Home Is Hard to Find

Mass Incarceration, Poor Communities, and Residential Mobility

Over the past three decades, the number of individuals incarcerated in prisons and jails in the United States has risen dramatically, from 0.2 percent of all U.S. residents in 1980 to 0.7 percent of U.S. residents in 2000 (Western 2006). Whereas in 1975 the population in jails and prisons on any given day was roughly 400,000 people, by 2003 this number had increased more than fivefold to 2.1 million people (Western 2006). As a result, more than 600,000 prisoners are released each year (Visher and Travis 2003). Released prisoners are disadvantaged educationally, economically, and socially; and incarceration has been linked to increasing inequality in the United States (Pager 2003; Western 2006). Incarceration is disproportionately experienced by young, low-skilled, African American men and has important consequences for their well-being. For example, declining labor force participation by young black men during the late 1990s, when a strong economy pulled other low-skilled workers into the labor market, has been attributed to incarceration and its effects (Holzer, Offner, and Sorensen 2005). Incarceration also increases the likelihood of experiencing severe health limitations (Schnittker and John 2007), and mortality rates are particularly high in the weeks following release from prison (Binswanger et al. 2007; National Research Council 2007). The steady flow of people into and out of prisons has played a role in the increasing inequality in recent decades, primarily by reducing opportunities for employment and lowering wages (Western 2006).

As a consequence of the dramatic rise in incarceration in the United States during the past 30 years, many communities are now grappling with the problem of reintegrating former prisoners. Successful reintegration is challenged by barriers to housing and employment and access to services, and by former prisoners’ disadvantaged positions with regard to education, work experience, social capital, and mental and physical health (Visher and Travis 2003). The challenges of reintegrating former prisoners do not fall on all communities equally. Poor urban communities bear a disproportionate share of the burden, as incarceration rates are particularly high in these communities (Pew Center on the States 2009; Sampson and Loeffler 2010; Clear 2007), leading to what some analysts have termed “million dollar blocks,” individual city blocks with such high incarceration rates that the government spends more than $1 million a year incarcerating their residents (Columbia Spatial Information Design Lab 2006). Furthermore, some evidence suggests that former prisoners are more likely to return to disadvantaged neighborhoods (Cadora, Swartz, and Gordon 2003; Lynch and Sabol 2004; Solomon and Thomson 2004), where resources and services are stretched thin and law enforcement supervision is high (Fagan, West, and Holland 2003). As a result, the criminal justice system is now as important an institution for poor communities and their residents as are the education system and the labor market. Understanding and meeting the challenges faced by poor communities requires an understanding of the role of the criminal justice system and its effects on individuals and communities.

Although most research on prisoner reentry has focused on the individual-level risk factors for recidivism and unemployment, some scholars have speculated that the neighborhoods where returning prisoners live may play a critical role in successful reintegration (National Research Council 2007). Several studies have now shown that returning prisoners who live in more disadvantaged neighborhoods are more likely to recidivate. Kubrin and Stewart (2006) found that tract-level concentrated disadvantage predicted recidivism in one Oregon county, while Mears and colleagues (2008) analyzed administrative data for the entire state of Florida and found that a county-level measure of resource deprivation was associated with return to prison for a violent or drug-related offense. Hipp, Petersilia, and Turner (2010) found that neighborhood disadvantage, social disorder, and access to social services predicted recidivism among parolees in California.

Another important question is whether moving back to the neighborhood where one lived before prison helps or hinders efforts to desist from crime. One perspective is that returning home is generally a negative influence on parolees, due to the proximity of criminogenic peers and opportunities for returning to crime or substance abuse. Another view is that the social support and social control provided by existing social networks in one’s former neighborhood may be protective. In a study of prisoner reentry in New Orleans after Hurricane Katrina, Kirk (2009) showed that former prisoners who returned to their pre-prison parishes upon their release from prison were more likely to recidivate than those who moved elsewhere.

In addition to the studies of neighborhood effects on criminal trajectories of returning prisoners, there is another stream of research that focuses on the consequences for neighborhoods of losing a large number of residents to prison and absorbing many returning prisoners. Clear (2007) refers to such churning between institutional and community supervision as “coercive mobility” and argues that the residential turnover created by mass incarceration leads to the breakdown of social cohesion in disadvantaged communities. As individuals move into and out of poor neighborhoods, where residential mobility is already high, social networks in the community become sparser, trust in neighbors declines, and collective action becomes less likely. In a study of communities and crime in Tallahassee, Florida, Clear and colleagues (2003) found that neighborhoods receiving returning prisoners at higher rates also had higher crime rates than neighborhoods receiving fewer prisoners. The relationship between neighborhood population loss due to prison admission and subsequent crime was more complex: moderate rates of population loss reduced crime, but after reaching a critically high level, further increases were associated with higher crime rates. They argued that the cumulative effect of churning people back and forth between prison and the community led to a breakdown in community social networks and ultimately eroded the capacity to control crime through informal social control. Relatedly, Hipp and Yates (2009) found that communities with more former prisoners had higher crime rates than communities with fewer former prisoners and that this effect was moderated by neighborhood social capital and voluntary organizations.

Obstacles to Securing Stable Housing among Parolees

The problem of housing insecurity among returning prisoners offers a compelling example of how institutional factors can compound socioeconomic disadvantage in an already marginalized group. In part, the challenge that returning prisoners face in finding a place to live is a reflection of their personal economic circumstances, including poor work and poor credit histories. Above and beyond these personal circumstances, however, returning prisoners face an additional set of institutional obstacles that make their housing security especially precarious (Geller and Curtis 2011; Metraux, Roman, and Cho 2007; Roman and Travis 2006). First, many private landlords require applicants to disclose their criminal history and are reluctant to accept applications from those with felony convictions (Helfgott 1997; Holzer 1996). Aggravating an already difficult economic situation, many parolees also face mounting fees for supervision, restitution, and child support (Alexander 2010), and their criminal records preclude them from obtaining most types of public assistance, public housing, and housing subsidies (Geller and Curtis 2011; Petersilia 2003). Returning prisoners even face obstacles in seeking temporary shelter with friends or family members who live in public housing. Under the U.S. Department of Housing and Urban Development’s “One Strike and You’re Out” policy (U.S. Department of Housing and Urban Development 1996), public housing tenants can be evicted if any other household members or guests are involved in drugs or other criminal activity on or off the premises, even without the tenant’s knowledge, making them very reluctant to offer assistance to anyone with a criminal record. ³

The conditions of parole also impose restrictions on where returning prisoners can live. For example, sex offenders are usually restricted from moving near schools or daycare centers. Some parolees are also subject to “electronic monitoring,” in which they wear an ankle bracelet, and must live in a house with a working phone that can transmit data (via modem) to the parole officer. In some states (including Michigan), parole agents closely supervise parolee’s living arrangements—often through home visits—and restrict them from living with others who have felony records or where there are firearms or drugs on the premises.

Despite the great risks for housing instability among returning prisoners, there has been relatively little research on their living arrangements and residential moves. Some studies have used national surveys of state prisoners to examine housing status prior to their imprisonment. One such study (Hughes, Wilson, and Beck 2001) found that 12 percent of state prisoners had been homeless at the time of their arrest, while another (Ditton 1999) reported that 9 percent had been living on the street or in a shelter at some point during the year prior to their arrest. In one of the few studies to use postincarceration data on a large sample of returning prisoners, Metraux and Culhane (2004) analyzed longitudinal data on all persons released from New York state prisons and paroled to New York City counties between 1995 and 1998 (n = 48,424) and found that 11.4 percent entered a New York City homeless shelter within two years of release. There have been fewer studies of housing instability and residential mobility among returning prisoners. One study, using survey data collected on 145 Illinois prisoners returning to the City of Chicago (La Vigne and Parthasarathy 2005), found that roughly 72 percent of the sample stayed in the same residence throughout the observation period of the study (which varied by individual but lasted between one to two years). In a study offering a much less sanguine view of housing security among formerly incarcerated men, Geller and Curtis (2011) analyzed housing instability among fathers from the Fragile Families and Child Wellbeing sample (n = 2,768) and found that men who had been incarcerated faced significantly higher odds of being homeless or experiencing some form of housing insecurity within the year prior to the study compared to men with no history of incarceration.

In one of the most comprehensive accounts of parole and prisoner reentry, Petersilia (2003) observed that “we know almost nothing about the exact housing arrangements of former prisoners” (p. 121). Unfortunately, this statement remains an accurate reflection of current research, even nine years later. The current study offers one of the first detailed views of living situations and residential trajectories of a large and representative sample of parolees, followed prospectively over a two-year time period.

Data

Through a collaborative arrangement with the Michigan Department of Corrections (MDOC), we compiled a unique dataset based on detailed administrative records on a cohort of 11,064 Michigan prisoners who were placed on parole in Michigan during 2003. ⁴ More than 90 percent of Michigan’s released prisoners are put on parole, one of the higher conditional release rates among American states. Our analyses in this article are based on a randomly selected sample (n = 3,221) of 29 percent of this population on which we collected more detailed data on post-prison trajectories—including the location of all places of residence—by coding narrative case notes that parole agents updated regularly on each parolee. ⁵ All of the variables used in our analysis have been extensively cleaned, which involved checking for duplicate records, for multiple people with the same ID number, and logical inconsistencies across variables, as well as detailed reading of the case notes where appropriate to resolve discrepancies across sources.

Our analyses here focus on variation by race, gender, age, and neighborhood poverty. Our sample is 8 percent female, 53 percent black, 45 percent white, and 2 percent other (mostly Mexican American). With regard to age, 18 percent are younger than age 25 in 2003, 17 percent are age 26 to 30, 17 percent are 31 to 35, 17 percent are 36 to 40, 13 percent are 41 to 45, 10 percent are 46 to 50, and 7 percent are 51 and older. ⁶

Returning to Pre-Prison Neighborhoods

The first stage of our analysis focuses on whether parolees return to the same neighborhoods where they lived before prison or to neighborhoods similar to their pre-prison neighborhoods in terms of tract poverty rate. In Table 1, we examine this question by looking at the geographic distance between parolees’ pre-prison addresses and the addresses where they resided immediately upon their release from prison and at each six-month interval over the following two years. We classified each postrelease residential address based on its geographic distance (in miles) from the pre-prison address. ¹³

OPEN IN VIEWER

Table 1

Distance between Pre-Prison Address and Post-Prison Addresses, by Time (Percentages)

	First Address	6 Months	12 Months	18 Months	24 Months	Ever in First 24 Months
0–0.50 miles	31.3	30.9	27.9	27.3	24.7	40.8
0.51–1.00 miles	3.9	3.7	4.4	4.5	4.6	9.3
1.01–2.00 miles	8.3	9.1	9.8	10.1	10.5	19.1
2.01–5.00 miles	16.4	18.2	19.6	18.3	20.6	33.4
>5.00 miles	40.2	38.0	38.3	39.9	39.6	58.9
Number of parolees with defined distances	3,030	2,250	1,858	1,565	1,318	3,221

NOTE: See Appendix Table A3 for residences with undefined distances to pre-prison address. The Appendix is available at http://ann.sagepub.com/supplemental.

Only 41 percent of parolees ever returned to their pre-prison neighborhood, if we define pre-prison neighborhood as within half a mile of their pre-prison address. This is an important corrective to the conventional wisdom that many returning offenders are moving back to the exact same location where they were living before prison. Not only are the majority of former prisoners not returning to their pre-prison neighborhoods, many live quite far from their pre-prison neighborhoods. At any given time point over the two-year observation period, at least 56 percent of all parolees in the sample lived more than two miles away from their pre-prison home, and at least 38 percent lived more than five miles away. Moreover, the percentage of parolees living within a half mile of their pre-prison residence declined slightly over time (from 31 percent at day one to 25 percent at 24 months), while an increasing percentage lived two to five miles from their pre-prison address (16 percent initially compared to 21 percent at 24 months). ¹⁴

Although relatively few parolees returned to their pre-prison neighborhoods, they encountered similar neighborhood contexts after prison compared to those where they lived before prison. Thus, we compared the concentration of poverty—classified as low (less than 10 percent), medium (10–20 percent), and high (more than 20 percent) in each parolee’s pre-prison neighborhood to that in his or her first post-prison neighborhood. Table 2 displays these comparisons, separately by race of the parolee. The majority of parolees returned to neighborhoods classified in the same poverty category as those where they lived before prison. However, there was also considerable variation between pre-prison and post-prison neighborhoods. More than 40 percent of blacks and more than 43 percent of whites lived in a different poverty category before prison than after prison.

OPEN IN VIEWER

Table 2

Pre-Prison to First Post-Prison Neighborhood Transition (with Row Percentages), by Race

	Low Poverty (<10%)	Medium Poverty (10–20%)	High Poverty (>20%)	Row Total
a. Black parolees
post-prison →
pre-prison ↓
Low poverty (<10%)	72	64	67	203
	35.5%	31.5%	33.0%	100.0%
Medium poverty (10–20%)	58	213	122	393
	14.8%	54.2%	31.0%	100.0%
High poverty (>20%)	85	262	683	1,030
	8.3%	25.4%	66.3%	100.0%
Column total	215	539	872	1,626
	13.2%	33.1%	53.6%	100.0%
b. White parolees
post-prison →
pre-prison ↓
Low poverty (<10%)	436	192	121	749
	58.2%	25.6%	16.2%	100.0%
Medium poverty (10–20%)	112	249	82	443
	25.3%	56.2%	18.5%	100.0%
High poverty (>20%)	54	68	139	261
	20.7%	26.1%	53.3%	100.0%
Column total	602	509	342	1,453
	41.4%	35.0%	23.5%	100.0%

In addition, there were striking racial differences in the degree to which parolees experienced continuity in exposure to concentrated poverty before and after prison. Among black parolees, the correspondence between pre- and post-prison neighborhood poverty rates was highest among those who lived in high-poverty neighborhoods before prison (66 percent of blacks in high-poverty neighborhoods before prison moved back to high-poverty neighborhoods after prison, whereas only 36 percent of blacks who lived in low-poverty neighborhoods before prison moved back to low-poverty neighborhoods). We found the opposite pattern among white parolees, where the probability of living in a post-prison neighborhood with the same poverty rate as the pre-prison neighborhood was highest for those who lived in low-poverty neighborhoods before prison (58 percent of whites who lived in low-poverty neighborhoods before prison moved back to a low-poverty neighborhood, whereas 53 percent of whites who lived in high-poverty neighborhoods before prison moved back to high-poverty neighborhoods). The racial disparities are even more striking when we compare the distribution of neighborhood poverty of blacks to whites either before or after prison, which shows that 63 percent of blacks lived in high-poverty neighborhoods before prison compared to only 19 percent of whites; while 54 percent of blacks moved to high-poverty neighborhoods immediately after their release from prison, compared to only 24 percent of whites.

Post-Prison Residential Mobility and Intermediate Sanctions

The next part of our analysis focuses on residential instability among parolees after prison. We begin by examining the types of residences where parolees live, since temporary or institutional housing is an indicator of residential instability. We are also interested in the degree to which parolees reside in “intermediate sanction” residences mandated by community supervision, as these are another potential source of residential instability.

Table 3 presents categories of institutional and noninstitutional residences where parolees lived and shows how the distribution of residence types changed with time (the cells show how many subjects lived in each type of residential setting at a specific number of months after their release from prison). More than 70 percent of subjects moved into a private residence at release, but the proportion living in private residences increased only slightly thereafter. A smaller but still substantial number (12 percent) of subjects were released to correctional centers operated by MDOC, where they had the ability to leave the facility to work, look for a job, or visit family, but they also had a nightly curfew and were sometimes kept on electronic monitoring. Our residence typology distinguishes between these correctional centers, which are typically but not exclusively used as an intermediate step between prison and living in the community on parole, and other custodial centers that are commonly used as intermediate sanctions for probation detention, technical rule violation, or reentry programming. ¹⁵

OPEN IN VIEWER

Table 3

Residence Type of Parolees, by Time

Residence Type (Percentage of Noncensored)	First Address	6 Months	12 Months	18 Months	24 Months
Private residence	72.0%	75.5%	74.3%	75.0%	75.8%
	(2,335)	(2,097)	(1,771)	(1,498)	(1,261)
State/local institution
Correctional center	11.8%	2.5%	1.3%	0.6%	0.5%
	(383)	(70)	(32)	(12)	(8)
Probation detention, TRV, or reentry programming	1.8%	2.7%	2.4%	3.0%	2.7%
	(37)	(98)	(80)	(47)	(46)
Jail		4.4%	5.9%	5.6%	4.1%
		(121)	(141)	(112)	(69)
Commercial placement
Transitional housing	1.9%	1.0%	1.0%	1.0%	1.2%
	(61)	(29)	(24)	(20)	(20)
Treatment program	1.8%	2.7%	2.4%	3.0%	2.7%
	(58)	(75)	(57)	(60)	(45)
Hotel	5.2%	0.5%	0.5%	0.6%	0.5%
	(170)	(14)	(12)	(12)	(8)
Other
Hospital	0.2%	0.3%	0.3%	0.5%	0.5%
	(5)	(9)	(6)	(9)	(8)
Shelter	3.2%	0.8%	0.5%	0.7%	0.8%
	(103)	(21)	(13)	(13)	(14)
Homeless		0.1%	0.2%	0.1%	0.1%
		(2)	(5)	(1)	(2)
Unknown residence	2.1%	8.6%	10.2%	10.7%	11.0%
	(69)	(240)	(243)	(214)	(183)
Censored		13.8%	26.0%	38.0%	48.3%
		(445)	(837)	(1,223)	(1,557)
Total	100%	100%	100%	100%	100%
	(3,221)	(3,221)	(3,221)	(3,221)	(3,221)

NOTE: N = 3,221 individuals; TRV = technical rule violation center.

Intermediate sanctions are responses by the criminal justice system to proscribed behavior by the parolee that fall short of reincarceration in prison. Sometimes these behaviors are illegal, such as drug use or petty theft; and sometimes they are violations of the rules of parole, such as alcohol consumption, curfew violations, failure to report to one’s parole officer, association with other parolees, or contact with crime victims. Intermediate sanctions are used to punish parolees for minor crimes or rule violations without returning them to prison, and they are often intended to stop such behavior from escalating to more serious offenses. For example, a parolee who fails to report to his parole officer and is suspected of using drugs may be sent to a residential drug treatment program or a technical rule violation center for a week or two to “detox.” Typically, a parolee will not receive such a sanction for a rule violation until he or she has accumulated a number of infractions or displays behaviors consistent with prior crimes. Only a small fraction (2 percent) of sample members were released to probation detention, technical rule violation (TRV), or reentry programming centers; but by a year after release, 2.4 percent were living in such a center. Another common intermediate sanction is spending time in jail. This occurs when a parolee is arrested by the police for a minor crime and either serves a short jail sentence (e.g., 90 days) or has the charges dropped before prosecution. Parolees may also spend a few days in jail as an intermediate sanction at the behest of parole agents. The proportion living in jail peaks at 6 percent at one year after release.

Other common places where former prisoners lived immediately after prison were hotels (5 percent) and homeless shelters (3 percent)—both are typical destinations for parolees who have no private residence to which they can return—but the proportion of parolees in these places declined over time. The “homeless” category captures people living on the street and is extremely uncommon in our data, although it is possible that some of the subjects with unknown residences were homeless, unknown to their parole agents. Transitional housing and treatment programs house parolees under programs operated by private entities, typically nonprofits under contract with MDOC. Treatment programs can also be used as an intermediate sanction by parole officers. The category of unknown residence captures instances when the parole agent either does not know where the individual is living (e.g., because the parolee has absconded) or does not record the type of residence associated with the address. The frequency of unknown residences increases considerably over time but never exceeds 11 percent of residences. We categorize subjects as “censored” in Table 3 if they had been discharged from parole or returned to prison. Overall, Table 3 documents a relatively high degree of institutional housing among former prisoners. Because such housing is intended to be temporary, its frequency is one piece of evidence of a high degree of residential instability in this population.

Table 4 shows how exposure to each of these residential settings varies across demographic subgroups, displaying both the percentage of sample members who ever resided in each type of residence and the percentage of time they spent in each type during the observation period. Our focus here is on the prevalence of so-called intermediate sanctions. Table 4 shows that most parolees experienced some type of intermediate sanction while on parole. For example, almost half of parolees in the sample spent some time in jail while on parole; 28 percent spent some time in a treatment program; 25 percent in a correctional center; and 23 percent in a center for probation detention, TRV, or reentry programming. Overall, 65 percent of parolees experienced at least one of these intermediate sanctions, and 53 percent experienced at least one intermediate sanction other than jail. Moreover, intermediate sanctions were common among all subgroups of parolees. Women were somewhat more likely than men to be sent to treatment and less likely to be sent to detention, TRV, or reentry programming centers. Because such facilities are gender segregated, this difference may simply reflect the availability of sanction options for women in certain geographic areas of the state. Jail and probation detention, TRV, or reentry programming centers are slightly more common among younger parolees.

OPEN IN VIEWER

Table 4

Residence Types by Demographic Groups (Percentages)

	Ever Resided in Residence Type
Residence Type	All Parolees	Whites	Blacks	Men	Women	Age 18–25	Age 26–35	Age 36–45	Over 45
Intermediate sanctions
Jail	49.1	47.7	50.5	49.2	48.2	56.0	50.2	46.1	44.8
Treatment program	28.3	28.7	27.8	27.3	39.4	24.7	24.8	32.5	31.4
Correctional center	25.2	23.3	27.0	25.2	25.9	25.0	25.3	24.8	26.1
Probation detention, TRV, or reentry programming	23.4	20.2	26.5	23.9	17.9	28.3	23.2	22.9	19.6
Any intermediate sanction	64.9	63.1	66.5	64.9	64.9	69.7	65.0	62.6	63.5
Any sanction other than jail	52.5	49.7	55.0	52.2	56.2	54.7	50.9	52.9	52.8
Other residence types
Private residence	96.2	96.0	96.3	96.1	97.2	97.8	97.5	95.8	92.6
Transitional housing	6.2	7.5	5.2	5.8	11.2	3.4	4.7	7.7	9.8
Hospital	9.0	8.7	9.3	8.5	15.9	3.9	7.0	10.8	15.4
Hotel	8.5	10.4	7.0	8.4	9.6	4.4	6.7	11.1	11.6
Shelter	8.9	9.5	8.4	8.7	10.4	4.6	5.7	11.4	15.1
Homeless	1.2	1.0	1.5	1.2	2.0	0.3	1.0	1.7	1.8
Unknown residence	32.9	27.0	37.8	32.4	38.7	31.1	32.1	33.4	35.4
	Percentage of Time in Residence Type for Average Parolee
	All Parolees	Whites	Blacks	Men	Women	Age 18–25	Age 26–35	Age 36–45	Over 45
Intermediate sanctions
Jail	4.6	4.6	4.5	4.7	3.7	5.8	4.4	4.3	4.0
Treatment program	3.0	3.4	2.7	2.9	4.2	2.1	2.5	3.6	3.8
Correctional center	1.9	1.8	1.9	1.9	1.9	1.8	1.7	2.2	1.7
Probation detention, TRV, or reentry programming	2.8	2.7	2.9	2.9	1.9	4.2	2.8	2.6	1.9
Any intermediate sanction	12.3	12.5	12.1	12.3	11.7	13.9	11.4	12.7	11.4
Any sanction other than jail	7.7	7.9	7.5	7.7	8.0	8.1	7.0	8.4	7.4
Other residence types
Private residence	74.4	76.7	72.4	74.5	73.2	77.4	77.8	71.9	68.7
Transitional housing	1.3	1.4	1.2	1.2	2.4	0.3	0.9	1.5	2.9
Hospital	0.4	0.3	0.5	0.4	0.5	0.1	0.2	0.4	1.2
Hotel	1.6	1.9	1.3	1.6	1.0	0.9	1.0	2.2	2.5
Shelter	1.3	1.5	1.2	1.3	1.3	0.6	0.6	2.1	2.2
Homeless	0.1	0.1	0.1	0.1	0.1	0.0	0.1	0.2	0.2
Unknown residence	8.6	5.6	11.2	8.5	10.0	6.8	8.2	8.9	10.9
Total	100	100	100	100	100	100	100	100	100

NOTE: N = 3,221 individuals; TRV = technical rule violation center.

How much time do parolees spend in intermediate sanctions? The bottom panel of Table 4 shows the percent of time spent in each residence type. Of the intermediate sanctions, the most time was spent in jails. Parolees on average spent about 5 percent of their nights in jail, or one out of every twenty nights while on parole. Treatment programs were the second most common, followed by detention, TRV, and reentry programming, and correctional centers. These relative rankings are similar across subgroups. Overall, about 12 percent of nights were spent in a residence for an intermediate sanction, suggesting that although common, stays in sanction residences were relatively short.

Next, we turn to another measure of residential instability, the number of residential moves per year. We ask how residences in intermediate sanction facilities are related to residential mobility among parolees. Table 5 shows two measures of residential mobility, the typical number of moves per year and the cumulative distance moved during the observation period. Table 5 also shows the degree to which residential moves were associated with moves into and out of intermediate sanctions. The median parolee experienced 2.6 moves per year, or one move every four and a half months. This is a very high level of residential mobility, a level not seen in most other populations. By comparison, the conventional threshold for a high degree of residential instability is more than one move per year (Geller and Curtis 2011).

OPEN IN VIEWER

Table 5

Residential Mobility and Intermediate Sanctions among Michigan Parolees, by Demographic Groups

	All Parolees	Whites	Blacks	Men	Women	Age 18–25	Age 26–35	Age 36–45	Over 45
Moves per year (median parolee)	2.59	2.62	2.58	2.54	3.00	3.00	2.50	2.65	2.44
Sanction moves per year	0.61	0.50	0.72	0.61	0.68	1.00	0.64	0.53	0.50
Proportion moves due to sanctions	0.24	0.19	0.28	0.24	0.23	0.33	0.25	0.20	0.20
Cumulative miles moved (median parolee)	25.50	36.43	19.88	24.37	38.47	33.72	23.59	23.66	24.28
Cumulative miles moved due to sanctions	7.79	6.58	8.55	7.86	7.38	11.21	8.21	6.50	5.77
Proportion miles moved due to sanctions	0.31	0.18	0.43	0.32	0.19	0.33	0.35	0.27	0.24

NOTE: N = 3,221 individuals.

Rates of residential mobility were similar for blacks and whites and for men and women, but younger parolees tended to experience more moves than older parolees. About one-quarter of residential moves was associated with moves into and out of intermediate sanction residences. This proportion is slightly higher for blacks (28 percent) and for younger parolees. Indeed, the higher rate of residential moves among parolees age 18 to 25 seems to be almost entirely attributable to intermediate sanction moves. Excluding intermediate sanction moves equalizes the number of moves across age groups at around two moves per year. The cumulative miles measure of residential mobility tells a similar story, although there is greater variability in miles moved. Whites tended to move longer distances than blacks, reflecting their greater concentration in the suburbs and rural areas, so the proportion of the total distance moved due to sanctions was lower among whites. Women moved farther than men overall, so the proportion of the total distance moved due to sanctions was lower for women. The proportion of distance moved due to intermediate sanctions was also lower for older parolees.

What are the consequences of intermediate sanctions for the stability of a parolee’s residence over time? We answer this question by examining the rate of return to the presanction address following release from an intermediate sanction facility or center. ¹⁶ As shown in Table 6, parolees return to the previous neighborhood following intermediate sanctions only three-quarters of the time. Returns were slightly more common among blacks than whites and among men compared to women. The rate of return seemed to vary somewhat by type of sanction, with correctional center sanctions most often leading to returns to the previous neighborhood, and detention, TRV, and reentry programming least often leading to returns (when an individual experienced multiple sanctions in sequence, we use the last sanction in the sequence to determine type). These return rates correspond to the typical sanction length, with longer sanctions associated with less likelihood of return. Parolees living in high-poverty neighborhoods before a sanction were least likely to return to the same address after the sanction, particularly when those sanctions were treatment programs or jail. Together, the results presented in Tables 4 through 6 indicate that a considerable amount of residential mobility is generated by intermediate sanctions.

OPEN IN VIEWER

Table 6

Median Days in Sanction and Returns to Previous Residence Following Intermediate Sanction by Type of Sanction and Demographic Groups

	Type of Sanction
	Treatment Program	Jail	Correctional Center	Probation Detention, TRV, or Reentry Programming	All Sanctions
Percentage of sanctions	24.1%	37.6%	20.3%	17.9%	100.0%
Median days in sanction	35	20	5	98	24
Return to presanction neighborhood (percentages)
All parolees	74.7	77.1	87.7	64.5	76.5
Whites	70.6	76.7	82.5	54.7	72.6
Blacks	77.8	77.2	90.7	70.1	79.2
Men	75.2	77.9	88.0	65.2	77.1
Women	70.8	65.1	82.2	50.0	68.1
Age 18–25	76.0	80.1	86.5	66.7	77.8
Age 26–35	75.4	76.9	89.4	58.5	76.3
Age 36–45	75.0	75.8	86.0	70.8	76.8
Over 45	71.9	75.2	88.4	60.1	74.8
Presanction neighborhood (percentages)
Low poverty (<10%)	75.5	79.6	90.3	63.6	78.2
Medium poverty (10–20%)	80.3	81.6	87.2	63.0	79.3
High poverty (>20%)	71.6	72.6	86.6	65.7	74.1

NOTE: N = 4,614 sanctions; TRV = technical rule violation center.

The implications of returning to a presanction address after the sanction likely depend on the neighborhood where the address is located. Higher-poverty neighborhoods are expected to present fewer opportunities for employment and greater risks of involvement in drug abuse or crime. Table 7 examines the presanction and postsanction neighborhoods among individuals who changed addresses following a sanction. The top panel presents results for blacks and the bottom panel presents results for whites. Among both blacks and whites who moved residences after an intermediate sanction, there is evidence that postsanction neighborhoods have higher poverty rates than presanction neighborhoods. For instance, among blacks who were living in a low-poverty neighborhood before the sanction, almost 60 percent moved to high-poverty neighborhoods after the sanction, while only 10 percent of blacks originating in high-poverty neighborhoods moved to low-poverty neighborhoods after the sanction. Similar though weaker patterns are evident among whites. Slightly more than 30 percent of whites who lived in a low-poverty neighborhood prior to the sanction moved to a high-poverty neighborhood after the sanction. Of those who originated in a high-poverty neighborhood, only 26 percent moved to a low-poverty neighborhood after the sanction.

OPEN IN VIEWER

Table 7

Presanction to Postsanction Neighborhood Transitions among Sanction Movers (with Row Percentages), by Race

	Low Poverty (<10%)	Medium Poverty (10–20%)	High Poverty (>20%)	Row Total
a. Black parolees
postsanction →
presanction ↓
Low poverty (<10%)	10	19	43	72
	13.9%	26.4%	59.7%	100.0%
Medium poverty (10–20%)	21	28	67	116
	18.1%	24.1%	57.8%	100.0%
High poverty (>20%)	34	54	263	351
	9.7%	15.4%	74.9%	100.0%
Column Total	65	101	373	539
	12.1%	18.7%	69.2%	100.0%
b. White parolees
postsanction →
presanction ↓
Low poverty (<10%)	107	53	70	230
	46.5%	23.0%	30.4%	100.0%
Medium poverty (10–20%)	43	50	35	128
	33.6%	39.1%	27.3%	100.0%
High poverty (>20%)	41	35	80	156
	26.3%	22.4%	51.3%	100.0%
Column Total	191	138	185	514
	37.2%	26.8%	36.0%	100.0%

Conclusion

The United States is now in an era of “mass incarceration” (Mauer 1999; Garland 2001), with incarceration rates that are unprecedented in our own history and compared to other nations. Incarceration is disproportionately experienced by poor urban minority men, and its effects extend beyond these individuals to their communities. Rates of incarceration and community supervision are so high in some poor urban neighborhoods that the criminal justice system has become a key institution that structures social and economic life in these communities. In this article, we draw on unique data on a cohort of parolees in Michigan to offer one of the first detailed portraits of the residential trajectories of returning prisoners. We focus on the extent to which former prisoners return to their pre-prison neighborhoods, on the types of neighborhoods and residences where they live, and on the role that intermediate sanctions play in generating residential mobility. We are aware of no prior study that examines pre-prison residences and neighborhoods, residential histories after prison, the relationship between the two, or the role of intermediate sanctions in residential mobility.

Two main sets of findings emerged from this analysis that challenge, in some ways, conventional wisdom about the living situations and housing instability of returning prisoners and emphasize the ways that both are affected by the institutions of the criminal justice system. First, our results sharply conflict with the conventional wisdom that former prisoners will return to former neighborhoods after prison. In fact, less than one-third of parolees in our sample returned to an address within a half mile of their pre-prison residence after release, and by two years after release, less than one-quarter were living within a half mile of their pre-prison address. The implications of these results for former prisoners are far from clear. On one hand, new homes and new communities after prison may mean lack of access to social and economic supports that are potentially important for reentry and reintegration. On the other hand, returning to “old places and old faces,” environments that were potentially criminogenic in the past, may increase opportunities for and tendencies toward renewed involvement in substance use and crime. The implications of returning “home” after prison likely vary by the characteristics of that environment. Moreover, high rates of residential mobility mean that future studies need to look beyond the first neighborhood after release when assessing the effects of neighborhood context on the outcomes of former prisoners, and further research is required to understand the implications of residential mobility and the predictors of neighborhood and household environments among former prisoners.

Why do so few former prisoners return to their pre-prison residences or neighborhoods? Although this question lies beyond the scope of the analysis here, we offer some hypotheses based on qualitative interviews that we conducted with twenty-four former prisoners released in southeast Michigan in late 2007 and early 2008. First, the families and communities from which former prisoners come have higher than average rates of residential mobility. During the time that a family member is in prison, the household may move. This is particularly likely when the prisoner provided the main source of income or owned or rented the family’s dwelling. Second, family members may be unwilling to continue to support a former prisoner, particularly when he or she has been in and out of prison multiple times or brings substance abuse or other criminal behavior into the household. Family relationships may fray when a family member has frequent contact with the criminal justice system and frequent and prolonged separation due to incarceration, particularly for extended kin. In such cases, treatment programs, correction centers, hotels, or homeless shelters may be the only postrelease housing options, and they are less likely to be in or near pre-prison neighborhoods. Third, former prisoners preparing for release may not be well served by returning to pre-prison social environments. Sometimes the former prisoner recognizes this and intentionally stays away from potentially criminogenic environments. In other cases, parole boards or parole agents may make this determination and require or encourage a new start in a new environment.

A second major finding concerns the volume of residential mobility among parolees and the degree to which this reflects the experience of parole supervision through intermediate sanctions. For example, 65 percent of parolees in our sample were moved away from their residence at least once for an intermediate sanction. Also, former prisoners experienced very high rates of residential mobility on average—higher than any other population of which we are aware—and roughly one-quarter to one-third of this mobility was generated by the criminal justice system through intermediate sanctions such as jail, treatment programs, and residential facilities for parole rule violators. Moreover, a substantial share of parolees who experienced an intermediate sanction (between one-quarter to one-third, depending on the sanction) did not return to their prior residence after the sanction. In short, because parolees experience considerable residential mobility associated with intermediate sanctions, the effects of the criminal justice system on residential turnover are likely more extensive than previously recognized, extending beyond moves to and from prison.

We are agnostic about the implications of this residential mobility for community well-being. It could bring further social disorganization to already disadvantaged communities, but Clear’s (2007) coercive mobility theory, particularly the argument that extremely high levels of movement to prison actually increase crime, has not been subjected to many empirical tests; nor has prior research on coercive mobility considered residential mobility due to intermediate sanctions. Moreover, we note that intermediate sanctions are intended to remove former prisoners temporarily from communities before they progress to more serious crimes. They represent both an alternative to the more consequential sanction of returning a parolee to prison and an attempt by parole agents to prevent crime before it happens (or when it does happen, before it becomes a more serious crime) or before substance use occurs. Whether intermediate sanctions benefit or harm either the individual or the community are open questions. The frequency with which they are experienced, however, suggests that these are potentially important questions for future research.

Moreover, because most prisoners are not returning to their former residences, many of the parolees who enter disadvantaged neighborhoods do so as newcomers and are therefore less likely to have social resources to buffer the detrimental effects of living in disadvantaged neighborhoods. This also means that the coercive mobility that comes from moves to and from prison may be more consequential for social cohesion than previously recognized, since it is not always the same individuals who are leaving and then returning to neighborhoods with high rates of incarceration. Indeed, there is something of a paradox when it comes to residential mobility and the neighborhood environments of former prisoners. Although these individuals experience high rates of residential mobility, a small proportion of neighborhoods house the vast majority of former prisoners. This is only possible if former prisoners are moving frequently from one disadvantaged neighborhood to another. Which former prisoners are able to escape high-poverty neighborhoods is an important question for future research.

Social and economic conditions as well as criminal justice policies vary considerably from state to state. Michigan is characterized by high unemployment, declining opportunities for employment in low-skill positions, and high rates of racial and economic residential segregation. Michigan also has few Latino or Asian residents, so we can only analyze racial differences between blacks and whites. In terms of criminal justice policies and practices, we note that Michigan’s rates of incarceration and parole are close to the national averages. Michigan also accounts for a nontrivial share (4–5 percent) of the nation’s parole population. Our findings regarding the importance of intermediate sanctions may be particularly sensitive to state-specific resources and policies related to parole supervision, particularly parole revocation. The rate of return to prison among Michigan parolees was 17 percent in 2006, the same as the national average for state parolees (Sourcebook of Criminal Justice Statistics Online 2011). Yet parole revocation rates are likely subject to countervailing forces. For example, more stringent supervision may result in higher revocation rates, but more stringent supervision coupled with greater use of intermediate sanctions may result in lower parole revocation rates. Nevertheless, one strength of our study is that it draws on data from multiple cities, metro areas, and rural areas for an entire state.