At a time when America has become more racially diverse, extreme residential segregation on the basis of race nonetheless persists. The cumulative effects of this segregation on people of color are profound. Research has consistently demonstrated that Black and Latinx people living in racially segregated communities, with the concentrated poverty that often accompanies such segregation, have limited life opportunities. Residential segregation impacts access to quality education, employment opportunities, government services, and social capital. Residents of racially segregated communities also experience increased contact with the criminal legal system, one of the critical drivers of unequal opportunity in America. Although the nature of racism in housing continues to change, government housing policies continue to sustain racial segregation, often working to re-segregate communities that had managed to achieve some level of integration.
One troubling aspect of modern housing policy is the entanglement with societal assumptions about the criminality of people of color and the criminal legal system’s steady march toward mass criminalization. Mass criminalization has been defined as “the criminalization of relatively nonserious behavior or activities and the multiple ways in which criminal justice actors, norms, and strategies shape welfare state processes and policies.” Many policymakers are acting on their assumptions that poor people and people of color are criminals and dangerous. Despite a growing consensus about the problem of mass incarceration and high-profile efforts to reverse America’s race-driven over-reliance on imprisonment, mass criminalization has seeped into virtually every aspect of society.
In a growing number of communities around the country, the increased contact with the criminal legal system combines with mass criminalization to further lock people of color out of housing in predominantly White communities, further producing and sustaining racial segregation. Crime-free housing ordinances and programs (“crime-free housing ordinances”) are some of the most salient examples of this phenomenon. Crime-free housing ordinances are local laws that either encourage or require private landlords to evict or exclude tenants who have had varying levels of contact with the criminal legal system. They are part of the expanding web of “zero tolerance” policies adopted by private landlords and public housing authorities that ban renting to individuals with a criminal history. Though formally race neutral, these laws facilitate racial segregation in a number of significant ways and their impact on racial segregation is a matter of great public concern.
I. Racial Exclusion Through Crime-Free Housing Ordinances
Exclusionary local laws and policies are among the primary mechanisms that predominantly White communities utilize to ward off racial integration. The long history of racially exclusionary localism reaches back to “sundown towns,” which excluded Black people through ordinances and policies, exclusionary covenants, threats, and harassment by local law enforcement officers. Hundreds of cities across America have been sundown towns at some point in their history. Not only were Black people barred from living in these towns, but Black people who entered the town or were found there after sunset were subject to harassment, threats, and acts of violence.
Facially-discriminatory local laws that bar Black people are, of course, now illegal. However, homogeneous municipalities continue to act on their worst racial biases by adopting exclusionary housing policies that masquerade as race-neutral principles of rational planning and home rule. Many local communities exercise their local power to relegate poor people of color to marginalized, resource-starved neighborhoods, away from the economic prosperity of their own communities. Crime-free ordinances are a step in the evolution of exclusionary localism.
A. Rise of Crime-Free Ordinances
Crime-free ordinances have roots in the law enforcement community. Historically, they are police-sponsored programs that seek to create closer collaboration between police departments and landlords. The first ordinances were created by the International Crime Free Association (“ICFA”), an organization founded in 1992 by a member of the Mesa Police Department in Mesa, Arizona. The stated goal of the ICFA is to use “law enforcement based crime prevention” to keep illegal activity, and the tenants believed to bring it, off of rental property. The efforts of the ICFA have led to the adoption of crime-free ordinances across the United States. According to one estimate, approximately 2,000 municipalities across forty-four states have adopted a crime-free housing ordinance.
Under the authority of crime-free housing ordinances, landlords are instructed or encouraged to refuse to rent to prospective tenants with a criminal history, including a history of arrests without conviction, regardless of whether that record suggests a present risk to the rental property or the safety of other tenants. In some municipalities, landlords are also encouraged to deny rental applications from individuals who were previously evicted because of suspicions that they engaged in criminal activity. The core components of crime-free housing ordinances are lease addendums that allow or require landlords to evict tenants who they believe have engaged in or facilitated criminal behavior. Under the ICFA model addendum, a tenant risks eviction if he or she has engaged in or facilitated any criminal activity. The model lease addendum does not define what constitutes criminal activity for purposes of the agreement. It provides that:
a single violation of any of the provisions of this added addendum shall be deemed a serious violation, and a material and irreparable non-compliance. It is understood that a single violation shall be good cause for immediate termination of the lease . . . .
Many municipalities around the country have adopted a version of this model lease addendum. Significantly, a resident usually does not have to be convicted in order to be evicted. A common crime-free lease addendum provision states “[u]nless otherwise provided by law, proof of violation shall not require a criminal conviction, but shall be by a preponderance of the evidence.” This creates the possibility that a mere arrest—or even a stop that results in neither arrest nor conviction—might be sufficient to evict someone from their home. When a tenant violates the crime-free lease addendum, many of these ordinances either give the landlord the authority to evict for these activities, including those committed by guests or other members of the household, or mandate the tenant’s eviction.
B. Adoption in Response to Increasing Racial Diversity in the Community
Crime-free housing ordinances are often adopted in response to burgeoning racial diversity, not burgeoning crime. Indeed, there is evidence that racial segregation may be more than an unfortunate by-product of the adoption of some crime-free ordinances. Faribault, Minnesota is an instructive example. Faribault passed a crime-free ordinance in 2014 with a goal of getting rid of “problem tenants” living in downtown Faribault. There is reason to believe that the language of “problem tenants” operated as a “dog whistle.” Nearly all of the racial and ethnic minority households in downtown Faribault live in rental housing.
The Black population of Faribault, composed almost entirely of Somali immigrants and refugees, nearly tripled between 2000 and 2010. The 2010 census showed an increase of 214% in Faribault’s Black population since 2000 and a 263% increase in the Black population living in the downtown area of Faribault during the same period. The overall number of Black households increased 542% in that decade. Although residents of Faribault began complaining about increases in crime during this period, the overall crime rates in Faribault did not, in fact, increase dramatically.
In 2013, resident complaints of increased drug activity and theft in downtown Faribault had become a hot button issue, but police reported that records did not support any claims of an increase. The police chief reported that he believed the issues were largely a result of cultural differences in the way Somali residents of Faribault “used public space” and noted that the police department would seek to decrease pedestrian traffic in the target areas.
The decision to exempt certain properties from the Faribault crime-free ordinance also evidences an intent to focus on the growing Black population while lessening the impact on the White residents of Faribault. In Faribault, the ordinance exempts single-family dwellings occupied by a relative of the owner. This provision is more likely to exempt White residents, who are more likely to own their own home than to live in rental property. This exemption allows White property owners to rent to their presumably White relatives without concern for those relatives’ history with the criminal legal system. Similar exemptions are included in crime-free ordinances around the country.
Finally, signs of discriminatory intent are also evident in some of the statements local officials made when discussing and adopting crime-free ordinances. Local officials have proudly proclaimed their intent to “move the bad guys out of town,” or to keep out “undesirables,” or the “criminal element.” These statements are often coded expressions of racial animus. Indeed, in housing discrimination cases, courts have found similar statements to be “camouflaged racial expressions.”
II. Racial Exclusion and the Discriminatory Reality of Crime-Free Ordinances
While seeking to maintain the safety of rental properties is a laudable goal, the widespread adoption of crime-free housing ordinances raises a host of public policy and civil rights concerns. Those concerns include the ways in which crime-free housing ordinances advance racial segregation, import harmful policing practices into the private housing market, expand the effect already destructive effects of the broad and over-inclusive definition of criminal activity, and promote destructive narratives around people with criminal legal system contacts.
A. Advancing Racial Segregation
By using contact with the criminal legal system as a tool for exclusion, documented racial biases in policing and the criminal legal system are imported into the private housing market. Through crime-free ordinances, the criminal legal system becomes wrapped around the entire housing process, forcing individuals with criminal legal system contacts—disproportionately Black people—to find housing elsewhere. Furthermore, the racial impact of crime-free housing ordinances will reach far beyond the individual resident. Government housing policy is never neutral in its impact on racial segregation; the policy will either exacerbate segregation or help to reverse it. Rejection of a housing application or eviction based on almost any type of criminal legal system exposure will further systemic racial exclusion because of the racial disparities in who has a criminal record. The impact is heightened because of the breadth of crime-free ordinances. The exclusions are not only based on convictions but, by design and implementation, on any contact with the criminal legal system—from claims that a person is suspicious, to stops, to arrests, to convictions. The exclusions do not only apply when an individual is seeking to move into the community but will force the evictions of individuals already living there and deter others from applying for housing in that community in the first place.
A housing system based on whether a person has involvement with the criminal legal system effectively functions as a racialized system. This is because there are racial disparities at every stage of the criminal legal process. Most of the people in prison are Black. Overall, Black men are seven times more likely to go to prison than White men. Black women are eight times more likely to go to prison than are White women. The disparities are most stark in state prisons. In Iowa, Minnesota, New Jersey, Vermont, and Wisconsin, incarceration rates are more than ten times higher for Black residents than for White residents. In eleven states, at least one in twenty adult Black males is in state prison. And, “on any given day, nearly one third of Black men in their twenties are under the supervision of the criminal justice system.” As Dorothy Roberts puts it, “African Americans experience a uniquely astronomical rate of imprisonment.”
By relying on criteria destined to exclude people of color at disproportionate rates, crime-free ordinances will perpetuate and increase segregation in the communities that adopt them. And, just as bastions of affluence in certain communities concentrate disadvantage elsewhere, concentrating Whiteness in a community will make other communities more segregated. Accordingly, the ordinances will predictably reinforce and perpetuate segregation in surrounding communities by exiling people of color, forcing them to seek housing in already segregated communities and recreating conditions in those communities that are among the drivers of systemic segregation.
It is axiomatic that housing policy in one community will have a ripple effect in surrounding communities. Evictions and exclusions based on criminal legal system contacts will force those excluded—most likely to be people of color and their families—to seek housing in those surrounding communities. Given the realities of housing patterns in the United States and deeply entrenched systems of segregation, people of color excluded by crime-free ordinances will likely be squeezed into predominantly minority communities, reinforcing racial segregation. People of color who are excluded by crime-free ordinances in one community may also avoid seeking housing in other predominantly White neighborhoods for fear of intolerance, prejudice, and violence, a fear likely reinforced by their experience seeking housing in or eviction from communities with crime-free ordinances. By concentrating Whiteness in some communities, we concentrate people of color in others.
B. Importing Harmful Policing Practices into the Private Housing Market
Black people are disproportionately surveilled and stopped by the police. Crime-free housing ordinances provide this disparate policing with a broader province of impact and influence. The ICFA has advertised that crime-free ordinances are law enforcement-driven and based on principles of policing. In utilizing principles of policing in both design and implementation, crime-free ordinances import racially discriminatory policing practices into the private housing market.
This country has a general problem of racialized policing—racialized encounters, stops, frisks, arrests, and violence. Under the reign of contemporary policing, people of color, particularly Black men, are stigmatized, brutalized, and burdened with fines and arrest records at an alarming rate. Under contemporary policing models, officers too frequently rely on racial stereotypes of people of color that make them presumptively people of interest to the police. By linking an individual’s encounters with police officers to the possibility of being denied access to private housing, crime-free ordinances add a new dimension to the sad history of race impacting access to housing. Crime-free ordinances allow the racial bias, both explicit and implicit, that has woven itself into the fabric of American policing to more easily weave itself throughout the private housing market.
Policing-based housing policies can lead to exclusions or evictions because of police contacts resulting from the weaponization of police by community members suspicious or resentful of people of color, solely because of their race. The phrase “Living While Black” has been used to encompass the innumerable ways people of color, and Black people in particular, are viewed with suspicion and required to justify their presence in spaces where they are seen as not being in the norm. When in White spaces, Black people are often required to provide justification and proof—to police or other citizens—that they belong. When those suspicious of Black people in White spaces call the police to enforce those exclusions, these criminal legal system contacts can pose numerous risks to the health and safety of those being policed and can result in “criminal records” justifying housing exclusion.
The combination of expanding policing-based housing policies and the spike in police officers responding to complaints about Black people living their lives in White spaces increase the likelihood that Black people will experience exile and that White residential spaces will be preserved. Together, they risk working in the same way as racially restrictive covenants—community-based contractual agreements that prohibit the sale, rental, or occupancy of a residence to certain groups of people.
C. Expanding the Destructive Reach of Mass Criminalization
A significant part of the danger of crime-free housing ordinances is the broad and over-inclusive definition of criminal activity common in those ordinances. This allows the system to ensnare people who have not engaged in activities that meet traditional notions of “crime” and who have not had any meaningful contact with the criminal legal system. This problem is exacerbated because it occurs against a backdrop of mass criminalization in the United States. The entanglement of policing-based housing policies and mass criminalization is pushing already marginalized people further to the edges of society. The problem gets worse the more we criminalize relatively innocuous behavior, thus swelling the numbers of people subject to the web of policing-based housing restrictions.
Scholars and advocates have drawn much needed attention to the individual and community-based harms of mass incarceration. However, mass incarceration represents just one part of a much larger social justice crisis: mass criminalization. The criminal legal system continues to expand its reach beyond criminalization by redefining crime and criminals. Today, America arrests over ten million individuals a year. Four million people in America are currently “on probation, parole, or otherwise under the control of the criminal legal system without being incarcerated.” Mass criminalization has attached criminal sanctions to spitting in public places, violating subway rules, sleeping in public places, jaywalking, riding a bike on the sidewalk, and removing trash from a bin. The result is that the criminal legal system intersects with our lives frequently and harshly.
Mass criminalization reflects a social and political process through which society determines who will be subjected to punishment by the government—creating new categories of criminals and attaching moral judgments. Accordingly, mass criminalization is achieved not only through the proliferation of laws and harsh punishments, but also through the proliferation of tough on crime rhetoric and criminalizing narratives. We have created narratives about who perpetuates crime and how crime begins and then takes hold of communities. These narratives have “crept into popular and political language,” and in the process, the values of freedom and equality take a back seat to the desire to repress this perceived crime. This regime fuels a culture of fear—of crime and those believed to perpetuate it—leading to control and exclusion.
Mass criminalization is permeated with racial bias. Indeed, the narrative linking race and crime has endured for centuries, continuing to act as a measure of Black people’s “fitness for modern life.” Through racialized narratives, negative labeling, and media coverage, Americans’ fear of crime has morphed with America’s fear of Black people. For many, crime statistics define Black humanity, with narratives asserting that poor Black people embrace and encourage a culture of crime. In conversations ranging from criminal justice reform, to education, to housing, these racialized narratives fuel a general state of anxiety and fear and brand Black people, and people of color more broadly, as dangerous and inhuman.
The mark of criminality exists for people of color even in the absence of actual criminality. Mass criminalization feeds the myth of criminality and validates unwarranted fear. In turn, crime-free housing ordinances weaponize that fear. Indeed, mass criminalization is particularly harmful in the housing context because private landlords and municipalities define criminality so broadly, even stretching it to include cases where there has not been an arrest. Given the close connections between law enforcement, the criminal legal system, and race and poverty, for many poor people of color exclusion is almost fate.
D. Promoting Destructive Messages About People with Criminal Legal System Contacts
The narratives about formerly incarcerated people that are often embedded in the structure of crime-free ordinances and promoted through training and descriptive materials can deepen the social stigma experienced by formerly incarcerated people returning home and contribute to the cycle of recidivism. In the past several years, correctional facilities have released “record numbers” of incarcerated people who then sought to successfully reenter their communities. However, several structural barriers hinder their efforts: bars to government benefits, voting disenfranchisement, exclusion from public housing, restrictions on employment, and disqualification for educational grants. This list of collateral consequences is compounded by social stigma and stereotyping that feed private discrimination and limit the ability of formerly incarcerated people to make connections with their community and build supportive networks. The messages crime-free ordinances convey about formerly incarcerated people perpetuate this social stigma and stereotyping.
Moreover, the stigma surrounding formerly incarcerated people is often intertwined with racial stereotypes. Unfortunately, “for many Americans, crime has a Black” or Brown face. Thus, formerly incarcerated Black people generally fight against “double stigma” related to their status and historical narratives of Black people as dangerous and violent. Indeed, the narrative of Black dangerousness and excessive criminality remains a relentless part of our culture, impacting debates about race from criminal justice reform to parenting, education, and housing.
The stigma born of this narrative is particularly harmful because of its impact on housing options for formerly incarcerated people. Housing is critically important for successful reentry and can present a significant challenge for individuals returning from incarceration. Many rental property owners state they would not knowingly rent their property to someone with a criminal conviction and often deny rental applications based on the fact that an individual has a criminal record. This is also a problem imbedded in the structure of these programs and ordinances. Most crime-free programs require or encourage landlords to conduct criminal background checks for all rental applicants. The screenings established by these ordinances and those advocated by the trainings have the effect of sending the message “that all tenants with criminal histories are likely to bring criminal activity” with them.
III. Examining the Real-World Effects of Crime-Free Housing Ordinances
A. Faribault, Minnesota
The Faribault, Minnesota crime-free ordinance went into effect in 2015, with goals far beyond crime reduction. The ordinance seeks to assure “that rental housing in the City of Faribault is decent, safe, and sanitary and is so operated and maintained not to become a nuisance to the neighborhood or to become an influence that fosters blight and deterioration or creates a disincentive to reinvestment in the community.” The Faribault ordinance is among the harshest in the country. First, the ordinance creates a requirement that all owners of rental housing obtain a license from the city to operate a rental dwelling and lays out a series of requirements for obtaining and retaining that license, including compliance with the crime-free housing provisions of the ordinance. The ordinance also provides a number of reasons that a license can be revoked or not renewed. Among the reasons are “[f]ailure to actively pursue the eviction of a tenant or otherwise terminate the lease with a tenant who has violated the provisions of this Article or Crime Free Drug Lease Addendum or has otherwise created a public nuisance.”
The ordinance also requires all licensees to conduct criminal background checks, going back at least three years, on all prospective tenants or occupants who are eighteen years or older. The landlord must retain the results of the criminal background check for at least one year after the check, even if the prospective tenant is rejected, and for a year after the tenancy ends. The ordinance does not, however, provide any guidance on what criteria should be used in assessing and reviewing the criminal background check. The Faribault Police Department does, however, provide such guidance, and encourages landlords to look beyond criminal convictions and consider arrests and even police contact. A Faribault police officer who runs the crime-free program on behalf of the Faribault Police Department said:
In our crime-free classes that property owners and managers are required to attend, I explain that, sometimes during the criminal history checks, you can find out more information by coming to the PD and asking for public data that would show police contacts . . . . There may not be a conviction, but there is a wealth of information.
Faribault’s ordinance also provides the text of a crime-free lease provision that is mandatory for every new or renewed lease. The text of the mandatory lease provision prohibits a “[r]esident, any members of the resident’s household, or a guest or other person under the resident’s control” from engaging in “illegal activity, including drug-related activity” on or near the premises, as well as acts “intended to facilitate illegal activity,” on or near the premises. Finally, it prohibits members of the household from engaging in the manufacture, sale, or distribution of illegal drugs anywhere. A single violation of the crime-free lease addendum “shall be deemed a serious violation and material non-compliance with the lease.” Importantly, the ordinance does not define what constitutes illegal activity—making any violation of the law grounds for eviction after a single violation.
The ordinance also prohibits disorderly conduct on all licensed premises, and makes it the responsibility of the landlord to prevent disorderly conduct by tenants and their guests, including through eviction. The ordinance specifically provides that to find that disorderly conduct occurred, “[i]t shall not be necessary that criminal charges be brought in order to support such finding nor shall the dismissal or acquittal of such a criminal charge operate as a bar to any action under this Section.”
Finally, the ordinance gives the Faribault police the power to order eviction of a tenant pursuant to the lease addendum without an arrest or conviction. If the police department determines that a premises or dwelling unit was used in violation of the crime-free provisions, or that a tenant or occupant is in violation of the provision, the licensee must terminate the tenancy of all tenants occupying the unit and may not enter into a new lease with any of them for a period of one year.
Consider the case of Thelma Jones. Ms. Jones is a Black woman who lives in Faribault, Minnesota. After living in Faribault for almost a decade and in her then-current home for five years, Ms. Jones was informed by her landlord that she and her children had two weeks to move out. At the time of her eviction notice, Ms. Jones’ landlord had been charged with misdemeanors under Faribault’s crime-free housing ordinance for failing to meet requirements of the ordinance. The landlord informed Ms. Jones that the police advised the landlord to remove Ms. Jones from the house, stating that the police had “responded to complaints at this home 82 times” and characterized Ms. Jones’ home as a location of “ongoing criminal activity.”
In fact, Ms. Jones and her children did not have any criminal convictions to support this conclusion. However, there is evidence that police had come to Ms. Jones’ home repeatedly because of harassing calls to the police by her White neighbors. On one occasion, police responded to calls because Ms. Jones was hosting a family barbecue. On another occasion, police were called when Ms. Jones hosted a child’s birthday party. Police even responded to calls when her children were outside playing on a trampoline. Ms. Jones and her family were not evicted because they were engaged in criminal activity. The Jones family was evicted because they were not welcomed by their White neighbors.
B. Orlando, Florida
Orlando, Florida provides an example of a seemingly voluntary crime-free housing program with similarly exclusionary results. In 2009, Orlando adopted a voluntary Crime Free Multi-Housing Program. An Orlando police officer described the program as designed to “squeez[e] out all of the people who just don’t want to do right, so good people can have a nice, quiet place to live.” The program has three components. First, property owners and managers who want to participate in the program can attend an eight-hour seminar presented by the police department. During this training, property managers and owners learn about crime prevention theory, lease agreements and eviction issues, and on-going security management. The second component of the program is called Crime Prevention Through Environment Design. Here, the Orlando Police Department assesses the property to confirm it meets the security requirements of the program. The final component, and the centerpiece of the program allows a property to advertise that it has “Full Certification” if the property management agrees to include a crime-free lease addendum, modeled on the ICFA model lease addendum, in all of its leases to help with the removal of “problem tenants.” The Orlando lease addendum provides that once a property owner or manager is notified that a tenant has been arrested, he or she can fill out an eviction form and give the accused resident seven days to move out. A resident does not have to be convicted in order to be evicted. In one example, an Orlando resident was caught smoking marijuana and was evicted from his apartment within the week for violating the crime-free lease addendum.
By becoming fully-certified, Orlando properties have access to the Orlando GOAL Database, which provides data about “police actions” involving their tenants and others in their community. Specifically, the database lists “all persons arrested on Crime Free Certified Properties, as well as other properties in the database.” “All adult misdemeanor and felony arrests” and juvenile felony arrests “are entered into the database.” Certified properties also have access to a “Tenant Criminal Violation/Eviction List” which identifies all renters who have been evicted pursuant to the Crime Free Multi-Housing Program. The purpose of this list, according to the City of Orlando, is to “track the criminal violators so they cannot migrate from community to community.” The top of the list states:
The person(s) named on this list have been involved in Police-documented criminal activity and/or evicted because of Police-documented criminal activity. Involvement in any criminal activity while a resident of a Crime Free Multi-Housing Community is a violation of the Crime Free Multi-Housing Standard and is subsequently grounds for eviction.
The list is automatically sent to property managers every week.
Between 2009 and 2014, approximately 1,400 renters were evicted under the program. One of those people was Leroy Ebanks. When he was twenty-one years old, police suspected that Mr. Ebanks participated in breaking into a car. Police questioned him and he denied any involvement. In connection with his questioning, the police checked Mr. Ebanks’ criminal history which showed that he had two prior arrests but no convictions. The officers turned that information over to the rental complex where Mr. Ebanks lived and the building management immediately started eviction proceedings against him.
IV. Challenging the Desire to Exclude
Although many policymakers point to crime as a primary motivation behind the proliferation of policing-based housing policies and the desire to exclude, the problem is not actual crime but the myth of criminality. Crime-free housing ordinances indulge many of the dark prejudices at the heart of American history, including a desire to exclude anyone perceived to be a threat. The prevalence of housing exclusions based on any level of contact with the criminal legal system is consistent with America’s broader embrace of exile in response to perceived threats. Crime-free housing ordinances fit squarely within America’s history of responding to threats—real or imagined—through temporary or permanent removal from the community. In so many ways, the history of Black people in America is the history of control and exclusion. Central to that history are the legal and social limitations on how and where Black and other marginalized people can live; exclusions which have outlived both chattel slavery and legally-countenanced Jim Crow. Policing-based housing policies are only the newest tool that American communities have developed to define the boundaries of who is allowed to live and thrive within their borders, and thus who can shape and be shaped by living in these communities. By combining the brutal efficiency of mass criminalization, the racism of the criminal legal system, and the policies governing private rental housing, crime-free housing ordinances risk profound damage to the physical, economic, and psychological well-being of its victims.
Exclusion is antithetical to social and economic equality. No community has the right to banish marginalized people because of perceived threats to its identity or way of life. We have a collective responsibility to create space in our communities for everyone. In order to fulfill this responsibility, communities must grapple with the underlying motivations that feed the desire to exclude. They will need to confront their fears and prejudices in order to replace exclusion with inclusion.
Communities across the United States are beginning to acknowledge the racially disparate impact and far-reaching harms caused by exclusions based on criminal legal system contacts. As a result, there has been progressive movement towards inclusion in areas including employment and political participation. But, for the most part, we have not paid sufficient attention to the ways in which contact with the criminal legal system impacts people’s access to housing. These ordinances trade on criminal legal system contacts and effectively expel people of color. This expulsion is facilitating racial segregation. That, without more, is worrisome. But segregation carries with it additional harms, including inferior education, increased crime, and under- and unemployment. Understood in this way, crime-free ordinances are an engine for racial and economic inequality. Failure to challenge the proliferation of crime-free housing ordinances will exacerbate people of color’s exposure to poverty, crime, over-policing, and incarceration. Ensuring the safety of all communities is critically important and should be a priority. Everyone has a right to feel safe in his or her home or community. However, crime-free housing ordinances will not make communities safer. They will continue to divide, further entrenching racial bias and segregation.
About the Author
Deborah N. Archer is an Associate Professor of Clinical Law; Co-Faculty Director of the Center on Race, Inequality, and the Law; and Director of the Civil Rights Clinic at New York University School of Law. She is a nationally recognized expert in civil rights and racial justice. Archer previously worked as an attorney with the American Civil Liberties Union and the NAACP Legal Defense and Educational Fund, Inc., where she litigated in the areas of voting rights, employment discrimination, and school desegregation. She was also a member of the faculty at New York Law School and an associate at the firm Simpson, Thacher & Bartlett. Archer is a graduate of Yale Law School. She currently serves as a member of the Board of Directors and General Counsel to the Board of the ACLU and as the chair of the American Association of Law School's Section on Civil Rights. She previously served on the New York City Civilian Complaint Review Board, the nation’s oldest and largest police oversight agency, and the 2018 New York City Charter Revision Commission.
About the American Constitution Society
The American Constitution Society (ACS) believes that law should be a force to improve the lives of all people. ACS works for positive change by shaping debate on vitally important legal and constitutional issues through development and promotion of high-impact ideas to opinion leaders and the media; by building networks of lawyers, law students, judges and policymakers dedicated to those ideas; and by countering the activist conservative legal movement that has sought to erode our enduring constitutional values. By bringing together powerful, relevant ideas and passionate, talented people, ACS makes a difference in the constitutional, legal and public policy debates that shape our democracy.
 See e.g., Marguerite L. Spencer & Rebecca Reno, Kirwan Inst. for the Study of Race and Ethnicity, The Benefits of Racial and Economic Integration in our Education System: Why This Matters for our Democracy (2009) (discussing the ways in which socioeconomic and racial segregation decreases life opportunities); See Richard Rothstein, The Color of Law: A Forgotten History of How Our Government Segregated America 186–187 (2017) (discussing the fact that young Black people are more likely to live in poor neighborhoods than young White people); Sheryll Cashin, The Failures of Integration: How Race and Class are Undermining the American Dream 3 (2004).
 Devon W. Carbado, Blue-On-Black Violence: A Provisional Model of Some of the Causes, 104 Geo. L.J. 1479, 1487 (2016).
 See Deborah N. Archer, The New Housing Segregation: The Jim Crow Effects of Crime-Free Housing Ordinances, 118 Mich. L. Rev. 173, 177 (2019); Emily Werth, Sargent Shriver Nat’l Ctr. on Poverty Law, The Cost of Being “Crime Free”: Legal and Practical Consequences of Crime Free Rental Housing and Nuisance Property Ordinances 3–4 (2013) (discussing the types of requirements imposed on landlords in different types of crime-free housing ordinances). Crime-free housing ordinances are often adopted alongside nuisance ordinances that require the eviction of tenants alleged to create nuisance on the property, often measured by calls for police assistance to the property. Werth, at 4–5 (“Often municipalities will incorporate both the crime-free rental housing and nuisance property elements into one ordinance or adopt both types of ordinances simultaneously.”)
 Rothstein, supra note 1, at 42 (2017); James W. Loewen, Sundown Towns: A Hidden Dimension of American Racism 4 (2005) (“Many towns drove out their black populations, then posted sundown signs. . . . Other towns passed ordinances barring African Americans after dark or prohibiting them from owning or renting property; still others establishes such policies by informal means, harassing and even killing those who violated the rule.”).
 Loewen, supra note 4, at 4–7 (revealing that sundown towns have existed everywhere in the country).
 Rothstein, supra note 1, at 42 (stating that “police and organized mobs” enforced policies “forbidding African Americans from residing or even from being within town borders after dark”); Loewen, supra note 4, at 218.
 Fair Housing Act of 1968, 42 U.S.C. § 3601.
 See David D. Troutt, Katrina’s Window: Localism, Resegregation, and Equitable Regionalism, 55 Buffalo L. Rev. 1109, 1166 n. 228 (2008); Sheryll D. Cashin, Localism, Self-Interest, and the Tyranny of the Favored Quarter: Addressing the Barriers to New Regionalism, 88 Geo. L.J. 1985, 1993 (2000) (citing one study that found the desire for racial exclusion more strongly influenced local law than did the desire for better services and lower taxes); Paul Kantor, The Dependent City Revisited: The Political Economy of Urban Development and Social Policy 164 (1995).
 Cashin, supra note 8, at 1987 (“Marginalized populations, particularly the minority poor who are regulated to poverty-ridden, central neighborhoods, are largely excluded from participating in the favored quarter’s economic prosperity.”).
 Crime Free Programs, Int’l Crime Free Ass’n (“The Crime Free Multi-Housing Program started in 1992 in Mesa AZ, when Tim Zehring of the Mesa Police Department was tasked to design a safety program that would work in rental housing.”) (last visited November 9, 2019).
 See Archer, supra note 3, at 191–193 (discussing certain crime-free ordinances that require or encourage landlords to conduct extensive background checks that sometimes go beyond criminal convictions); Kathryn V. Ramsey, One-Strike 2.0: How Local Governments are Distorting a Flawed Federal Eviction Law, 65 UCLA L. Rev. 1146, 1182–84 (2018) (explaining that police can decide when to evict tenants but that no standards “govern the discretion of the police to require eviction”).
 See Archer, supra note 3 at 197–198 (outlining Orlando’s database for crime-free certified landlords); Crime Free Multi-Housing, City of Orlando (referring to the crime-free multi-housing program as “an important community policing tool” and noting that the police department provides a database accessible to eligible landlords that shows all persons arrested on Crime Free Certified Properties); Bianca Prieto, Crime-Free Apartment Program Starting in Orlando, Orlando Sentinel (Jan. 30, 2011) (“‘This program is [aimed at] squeezing out all the people who just don't want to do right, so good people can have a nice, quiet place to live,’ said Officer Derwin Bradley, who was tasked with starting program in Orlando. ‘Some families move from property to property just wreaking havoc.’").
 Crime Free Multi-Housing, supra note 13.
 Archer, supra note 3.
 Crime Free Lease Addendum, supra note 17; see also, e.g., Landlords Sue City Over ‘Crime-Free’ Ordinance, Am. Apartment Owners Ass’n (explaining that the city’s crime-free lease addendum says that landlords do not need a criminal conviction in order to find “proof of violation”) (last visited Mar. 23, 2018).
 Archer, supra note 3, at 193–195 (discussing addendums from different municipalities which suggest that tenants may be evicted for mere arrests or stops, sometimes even when these interactions with law enforcement take place outside of the landlord’s property). In some municipalities, the police department periodically provides owners and property managers of crime-free properties with a listing of individuals who have been arrested. See e.g., Prieto, supra note 15; Crime Free Multi-Housing, supra note 13 (explaining the information available in the database managed by the police department).
 Marie Claire Tran-Leung, Beyond Fear and Myth: Using the Disparate Impact Theory Under the Fair Housing Act to Challenge Housing Barriers Against People with Criminal Records, 45 Clearinghouse Rev. 4, 5 (2011); Schaumburg, Ill., Code § 99.10.05(F)(1) (mandating a crime-lease provision in every residential lease and providing that violations of the provision provide grounds for eviction).
 Memorandum from Andy Bohlen, Police Chief, Faribault to Brian J. Anderson, City Adm’r, Faribault (Oct. 10, 2013) (on file with author); Complaint at 1, para. 46, Jones v. City of Faribault, No. 18–CV–01643–JRT (D. Minn. June 13, 2018).
 Cmty. Partners Research, Inc., Rice County Housing Study: An Analysis of the Housing Needs of the Cities in Rice County at F–21, DF–6 (2012).
 Id. at F–22.
 Jones Complaint, supra note 23, at para. 7–9 (stating that crime rates generally decreased between 2000 and 2014).
 Rebecca Rodenborg, Faribault Police, Business Owners Take a Look at Downtown Crime, Faribault Daily News (Mar. 2, 2013).
 Memorandum from Andy Bohlen, supra note 23; see also Jones Complaint, supra note 23, at para. 7–13 (describing the racial animus that fueled the Faribault housing ordinance and other similar proposals intended to silence and control the Somali population).
 § 7–38(1)(a)(1)–(2).
 See e.g., Carol Stream, Fla. § 10–12–4(B)(1) (exempting landlords of the residential rental license requirement for single family dwellings occupied “by a member of the owner’s immediate family”); St. Louis Park, Minn., § 8–328 (exempting owners of having to attend a training program if their “only rental housing is either unoccupied or a dwelling unit homestead by a relative”).
 Katie Dahlstrom, DeKalb’s Crime Free Housing Program Gets Mixed Reviews, Daily Chron. (Feb. 26, 2014).
 See e.g., Smith v. Town of Clarkton, 682 F.2d 1055, 1066 (4th Cir. 1982) (finding reference to “an influx of ‘undesirables’” and concerns about “personal safety due to the influx of ‘new’ people” to be coded racial expressions).
 Ramsey, supra note 14, at 1183–84 (“CHOs also raise troubling questions about racial justice, especially when eviction decisions by the police department can be based only on an arrest. It is well-documented that the police are more likely to arrest people of color than white people.”); see also Human Rights Watch, No Second Chance: People with Criminal Records Denied Access to Public Housing 84–85 (2004) (noting racial disparities in the enforcement of criminal offenses in the United States); Ann Cammett, Confronting Race and Collateral Consequences in Public Housing, 39 Seattle U. L. Rev. 1123, 1141–42 (2016) (noting that recreational drug use exists among “all racial and economic groups, and is not more prevalent among African Americans” but that people living in public housing, especially Black tenants, “are especially vulnerable to surveillance and state intervention in the form of police presence, selective prosecutions, and disparate outcomes in criminal courts”).
 Rothstein, supra note 1, at 190.
 Valerie Schneider, The Prison to Homelessness Pipeline: Criminal Record Checks, Race, and Disparate Impact, 93 Ind. L.J. 421, 423–24 (2018) (providing statistics which demonstrate that Black and Latinx people are incarcerated at disproportionate rates when compared to White people).
 Dorothy E. Roberts, The Social and Moral Cost of Mass Incarceration in African American Communities, 56 Stan. L. Rev. 1271, 1272 (2004).
 Todd R. Clear, Imprisoning Communities: How Mass Incarceration Makes Disadvantaged Neighborhoods Worse 63 (Oxford 2007).
 Ashley Nellis, The Sentencing Project, The Color of Justice: Racial and Ethnic Disparity in State Prisons 3 (2016).
 Roberts, supra note 36, at 1272.
 Id. at 1272–1273.
 See generally Douglas S. Massey, Categorically Unequal: The American Stratification System 6, 18 (2007) (discussing the impact of “opportunity hoarding”).
 See Vicki L. Been et al., Supply Skepticism: Housing Supply and Affordability 4 (Oct. 26, 2017) (unpublished manuscript) (arguing that changes in housing demand and supply push people either out of a jurisdiction or force them to turn to less expensive housing in the same city and that the “failure of supply to respond to increase demand at the higher end will ripple through other submarkets as demand spills into these markets and increases their prices and rents”).
 Maria Krysan et al., Does Race Matter in Neighborhood Preferences? Results from a Video Experiment, 115 Am. J. Soc. 527, 533–534 (2009) (“An important factor shaping African-Americans’ racial residential preferences is concern about possible discrimination in predominantly white neighborhoods . . . .”).
 See Carbado, supra note 2, at 1479 (listing the variables that contribute to the systemic targeting of Black communities by police); see also Jeffrey Fagan et al., Stops and Stares: Street Stops, Surveillance, and Race in the New Policing, 43 Fordham Urb. L.J. 539, 560 (2016) (collecting studies); Tatiana Pina, New Study Shows Racial Disparities in Rhode Island Traffic Stops, Providence J. (Jan. 16, 2014) (presenting evidence showing that police in Rhode Island stopped drivers of color at a disproportionate rate); Matthew Block et al., Stop, Question and Frisk in New York Neighborhoods, N.Y. Times (July 11, 2010) (presenting statistics that show the disproportionate rates at which Blacks and Latinos were stopped under the NYPD’s stop and frisk policy).
 See Crime Free Programs, supra note 11 (“The Crime Free Programs are innovative, law enforcement based crime prevention solutions designed to help keep illegal activity off rental property”).
 I. Bennett Capers, Race, Policing and Technology, 95 N.C. L. Rev. 1241, 1255–56 (2017) (providing evidence of racial profiling in New York City, Los Angeles, Philadelphia, Minnesota, Maryland Boston, North Carolina, New Jersey, and other American cities); Bennett L. Gershman, Use of Race in “Stop-and-Frisk”: Stereotypical Beliefs Linger, But How Far Can the Police Go?, 72 N.Y. St. B.J. 42, 42 (2000) (explaining that a study done by the New York State attorney general’s office found that Blacks were over six times more likely to be stopped than Whites and Hispanics more than four times as likely); John J. Donohue III & Steven D. Levitt, The Impact of Race on Policing and Arrests, 44 J.L. & Econ. 367 (2001) (finding that the number of nonwhites arrested remains unchanged by changes in the number of white and nonwhite police officers).
 Devon W. Carbado & Patrick Rock, What Exposes African Americans to Police Violence, 51 Harv. C.R.– C.L. L. Rev. 159, 163 (2016) (explaining the variables that “converge to render African Americans vulnerable to repeated police interactions”); Kevin R. Johnson, Race Profiling in Immigration Enforcement, 28 Hum. Rts. 23 (2001) (discussing how the Supreme Court “opened the door to Border Patrol reliance on race” in conducting immigration stops).
 See Elijah Anderson, The White Space, 1 Soc. Race & Ethnicity 10, 11–15 (2015) (describing “white space[s]” as “settings in which black people are typically absent, not expected, or marginalized when present”).
 See, e.g., Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness 4 (2010) (“[I] came to see that mass incarceration in the United States had, in fact, emerged as a stunningly comprehensive and well-disguised system of racialized social control that functions in a manner strikingly similar to Jim Crow.”); Nazgol Ghandnoosh, The Sentencing Project, Black Lives Matter: Eliminating Racial Inequity In The Criminal Justice System 18 (2015) (“[M]ass incarceration’s hold on vast public resources and the obstacles erected for people with criminal records further erode the economic and social buffers that prevent crime.”); Dorothy E. Roberts, The Social and Moral Cost of Mass Incarceration in African American Communities, 56 Stan. L. Rev. 1271 (2004) (assessing the effects of mass incarceration on Black communities and concluding that incarceration does not benefit them); Ta-Nehisi Coates, The Black Family in the Age of Mass Incarceration, Atlantic (Oct. 2015) (discussing the effects mass incarceration has had on Black families).
 See Devon W. Carbado, Predatory Policing, 85 UMKC L. Rev. 545, 551 (2017) (“Mass criminalization enables the police to arrest African-Americans not only through the criminalization of non-serious conduct, but also through the diffusion of criminal justice officials, norms, and strategies into the structure and organization of the welfare state.”); Jenny Roberts, Expunging America's Rap Sheet in the Information Age, 2015 Wis. L. Rev. 321, 325 (2015) (“Although mass incarceration is perhaps the most serious and pressing problem with the criminal justice system in the United States, most criminal cases are misdemeanors and often do not result in jail or prison time. The problem is thus better characterized as one of mass criminalization.”).
 Andrea J. Ritchie & Beth Ritchie, Barnard Ctr. for Research on Women, The Crisis of Criminalization: A Call for a Comprehensive Philanthropic Response 1 (2017).
 Carbado, supra note 2, at 1487–1488 (listing the types of non-serious behaviors that localities have criminalized); Erik Luna, The Overcriminalization Phenomenon, 54 Am. U. L. Rev. 703, 707 (2005) (listing “non-criminal behavior” that people have been arrested for, like loitering and violating subway rules).
 Ritchie & Ritchie, supra note 53, at 9 (“Criminalization is the social and political process by which society determines which actions or behaviors – and by who – will be punished by the state”).
 Id. (“[H]ighly racialized and gendered narratives– whether they are about ‘thugs,’ ‘crack mothers,’ ‘welfare queens,’ or ‘bad hombres’ – are used to fuel a generalized state of anxiety and fear, and to brand people labeled ‘criminal’ as threatening, dangerous, and inhuman.”); see also Bernard Harcourt, The Collapse of the Harm Principle, 90 J. Crim. L. & Criminology 109, 113 (1999) (asserting that “claims of harm have become so pervasive that the harm principle has become meaningless”).
 See Jonathan Simon, Governing Through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear 3–7 (2007) (discussing how fear of violent crime has spread through American society and the outsized impact it has had on the poor and on certain racial groups).
 Anthony C. Thompson, Navigating the Hidden Obstacles to Ex-Offender Reentry, 4 B. C. L. Rev. 255, 258 (2004).
 See Simon, supra note 58, at 3.
 Id. at 6.
 Carbado, supra note 4, at 1489 (noting how “poverty and race intersect to create a pathway to criminalization”); see generally Khalil Gibran Muhammad, The Condemnation of Blackness: Race, Crime, and the Making of Modern Urban America (2010) (chronicling the history of Black criminality in the United States).
 Muhammad, supra note 62, at 1–2.
 Id.; see also Thompson, supra note 59 (discussing negative narratives around involvement in the criminal legal system, particularly when it comes to narratives of people of color involved in crime).
 Muhammad, supra note 62, at 1.
 Thompson, supra note 59.
 For example, criminalizing and dehumanizing narratives about immigrants of color have been a hallmark of President Trump’s campaign and presidency, including his infamous statement about Mexican migrants: “They’re bringing drugs. They’re bringing crime. They’re rapists.” Katie Rogers, Trump Highlights Immigrant Crime to Defend His Border Policy. Statistics Don’t Back Him Up, N.Y. Times (June 22, 2018); see also Gregory Korte & Alan Gomez, Trump Ramps Up Rhetoric on Undocumented Immigrants: ‘These aren't people. These are animals.', USA Today (May 16, 2018) (“’We have people coming into the country or trying to come in . . . . You wouldn't believe how bad these people are.’ . . . ‘These aren't people. These are animals.’") (quoting Donald Trump)).
 See generally Devah Pager, The Mark of a Criminal Record, 108 Am. J. Soc. 937 (2003) (discussing a study that confirmed the impact of racial and criminal stigma in the employment market).
 Michael Pinard, A Reentry-Centered Vision of Criminal Justice, 20 Fed. Sent’g Rep. 103, 103 (2007).
 Thompson, supra note 59, at 258; see also Jamil A. Favors, Note, Deconstructing Reentry: Identifying Issues, Best Practices, and Solutions, 21 U. Pa. J.L. & Soc. Change 53, 61–64 (2018) (discussing some of the difficulties faced by individuals reentering their communities); Clear, supra note 37, at 58–60 (discussing some of the employment restrictions that have been placed on individuals who were formerly incarcerated); see generally Deborah N. Archer & Kele S. Williams, Making America “The Land of Second Chances”: Restoring the Social Safety Net for Ex-Offenders, 30 N.Y.U. Rev. L. & Soc. Change 527 (2006) (discussing barriers to reentry).
 Collateral consequences, the indirect consequences that flow from a criminal conviction, include a “range of social and civil restrictions” and can have an astonishing impact on recidivism rates. See Michael Pinard & Anthony C. Thompson, Offender Reentry and the Collateral Consequences of Criminal Convictions: An Introduction, 30 N.Y.U. Rev. L. & Soc. Change 585, 590 (2006).
 Jody D. Armour, Race Ipsa Loquitor: Of Reasonable Racists, Intelligent Bayesians, and Involuntary Negrophobes, 46 Stan. L. Rev. 781, 787 (1994).
 Adrienne Lyles-Chockley, Transitions to Justice: Prisoner Reentry as an Opportunity to Confront and Counteract Racism, 6 Hastings Race & Poverty L.J. 259, 269 (2009); Clear, supra note 37, at 8–9 (“The social concept of the ‘dangerous young black man,’ so deeply ingrained in our nation’s consciousness, continues to fuel punitive politics.”).
 See Brief on Behalf of National Black Law Students as Amicus Curiae Supporting Petitioner in Buck v. Davis, 137 S.Ct. 759 (2017), 2017 WL 685534, at 5-16 (explaining the history of the view that Black people are uniquely violent and dangerous); see generally Pager, supra note 68 (presenting research which suggests that prospective employers treat Black applicants as if they have a criminal record, regardless of whether they disclose one or not).
 See Pinard & Thompson, supra note 71, at 595; Archer & Williams, supra note 70, at 543 (noting that parole officers often cite housing for parolees “as their biggest challenge.”); Lyles-Chockley, supra note 73, at 282 (“The denial of crucial social benefits – including food stamps and federally subsidized housing – is a final, and sometimes devastating, blow to ex-offenders attempting to successfully integrate into their communities.”); Thompson, supra note 59, at 278–79 (discussing how housing has been a barrier to reentry); Travis et al., Urban Inst., From Prison to Home: The Dimensions and Consequences of Prisoner Reentry 35 (2001) (discussing the barriers people released from prison face in finding housing).
 Thompson, supra note 59, at 278 (“Private property owners often inquire into the individual’s background and tend to deny housing to anyone with a criminal record.”); Travis et al., supra note 101, at 35 (“Landlords typically require potential tenants to list employment and housing references and to disclose financial and criminal history information. For these reasons, offenders are often excluded from the private housing market.”).
 Tran-Leung, supra note 22, at 5; See also Werth, supra note 3, at 15 (“[L]andlords that are encouraged or required to screen prospective tenants are likely to err on the side of rejecting anyone with a record.”).
 Faribault, Minn., Code § 7–36.
 Litigation has been filed to challenge the Faribault ordinance under the Fair Housing Act, the Equal Protection Clause of the Fourteenth Amendment, and the Minnesota Constitution. See Jones v. City of Faribault, No. 18–cv–01643 (D. Minn. June 13, 2018).
 Faribault, Minn., Code § 7–38(1).
 § 7–42(3)(g).
 § 7–44(4).
 § 7–44(4)(d).
 Gunnar Olson, Mixed Emotions: Landlords, Officials Measure Success Differently for Crime-Free Multi-Housing Program, Faribault Daily News (Mar. 29, 2017).
 Faribault, Minn., Code § 7–44(3).
 § 7–44(3)(a)–(b).
 § 7–44(3)(d).
 § 7–44(3)(e).
 § 7–43(1).
 § 7–43(5).
 § 7–44(5).
 Id. at 45, para. 197.
 Id. at 45, para. 199. The Faribault Police Department charged Jones’ landlord with two criminal misdemeanors: “Rental Dwelling Registration/Fail to Register” and “C[rime ]F[ree ]M[ulti-]H[ousing]-Fail to Attend Training.” Id. Crime-free ordinances are frequently adopted as part of a larger rental housing licensing program for all landlords in the municipality and require compliance with the crime-free ordinance provisions in order for landlords to obtain or maintain their residential operator’s license. See, e.g., Las Vegas, Nev. Code § 6.09.020(A) (2014); Schaumburg, Ill. Code § 99.10.05(A) (2015); Faribault, Minn. Code § 7–42 (2017). In some municipalities, violating these provisions is a misdemeanor.
 Id. at 46, para. 203.
 Id. at 45, para. 205.
 Prieto, supra note 15.
 Crime Free Multi-Housing, supra note 15.
 9 Investigates: Crime Free Multi-Housing Program Proves Successful, WFTV (Mar. 21, 2014).
 Crime Free Multi-Housing, supra note 15.
 9 Investigates, supra note 106.
 Crime Free Multi-Housing, Orange County Sheriff’s Office (“Property managers will be sent an automatic weekly report which will provide notification of arrests for criminal incidents occurring on your property as well as the arrests of tenants.”).
 9 Investigates, supra note 106.
 Prieto, supra note 15.
 See generally Deborah N. Archer, Exile From Main Street, 55 Harv. C.R.–C.L. L. Rev. (forthcoming 2019).