the City’s homelessness crisis. 2. Develop and implement consistent direction for officers who interact with people experiencing homelessness. 3. Improve data quality and usefulness of arrest data in RegJIN or a future database by: a. Eliminating or limiting the use of “free form” fields; b. Recording and retaining more “at time of incident” information about arrestees; c. Consulting with the City Attorney and social service providers to develop consistent indicators related to mental health, housing status, substance addiction and other aspects of vulnerability. 4. Work with Multnomah County criminal justice partners to identify and implement ways to minimize “failure-to-appear” warrants, which may increase suffering for vulnerable community members and lead to inefficient uses of police resources. 11 July 2019 Homeless Policy Review Methodology This review was conducted in response to a request from the Police Commissioner and Police Chief. IPR developed and conducted the review. To learn more about homelessness we reviewed multiple national studies on homelessness as well as research specific to Portland. We conducted listening sessions with people experiencing homelessness and talked with staff at service providers including:  Central City Concern  OutsideIn  Sisters of the Road  Transition Projects  A Home for Everyone  Native American Youth and Family Center  Right 2 Survive We also interviewed representatives from: 12  Multnomah County’s Joint Office of Homeless Services  Multnomah County’s Local Public Safety Coordinating Council  Laurelhurst Neighborhood Association  Lents Neighborhood Association  ACLU Oregon  Latino Network  San Francisco’s Coalition on Homelessness  San Francisco Police’s Healthy Streets Operations Center  Seattle Police’s Navigation Team  Portland Mayor’s Office  City Auditor - Audit Services  Portland Police Bureau  Command Staff  Neighborhood Response Teams  Behavioral Health Unit  Service Coordination Team July 2019 Homeless Policy Review We reviewed police bureau arrest data from 2017 and 2018. From those arrests we selected a representative sample of 727 arrest reports where the arrestee’s address was listed in the arrest data as transient, homeless, houseless, or a shelter. This group does not include "Unknown" as we did not consider that a positive affirmation of a person’s housing status. We reviewed those reports and analyzed the content to learn more about how the interactions originated, circumstances of the arrest, and other relevant information. 13 July 12, 2019 Mary Hull Caballero City Auditor 1221 SW 4th Avenue, Room 140 Portland, OR 97204 Dear Auditor Hull Caballero: Thank you for the opportunity to review and respond to the Independent Police Review Division’s report regarding how the Portland Police Bureau responds to persons in the houseless community. News reports raised concerns about police criminalizing persons experiencing houselessness. In response to these reports, I requested this review to improve transparency around this important topic that affects so many in our city. A review of our policies and procedures was needed. While your report did not identify any actions by the Portland Police Bureau that criminalized persons experiencing houselessness, we recognize your recommendations can strengthen our existing policies and procedures and improve data collection for future analysis. Thank you again for allowing for an ongoing collaborative effort to address such important issues in our community. I look forward to implementing these recommendations as well as our continued partnership. Sincerely, DANIELLE M. OUTLAW Chief of Police DMO/tws Portland Police Bureau Responses to IPR Recommendations to the Homeless Policy Review 1. Seek direction from the Mayor and City Council to identify its role in addressing the City’s homeless crisis. Agree, in practice. The Chief of Police and the members of the executive team meet regularly with the Mayor and other members to City Council to discuss the roles and responsibilities of the Police Bureau. Within 30 days, the Chief’s Office will meet with the Mayor’s Office to clarify the Police Bureau’s role. Any clarifications from this meeting will be communicated to the entire Police Bureau within 45 days. 2. Develop and implement consistent direction for officers who interact with people experiencing homelessness. Agree. It is imperative to our mission to provide our members with consistent direction and expectations for their interactions with persons experiencing homelessness. Within 30 days, the Office of the Inspector General (OIG) will meet with subject matter experts and review Directive 835.20 – Established Campsites on Public Property and Directive 840.00 – Arrest with Warrant. The OIG will provide policy recommendations to the Chief of Police or designee within 90 days. Within 90 days, the Community Services Division (CSD) will develop reference materials for use by officers and community members related to interactions with people experiencing homelessness. 3. Improve data quality and usefulness of arrest data in RegJIN or a future database by: a. Eliminating or limiting the use of “free form” fields; Agree. Within RegJIN, officers can enter the following words into the free form address field: Confidential, Refused, Houseless, or Unknown. Within 30 days, the Records Division will work with the Training Division to develop and distribute a Tips and Techniques clarifying the use of these terms within the free form address field. We believe this change will improve the quality and usefulness of the date within RegJIN. Additionally, the Police Bureau is working on a Request for Proposals (RFP) for a new records management system. We will incorporate this recommendation into our review of vendor responses. b. Recording and retaining more “at time of incident” information about arrestees; Agree, in practice. Members of the Police Bureau are trained to ask for current information at the time of arrest. A bureau-wide reminder will be incorporated into the Tips and Techniques generated in response to 3(a). c. Consulting with the City Attorney and social service providers to develop consistent indicators related to mental health, housing status, substance addiction and other aspects of vulnerability. 1 Agree. Within 30 days, the Community Services Division (CSD) will meet with our partners to develop a comprehensive list of indicators related to the aspects of vulnerability cited. Within 120 days, CSD will provide any recommendations from the group to the Chief of Police or designee. 4. Work with Multnomah County criminal justice partners to identify and implement ways to minimize “failure to appear” warrants, which may increase suffering for vulnerable community members and lead to inefficient uses of police resources. Agree to research. We recognize the impact of warrants on the vulnerable members of our community. However, the issuance of warrants rests with the court system. Within 30 days, the Assistant Chief of the Investigations Branch will establish a workgroup that includes our criminal justice partners, to research and identify strategies to minimize “failure to appear” warrants. Within 120 days, the Assistant Chief of the Investigations Branch will provide any recommendations from the group to the Chief of Police or designee. 2