An Analysis of Safe Parking Programs: Identifying Program Features and Outcomes of an Emerging Homelessness Intervention

Abstract

As vehicular homelessness increases in the United States, safe parking programs have proliferated. Yet little research exists on this emerging homelessness intervention. This three-year, mixed-methods study analyzed one of the largest safe parking programs in operation: the Jewish Family Service of San Diego Safe Parking Program (JFS SPP). Through analysis of quantitative data and interviews with 349 clients and 15 staff, this study yielded three major findings. First, JFS SPP has a 40% positive exit rate, with younger clients, women, veterans, and families more likely to exit into housing. Second, JFS SPP is preferred over shelters by persons who have used both services. This preference is especially important for older clients and clients with disabilities for whom safe parking is a safety net. Finally, increasing access to lots and services would benefit all clients, including families, seniors, and individuals with nontraditional schedules. Findings support policy recommendations for safe parking programs.

Keywords:

• Safe parking
• vehicular homelessness
• program evaluation
• housing
• San Diego

Vehicular homelessness has been steadily increasing in the United States, becoming a more visible issue as the number of persons residing in their cars, trucks, SUVs, vans, RVs, or boats continues to climb (Bernstein & Mokri, 2021; Giamarino, Blumenberg, et al., 2022). To address the issue of vehicular residency, safe parking programs have emerged in a number of cities (McElwain et al., 2021). Safe parking programs offer clients a safe and sanctioned place to park and sleep in their vehicles at night, along with access to services and resources that vary across programs. Safe parking programs allow individuals to evade policing related to homelessness and maintain possession of their primary source of transportation (Homelessness Policy Research Institute, 2018; So et al., 2016; Wakin, 2014). They potentially offer refuge and respite in the short term and a supported pathway to rehousing in the longer term. Yet very little is known about this relatively new strategy because safe parking programs are currently under-analyzed and lack official federal categorization (Ivey & Gilleland, 2018; McElwain et al., 2021). To date, no large-scale, independent evaluation has been undertaken to understand the profiles, experiences, and outcomes of clients using safe parking programs. This three-year study addresses this gap.

Between July of 2019 and September of 2022, our research team investigated one of the largest and most comprehensive safe parking programs in the United States: the Jewish Family Service of San Diego Safe Parking Program (JFS SPP), a safe parking program located in San Diego County, California. Given its reach and resources—as well as the fact that JFS SPP operates in an area with a high rate of vehicular homelessness—this program was an ideal site to investigate the characteristics, experiences, and trajectories of safe parking clients. Three research questions guided this study:

1. What are the effects of the safe parking intervention in terms of housing placements and/or improvements to health and well-being for clients, and do any sub-populations benefit more than others?

2. Where do safe parking programs fit within the broader homelessness service ecosystem?

3. How can safe parking programs improve their services to better assist all clients?

To answer our research questions, our team analyzed client-level data collected by JFS SPP, interviewed 349 current and former clients of the program, and interviewed 15 members of the program’s staff. Our three-year mixed methods study yielded rich data regarding JFS SPP client demographics, pathways into vehicular homelessness, client destinations upon leaving the program, and client experiences within JFS SPP. Our findings make the following three contributions to the limited scholarship on safe parking programs. First, during our study’s time period, JFS SPP placed 40% of its clients into housing. JFS SPP clients vary demographically in important ways that impact their access to resources. JFS SPP specifically helps clients who are able to work or utilize social or communal resources by offering a safe space to build up their savings and more quickly rehouse. Second, JFS SPP is highly preferred over emergency shelters by clients who have used both types of services, due to a greater sense of safety and community, and offers clients a greater ease of meeting one’s basic needs than would be available through street outreach efforts alone. This is especially important for clients who are not able to work, often due to their older age or chronic health challenges. Finally, JFS SPP can better serve all clients including seniors, persons with disabilities, families with young children, and individuals with nontraditional work schedules if it provides expanded access to its lots and resources.

Based on our findings, we argue that safe parking programs play an important and distinct role in the homelessness services ecosystem and efforts should be made to expand their operations, including their ability to operate 24/7 lots, hire more case managers, and provide ongoing training to staff. We further recommend that safe parking programs be officially categorized by the US Department of Housing and Urban Development (HUD 2017) as a homelessness intervention in order to establish proper channels for federal funding and enable standardized evaluation. In addition to our policy recommendations, we recommend that, until HUD implements universal data collection mandates on safe parking programs, further research on vehicular homelessness and safe parking programs be conducted as a stopgap measure by collaborative teams of university researchers and community partners.

Rising Rates of Vehicular Homelessness

Homelessness rates have steadily increased over the past few years due to soaring housing costs and the financial reverberations of the COVID-19 pandemic (Aurand et al., 2021; Hoeven, 2022). In this context of growing insecurity, vehicular homelessness, as a subcategory of homelessness, is increasing. Precise numbers are difficult to measure because HUD does not require the collection of data on vehicular residency, although some Continuums of Care (CoCs) collect this data on their own, as will be discussed below.¹ One way to approximate growth in vehicle residency is to track unsheltered homelessness, which has grown steadily since 2015 and includes individuals residing in their cars, trucks, SUVs, vans, RVs, and other recreational vehicles (de Sousa et al., 2022). In 2019, the US Census Bureau estimated that approximately 140,000 Americans resided in their vans, RVs, or boats, a 40% increase in such dwelling patterns since 2016, although the census estimate overlooked people living out of their cars, trucks, or SUVs and predated the impact of COVID-19 on vehicular homelessness (Ryan, 2022). While counts of vehicular homelessness that include residency in cars, trucks, or SUVs have more recently been captured at the CoC level, these efforts likely undercount the extent of the problem because people residing in their vehicles have both greater mobility and a strong desire to remain invisible (Giamarino, Blumenberg, et al., 2022). Furthermore, many CoCs simply do not collect this data (McElwain et al., 2021).

Despite the issues surrounding accurate measurement, it is known that vehicular homelessness recently increased in many urban areas, with CoCs who collect such data reporting high rates of vehicular habitation (Giamarino, Blumenberg, et al., 2022; Pruss, 2012; So et al., 2016). In the Seattle/King County CoC, 50% of the unhoused, unsheltered population reside in their vehicles (All Home, 2020).² In the Los Angeles CoC, 40% of unhoused, unsheltered residents are living out of their cars, RVs, vans, or campers (Los Angeles Housing Services Authority, 2019). As rates of vehicular homelessness have grown, so has interest in understanding this particular intervention and the populations it serves. Two recent studies identified characteristics of persons experiencing vehicular homelessness, finding that women, individuals with children, employed adults, white individuals, and older adults are more likely than other unhoused, unsheltered individuals to reside in their vehicles (All Home, 2020; Giamarino, Blumenberg, et al., 2022).

The increase in the number of people living out of their vehicles prompted many cities to regulate parking, resulting in a significant rise in municipal ordinances that either restrict vehicle habitation or entirely ban the practice (So et al., 2016). The National Law Center on Homelessness and Poverty (2019) found that across their sample of 187 cities, municipal ordinances restricting vehicle residency increased by 213% from 2006 to 2019. Municipal ordinances that target vehicle habitation have been on the rise in California specifically. For instance, while 103 municipal ordinances targeting vehicular habitation exist within the Los Angeles CoC, 41 of these were adopted after 2016, revealing a 66% increase in such ordinances over the past few years (Giamarino, Brozen, et al., 2022). Alongside the growing criminalization of vehicular homelessness, and in recognition of the impossible situation it creates for people living out of their cars, safe parking programs have begun to emerge in a number of cities. Recently published analyses of safe parking programs suggest that they assist clients in becoming rehoused (Homelessness Policy Research Institute, 2018; McElwain et al., 2021), although these findings are based on self-reports from providers. Additionally, some studies found that many individuals prefer safe parking programs to the traditional emergency shelter system (So et al., 2016; Wakin, 2014). Notwithstanding this early and important research, no full-scale evaluation of the safe parking intervention has yet been undertaken. This study fills that gap by analyzing client outcomes for one of the largest and most well-resourced safe parking programs currently in operation, examining for whom and in what ways it is helpful, and clarifying the strengths and limitations of the safe parking intervention model.

An Overview of Safe Parking Programs

Safe parking programs are a relatively new intervention within the homelessness servicescape. The first such program, established in Santa Barbara in 2004, emerged to meet the needs of the growing numbers of residents residing in their vehicles (McElwain et al., 2021), a trend that has likewise been increasing across the nation but particularly within West Coast locations that are facing the brunt of the nation’s housing crisis and that exhibit the highest rates of unsheltered homelessness (de Sousa et al., 2022; Schmid, 2021). Increasing rates of homelessness in general and vehicular homelessness in particular have been noted in California specifically. While approximately 12% of the nation’s population resides in California (US Census Bureau, 2020), 29% of those experiencing homelessness in the United States are in the state, with the majority (67%) living unsheltered on the streets or in their cars (de Sousa et al., 2022). As housing prices continue to skyrocket nationally, heavily impacted states are seeing people moving into their vehicles as an option of last resort (Giamarino, Blumenberg, et al., 2022). In response, safe parking programs have opened in areas most affected by growing housing precarity and where the weather typically permits year-round vehicle residency. In 2021, 43 communities across the nation offered a safe parking program, 93% of which were located on the West Coast (McElwain et al., 2021).

The large-scale increase of safe parking programs reflects both growing need in the form of rising rates of vehicular homelessness and an increase in municipal ordinances that make it illegal to sleep in one’s car overnight (Giamarino, Brozen, et al., 2022). For clients of safe parking programs, having a sanctioned place to park and sleep enables them to hold onto their vehicle amidst this context of criminalization. By contrast, if people living out of their vehicles park and sleep on public streets, the likelihood that they will be ticketed or towed for parking violations increases (Pruss, 2012). Ticketing and towing are serious threats to a valuable resource, one that both serves as shelter and provides transportation to work as well as needed services and potential housing across a broader geographic area. The accumulation of multiple tickets or, worse, having a vehicle towed and impounded, puts individuals at risk of losing their vehicle, as many experiencing homelessness are unable to pay the fines and fees associated with vehicle recovery (So et al., 2016). According to recent research conducted by UC San Francisco Benioff Homelessness and Housing Initiative researchers, fear of losing a vehicle prevents individuals residing in oversized vehicles from seeking out social and health related services as they are hesitant to leave their most important resource unattended (Pruss et al., 2022). Since some safe parking programs allow oversized vehicles, this crisis intervention strategy can help to assuage those fears (McElwain et al., 2021).

In addition to assisting clients in protecting their most valuable resource, safe parking programs offer comparative safety; safe parking programs are a safer alternative to parking and sleeping on a public street for individuals residing out of their vehicles (Giamarino, Brozen, et al., 2022). This is especially true for vulnerable groups including women and families with children, who tend to be prioritized by safe parking programs (McElwain et al., 2021).

Safe parking programs provide a crisis-level service to individuals experiencing homelessness, with fewer restrictions than the traditional shelter system. Most shelters only serve single males or females, forcing opposite-sex couples and families to split up (Skinner & Rankin, 2016). Many people who identify as gender nonbinary report facing significant bias and harassment within this system, a reality that causes many within this community to avoid shelters altogether (Davis, 2023; National Center for Transgender Equality, 2016). Pets, who represent primary companions for many unhoused individuals, are not allowed at many shelter sites (Donley & Wright, 2012), and many individuals chafe against the loss of autonomy inherent within the shelter system (Hoffman & Coffey, 2008; Stuart, 2016). Furthermore, members of vulnerable groups, such as women, families with children, senior citizens, and persons with disabilities tend to feel unsafe in shelters (Batko et al., 2023; Grenier et al., 2016) or find that their needs are not met (Serving Seniors, 2021). For potentially all of these reasons, safe parking programs are often preferred over shelters by individuals still in possession of their vehicles (Wakin, 2014). Despite serving as an alternative crisis intervention option for people experiencing homelessness, safe parking programs find it challenging to secure adequate funding because they have not been identified by HUD as eligible for funding through the CoC or Emergency Shelter Grants programs (Weare et al., n.d.).³

Although the research on safe parking programs is limited, a recent survey of the 43 programs operating in 2021 conducted by McElwain et al. (2021) revealed that safe parking programs vary considerably in size, structure, and services. Some programs operate multiple lots, are better resourced, and offer anywhere from 21 to 101 parking spots, as well as case management services. Others operate a single lot, oversee anywhere from 6 to 60 spaces, and are not adequately funded to provide case management. In addition to differences in the availability of case management, amenities and services vary significantly across programs. Approximately 70% of safe parking programs provide access to toilets and 60% provide showers, meals, Wi-Fi, and/or electronic charging stations. Close to 60% provide financial support for vehicle repairs, auto insurance, and registration, and a similar percentage provide housing placement assistance. In addition to variation in amenities and services offered, program hours similarly vary, with more than half remaining open 24 hours a day, and the rest requiring clients to leave early in the morning and reenter by a specific time in the evening. Safe parking programs engage diverse strategies for creating a safe environment, including organizing client safety patrols, employing staff, or contracting with local police or private security groups (Homelessness Policy Research Institute, 2018).

McElwain et al.’s (2021) survey revealed that an average rate of 34% of all safe parking clients exit from their programs into permanent or temporary housing. Importantly, exit rates differed significantly by site and were based entirely on self-reports from program leadership, with no clarity of time frame and no definition of permanent or temporary housing. Due to the lack of comprehensive, contextualized research on this intervention, it is currently unclear whether program features, methods of measurement, or other factors including differences in the cost of living across CoCs impact variance in exit rates. Notwithstanding these limitations, the above-cited literature does suggest that safe parking programs assist clients in becoming rehoused by offering them a safe and free place to park while they increase their income and utilize housing navigation services. At the same time, many safe parking programs fall short of the established need within their CoC, offering space to only a fraction of individuals and families residing in their vehicles. Most programs exclude persons living in oversized vehicles such as RVs as well as individuals who are not in possession of a driver’s license, insurance, registration, or a working vehicle (Giamarino, Blumenberg, et al., 2022). Furthermore, almost half of all safe parking programs are not open 24 hours, forcing clients to find other places to park during the day (McElwain et al., 2021).

While the report by McElwain et al. (2021) identified whether or not specific target populations were served (oftentimes families with children were prioritized), significant questions remain regarding safe parking client demographics, trajectories, and experiences. Furthermore, safe parking programs currently lack standardized benchmarks, a reflection of the fact that they are inadequately categorized by HUD. This makes comparison across different programs extremely difficult. Within the McElwain report, many safe parking programs used moves into both temporary and permanent housing as a means of assessing program success. This mirrors the strategy employed by HUD to measure successful exits from street outreach programs.⁴ Indeed, when we began our study, safe parking programs were initially categorized as a form of street outreach by San Diego County’s CoC. Currently they are categorized by this CoC as comparable to an emergency shelter. Because of these inconsistencies, comparing exit outcomes across safe parking programs poses significant challenges. As such, we describe how JFS SPP differs from, and offers an alternative to, other crisis-level interventions within the San Diego County CoC.

Introduction to the Case Study

This study was conducted in San Diego County, which is located in a larger binational border region. The CoC for San Diego County encompasses 18 city jurisdictions as well as unincorporated areas. San Diego County has a population of approximately 3.3 million people in a geographic area of 4,261 square miles. In 2021 the median household income for the county was $88,240. In terms of racial and ethnic composition, 74.4% of individuals who reside in the county identify as white, 5.6% identify as Black or African American, 1.4% identify as American Indian or Alaska Native, 13.1% identify as Asian, 0.6% identify as Native Hawaiian or Pacific Islander, 5% identify as two or more races, and 35% identify as Hispanic or Latino (US Census Bureau, 2020). According to RentCafe, the current (2024) average rent within the City of San Diego, the largest jurisdiction within the county in terms of geographic size and population, is $2,997. A recent report highlights that the average household within this region spends well above 30% of their income on housing, the standard set by HUD to determine housing cost burden (Corporation for Supportive Housing, 2019). Within the county, each jurisdiction has its own policies and infrastructure to support populations experiencing homelessness, and they vary considerably. Furthermore, relatively little policy coordination occurs at a regional scale.

For a number of reasons, San Diego is an important site in which to conduct research on safe parking programs. The HUD Point-in-Time Count conducted in January of 2023 identified 10,264 people experiencing homelessness in San Diego County (San Diego Regional Task Force on Homelessness, 2023).⁵ This count marked at least a 14% increase in homelessness in the region from the year before and established San Diego as an area that contains one of the largest populations of unhoused persons in the country (de Sousa et al., 2022), the majority of whom live unsheltered (San Diego Regional Task Force on Homelessness, 2022a). Furthermore, as homelessness in general and unsheltered homelessness in particular have increased precipitously across California, vehicular homelessness has likewise been increasing within the state, owing to the combination of an ongoing housing affordability crisis coupled with milder weather that makes it possible for people to live out of their vehicles all year long (Giamarino, Brozen, et al., 2022).⁶ Although HUD does not require data collection on vehicular homelessness and thus no formal count exists for the San Diego CoC, public perceptions of increased vehicular homelessness have resulted in the recent enactment of municipal ordinances in many jurisdictions throughout the county that ban overnight sleeping in vehicles (City News Service, 2019; Johnson, 2019; Koenig, 2020).

In the City of San Diego, municipal code 86.0137 was revised in 2019 after briefly being repealed, both banning vehicular habitation anywhere in the city between the hours of 21:00 and 06:00, except at select designated lots, and prohibiting parking within 500 feet of any residence or school at any time (Prohibition of Use of Streets for Storage, Service, or Sale of Vehicles or For Habitation, 2023). Catalyzed by the recent criminalization of vehicular homelessness in the region, some jurisdictions in the county have made a clear commitment to making safe parking programs available. In 2019, the year the City of San Diego enacted the new ban on vehicular habitation, the city additionally committed funding to the largest safe parking program in the county, JFS SPP.

The organization that operates JFS SPP, Jewish Family Service of San Diego, is one of the region’s oldest and largest nonprofit human services agencies. Established in 1918, it serves close to 40,000 clients from diverse backgrounds each year. It offers an array of services ranging from programs for older adults to legal aid and shelter for asylum seekers. In February of 2019, this organization took over the operation of three safe parking lots located in the City of San Diego, and in 2020 it expanded into the northern part of San Diego County with the creation of an additional lot in the small and affluent City of Encinitas after a contentious city council vote.

Individuals interested in using JFS SPP are encouraged to call the main number and often learn about this program through 2-1-1,⁷ a web search, referrals from other organizations, law enforcement, or word of mouth from someone already using the lots. Once individuals receive a call back, they are screened for eligibility and offered an appropriate and available location in the order that calls are received. Walk-ins are not encouraged as JFS SPP is very impacted. Clients utilizing JFS SPP must possess a working vehicle but do not need to have a current driver’s license, vehicle registration, or insurance. Clients must consistently attend the program to ensure their ongoing enrollment (after three days of no attendance and no communication with JFS SPP, clients may be dismissed from the program). While none of the lots can accept registered sex offenders, only one has the added restriction of not enrolling individuals with active warrants. Beyond this, the barriers for entry to JFS SPP are low. Pets, as long as they are under the owner’s control and non-aggressive, are allowed at all program sites. Although drug use is prohibited on the lots, clients are not asked about or screened for substance use as a requirement for entry. At the time of our study, clients utilizing JFS SPP had to enter the lots between 18:00 and 21:00 and exit the lots by 07:00 the next morning, or 08:00 on the weekends, and had to adhere to quiet hours in the evening. At the time of our study, clients had access to showers at one of the lots but because of the limited facilities, the number of showers that could be taken each week was limited, and the total time for each individual using the shower space was 15 minutes.

While only one JFS SPP lot allows RVs, the program is open to all clientele and does not specifically serve one particular population. Across all four locations, JFS SPP can support 231 vehicles a month and currently staffs 23 people.⁸ Although the four sites vary somewhat in their offerings due to contracts, sources of funding, and specific site facilities, at the time of this research all of the lots had security guards, toileting, handwashing, Wi-Fi, some dinners and other donated food, staff support from 18:00 to 21:00, assigned case managers, and referrals to other programs and resources as needed. Case managers are assigned after clients establish consistent attendance, and this usually occurs one to two weeks after initial entry based on the capacity of the lot. Once a case manager is assigned, they are expected to have bi-weekly meetings with their caseload, although meetings can be more frequent depending on the level of need of the individual or family. Compared to other safe parking programs identified by McElwain et al. (2021), JFS SPP is one of the larger and more well-resourced safe parking programs currently in operation in the US, making it an ideal site to evaluate the effectiveness of such programs.

Methodology

Data collection for this project occurred from July of 2019 to September of 2022. During this time frame, our team used a mixed-methods approach to evaluate JFS SPP, combining quantitative analysis of client data with qualitative data from individual and group interviews with clients and staff. It additionally provided our team with more insights than would have been possible if relying on any one method alone. Thus, it enabled us to see big-picture trends and to probe for depth of understanding in terms of why certain clients have an easier or harder time moving from safe parking into housing or how the intervention assists clients in unique ways that distinguishes it from other types of homelessness services. To our knowledge, this is the first evaluation of the safe parking model to incorporate primary source data about the client experience. An additional strength of this work is its collaborative nature. This project was undertaken with support from, and in collaboration with, JFS SPP leadership. Part of the data collection for this study consisted of holding numerous discussions with the JFS SPP leadership team to ascertain their theory of change as well as the aims and constraints of their program. Our collaboration with JFS SPP additionally enabled us to work directly with their team in order to identify gaps in their knowledge that could be closed through research. It also paved the way for us to secure client-level data and it facilitated our interview recruitment efforts.

In addition to collaborating with JFS SPP, the project also included a pedagogical component. Each year, approximately 20 students at UC San Diego enroll in a two-quarter course series on homelessness co-taught by two authors of this paper. In the second quarter of the series, students in the course are trained to conduct and analyze oral history interviews and group interviews with clients in trauma-sensitive ways. Collaborating with students both expanded our data collection capabilities and informed the next generation of urban planners, affordable housing developers, social workers, medical practitioners, and social science researchers on the root and proximal causes of homelessness, the homelessness services ecosystem, and the effectiveness of various policy interventions.⁹ A major limitation of our study is that we could not compare JFS SPP client demographic data against demographic data for all individuals residing in their vehicles in San Diego County, simply because such data is unavailable. This limitation points to a need for more robust data collection within CoCs.

Quantitative Data Collection and Analysis

For the quantitative component of the study, our team analyzed data on JFS SPP clients from the Homelessness Management Information System (HMIS), a data repository system mandated by HUD that stores information on homelessness programs including program and client-level data. The San Diego Regional Task Force on Homelessness (SDRTFH), the organization that oversees San Diego’s CoC, manages HMIS data. A limitation of HMIS data is that its collection varies greatly from one jurisdiction to the next. Furthermore, program-level data on safe parking programs do not exist within the HMIS system, due to the lack of official categorization of such progrms at the federal level (McElwain et al., 2021). To clarify, while other organizations can categorize their program interventions as temporary shelters, street outreach, or safe havens, providers operating safe parking programs do not have a distinct program category from which to choose. As a workaround, JFS SPP collects data on its clients for HMIS but places such data into the “Other” category provided by HUD. This lack of precise categorization makes it difficult for comparisons to be made in terms of program success rates, both across safe parking programs and between safe parking programs and other interventions. An additional issue with HMIS data is that it is often incompletely reported, due to such issues as a lack of training for providers managing data entry (Community Solutions, 2023). This was certainly true when reviewing HMIS data collected by JFS SPP. However, because JFS SPP does enter data into HMIS to the best of its ability, it allowed our team to analyze client-level data from this program complete with the limitations that affect all HMIS data.

Our team specifically analyzed demographic and exit data for clients who were enrolled at the four SPP sites in operation between February 1, 2019 (when the organization that runs JFS SPP took full control of operating the safe parking lots for the City of San Diego), and March 31, 2021.¹⁰ We identified HMIS data for clients who self-identified as the head of their household¹¹ and who had been enrolled in at least one of the four sites through March 31, 2021. In addition to focusing on heads of household to simplify the analysis of our data (assuming that exits of heads of household reflect exit types for all members of their households), we additionally focused on demographic and program exit data collected the last time each head of household entered JFS SPP. While this approach simplified the analysis of our data, we accounted for the number of stays each head of household had in JFS SPP within both our descriptive and inferential statistics to capture how frequently households move in and out of the program as well as how such activity relates to various types of exits from the program. We used multinomial logistic regression to determine how the relative odds of having positive or negative exits from the program (versus no exit from the program) vary by client characteristics and time in the program.

We began by identifying 925 unique heads of household for which we had HMIS data. Reviewing these data, we found that 55% of all JFS SPP households who exited the program departed to unknown destinations.¹² It is typical among programs that assist people experiencing homelessness to have a significant portion of their clients leave to unknown destinations (HomelessData, 2020). For example, 48% of all bridge shelter clients in San Diego similarly exited these programs to unknown destinations (Focus Strategies, 2015). This reality creates a common data limitation for homelessness researchers, including our team. However, the JFS SPP rate was still comparatively high. This may be due to the fact that, within safe parking programs more generally, clients have greater potential for mobility due to vehicle ownership and can move on to other locations without informing staff. Due to this data limitation, we include descriptive statistics for all JFS SPP heads of households in the Online Appendix, but our analysis focuses only on heads of households with known exits (n = 362). Importantly, clients with known and unknown exits were very similar in terms of demographic characteristics. The few statistically significant differences suggest that clients with unknown destinations are younger, do not receive income from benefits, and are less likely to live with a chronic health condition or disability (see the Online Appendix). Thus, our sample likely skews more toward the experiences of older clients, clients who are more likely to receive income from benefits, and clients who are more likely to live with a chronic health condition or disability. This data limitation should be taken into consideration when reviewing the findings of this study.

Client exits are the outcome of interest for this analysis. We structured the dependent variable into negative and positive exits and used program continuation (or no exit) as the comparison category. We classified exits into permanent housing or temporary housing as positive and exits into sheltered or unsheltered homelessness as negative exits from the program.¹³

We made the deliberate choice to exclude emergency shelters and safe havens from our positive exit count for two reasons. First, we considered a move from a safe parking program to an emergency shelter to essentially be a horizontal move rather than a “step up” to a safer, more secure setting, as it would be for someone living rough on the streets. Second, many clients and former clients we interviewed had very poor opinions of local shelters. We make note of this decision because it resulted in a lower positive exit rate than would have been collected if we closely followed HUD’s guidance for categorizing program success for street outreach efforts.¹⁴ Our specific categorizations of positive and negative exits from JFS SPP can be seen in Table 1.

For our analysis, the following were structured as binary variables, where 1 equals the presence of the condition or characteristic: disability (all types), physical disability, both household variables (with and without children), veteran, mental health condition, chronic illness, and substance abuse. Gender and race are also binary variables, where female equals 1 (male = 0) and non-white = 1. Age is an ordinal variable of 10-year increments. Times homeless is a count based on the number of times a person reports being homeless as recorded in HMIS. Weeks in the program is a continuous variable, with the number of weeks a person was, or has been, enrolled in JFS SPP reported. We employed a multinomial logistic regression to understand the correlates of exit type in order to account for the reality that the two client exit types (positive or negative) and program stasis are not independent, or unrelated, outcomes.

Qualitative Data Collection and Analysis

To complement and expand upon our quantitative data analysis, our team conducted both individual and group interviews with JFS SPP clients and staff. Between September of 2019 and June of 2022, the authors, aided by students, conducted 180 oral history interviews with current JFS SPP clients across all four sites.¹⁵ One of the authors also conducted 69 interviews with former clients across all four sites between January and September of 2022. Oral history interviews with current clients and interviews with former clients were conducted to provide greater depth of understanding of the root and proximate causes of vehicular homelessness as well as clients’ perceptions of the strengths and limitations of JFS SPP. Interviews with former clients were also a chance to better understand what happens to clients after they leave the program as well as how program exits vary across client characteristics. In addition to individual interviews, two of the authors held six group listening sessions with clients across all four sites in the second year of the study, engaging with a total of 55 clients. The aim of these sessions was to invite people to share their perspectives on what helps or hinders their attempts to regain housing and stability, whether JFS SPP was helpful to them, and how services might be improved.

In order to provide additional “up close” perspectives on the challenges and benefits of JFS SPP, group interviews were held with staff at each of the four sites in the fall of 2020. We engaged a total of 15 staff, asking them to share their views on program effectiveness, barriers clients face to becoming rehoused, and staff needs and concerns. In September of 2021, we conducted a final round of “share out and feedback” sessions at each of the four JFS SPP sites with an audience of current clients, as a way of both vetting and deepening qualitative findings from the prior two years and assuring that we did not miss anything important. A total of 45 individuals from across the four lots attended these sessions, listened, asked questions, and offered their insights. Every individual who participated in an individual or group interview received a $25 gift card.

The authors used a general inductive approach to analyzing qualitative data, which accounts for both deductive and inductive analysis (Bryman & Burgess, 1994; Dey, 1993). We scanned for information that responded to our research questions and searched for deductive codes derived from the literature on homelessness and homelessness services. Beyond this, we read each transcript or set of notes with an eye to potential inductive codes. We used interviews with current clients collected in years 1 and 2 of the study to inform our interviews with former clients in year 3. The focus of this later set of interviews was not only on pathways into vehicular homelessness and the strengths and limitations of JFS SPP but also on trajectories into housing or sheltered or unsheltered homelessness upon leaving the lots.

Once coding was complete, we analyzed the coded transcripts based on our HMIS data analysis, examining which client demographic characteristics correlated with positive or negative exits from the program. In this stage, we analyzed the relationship between client characteristics on the one hand (specifically, the financial resources and social ties available to clients) and overall satisfaction with JFS SPP on the other. To help us with this process, we engaged the use of memoing, creating memos across a variety of topics including age, gender, race/ethnicity, and histories of homelessness (Birks et al., 2008). Analysis was also an iterative process. When themes were identified, we went back through the transcripts to ensure that they were consistent across the series of interviews (Noble & Smith, 2014).

Findings

Who Uses JFS SPP? Identifying Client Demographic Trends and Pathways Into Vehicular Homelessness

Our analysis of HMIS data offers evidence that strongly counters many negative stereotypes that exist regarding persons experiencing homelessness, including those around chronicity, widespread unemployment, and rates of mental illness and addiction (Buch & Harden, 2011). The majority of JFS SPP clients (73%) reported that they were experiencing homelessness for the first time.¹⁶ Contrary to common presumptions about high levels of psychiatric and substance use disorders among the unhoused population, JFS SPP clients have significantly lower rates of mental illness and substance use disorders compared to the general unhoused population in San Diego County and lower rates of addiction compared to the wider US population (see Tables 2 and 3).¹⁷ These findings are in alignment with a recent study out of Los Angeles that highlighted the specific demographic characteristics of individuals who live in their vehicles versus those who experience other forms of unsheltered homelessness (e.g., living in a tent), including the fact that they exhibit lower rates of chronicity (Giamarino, Blumenberg, et al., 2022).

Table 1. Classification of exits from safe parking used in this study.

Exit category	Exit examples	Exit demarcation for the study
Permanent Housing	Rental by Client, No Ongoing Housing Subsidy Rental by Client, with Subsidy Staying or Living with Family, Permanent Tenure Staying or Living with Friends, Permanent Tenure	Positive
Temporary Housing	Staying or Living with Family, Temporary Tenure Staying or Living with Friends, Temporary Tenure Transitional Housing for Homeless Persons	Positive
Sheltered Homelessness	Emergency Shelter Hotel or Motel Paid for with Emergency Shelter Voucher Safe Haven	Negative (This is in contrast to the Department of Housing and Urban Development’s categorization of exits to sheltered homelessness as positive exits from street outreach)
Unsheltered Homelessness	Place Not Meant for Human Habitation (Vehicle not in Safe Parking, Tent, Et Cetera)	Negative
Institutions	Psychiatric Hospital or Other Psychiatric Facility Jail, Prison, or Juvenile Detention Facility Substance Abuse Treatment Facility or Detox Center	Dropped exits to institutions from our data set (the Department of Housing and Urban Development categorizes exits to institutions as positive exits from street outreach, with the exception of exits to jail, prison, or a juvenile detention facility)

Note. Exit categories and examples come from the Department of Housing and Urban Development’s classification scheme.

Table 2. Descriptive statistics for Jewish Family Service of San Diego Safe Parking Program clients.

Characteristic	n = 362
Age (per 10 years)
0 to 10	0 (0%)
10 to 19	2 (0.6%)
20 to 29	27 (7.5%)
30 to 39	58 (16%)
40 to 49	58 (16%)
50 to 59	102 (28%)
60+	115 (32.1%)
Income/month ($100s)	11 (0:64) SD(10)
Physical disability	115 (32%)
Times homeless
1	265 (73%)
2	34 (9.4%)
3 or more	63 (17.4%)
Weeks in program	18 (0:107) SD(23)
Gender
Male	195 (54%)
Female	167 (46%)
Race
White	199 (55%)
Non-white	163 (45%)
Household type
Single adult	295 (81%)
Household without children	41 (11%)
Household with children	26 (7.2%)
Veteran	42 (12%)
Chronic illness	74 (20%)

Note. n (%); mean (minimum:maximum) SD(standard deviation).

While JFS SPP clients exhibit lower rates of mental illness and addiction than the general unhoused population in San Diego County, they do report somewhat higher rates of physical disability than the general US population. While 32% of JFS SPP clients live with some type of physical disability, only about 20% of the general population of American adults report a physically disabling condition (National Center for Health Statistics, 2020). However, according to a recent study by UC San Francisco’s Benioff Homelessness and Housing Initiative, over half of all persons experiencing homelessness in California struggle to work due to their age, health, or mental or physical disability (Kushel & Moore, 2023). In San Diego, approximately 63% of those experiencing homelessness live with a mental or physical disability (Regional Task Force on Homelessness, 2023). While the Benioff study and local data do not disaggregate types of disabilities, a recent report demonstrates that 57% of seniors experiencing homelessness in San Diego live with a physical disability (Serving Seniors, 2021). Thus, it is possible that rates of physical disability are lower for JFS SPP clients when compared to the general population experiencing homelessness in the San Diego County CoC, especially considering how many seniors are experiencing homelessness in this region (Serving Seniors, 2021). This possibility might reflect the fact that individuals with physical disabilities are less likely to own or operate a vehicle (Bureau of Transportation Statistics, 2022).

We found that a large proportion of JFS SPP clients (60.1%) are over the age of 50. This trend highlights a growing population of older adults experiencing homelessness in California in general and in San Diego specifically (Ibarra, 2023; Serving Seniors, 2021). In 2022, the Point-in-Time Count found that 24% of individuals experiencing homelessness in San Diego were over the age of 55 (San Diego Regional Task Force on Homelessness, 2022a). Thus, not only is homelessness among older adults growing in the region, but it is particularly high on the JFS SPP lots. While we do not have access to data on all individuals experiencing vehicular homelessness in the San Diego County CoC, we can speculate, based on the literature and our conversations with clients, that the higher proportion of older adults on the lots reflects the greater sense of safety they have over parking on a public street or staying in an emergency shelter (Serving Seniors, 2021).

Since the average age of JFS SPP clients trends older, this finding, combined with the fact that persons with disabilities are more likely to experience homelessness, may help to account for higher rates of physical disability among this group when compared to the general US population (Serving Seniors, 2021). However, as mentioned above, rates of physical disability may be lower across JFS SPP clients when compared to the population experiencing homelessness in the San Diego County CoC. Conversely, while approximately half of all adults in the United States live with a chronic illness as indicated by their doctors (Boersma et al., 2020), only 16% of JFS SPP clients identified as living with a chronic health condition. Because JFS SPP data is reliant on self-reporting, it is likely that, if data were collected through biometric markers or diagnoses from health professionals, the incidence of chronic illness would be higher, especially considering that experiences of homelessness produce negative health consequences (Baggett et al., 2010).¹⁸

While most heads of households utilizing the program are single adults (81%), 7.2% are members of families with children. This is important considering the fact that, across the entire unhoused, unsheltered population in the San Diego CoC, only 1% identify as individuals who are part of families, while families comprise approximately 20.8% of the entire population experiencing sheltered or unsheltered homelessness in San Diego (San Diego Regional Task Force on Homelessness, 2022a).¹⁹ Thus, our findings align with both national trends that see more families utilizing shelters, and those of Giamarino, Blumenberg, et al. (2022), who found that unsheltered families are more likely to reside in vehicles than in tents or other makeshift structures.

Gender distribution was another noteworthy finding. While 54% of JFS SPP clients identify as male, 46% identify as female.²⁰ The larger percentage of men on the lots reflects national and local trends regarding gender and unsheltered homelessness. However, the gender disparity is much smaller in the JFS SPP population. Women are represented within JFS SPP clientele at rates much higher than they are within the unsheltered, unhoused population in general both locally and nationally (de Sousa et al., 2022; San Diego Regional Task Force on Homelessness, 2019). This finding most likely reflects an increased sense of safety that women living out of their vehicles experience on the lots as unhoused women are more vulnerable to physical and sexual violence (Li & Urada, 2020).

JFS SPP clients additionally represent a diversity of racial and ethnic backgrounds including white, Hispanic, Black/African American, multiracial, Asian, American Indian/Alaskan Native, and Native Hawaiian/Pacific Islander. However, it is important to note that clients who identify as Black, American Indian/Alaska Native, or Native Hawaiian/Pacific Islander are represented on the JFS SPP lots at significantly higher rates compared to their percentages in the general local population (see Table 4).

Table 3. Mental illness and addiction rates for Jewish Family Service of San Diego Safe Parking Program clients as compared to other populations.

	Jewish Family Service clients	Individuals experiencing homelessness in San Diego Continuum of Care^a	General United States population
Mental illness	19%	21%	16%^b
Addiction (alcohol, drugs, or both)	1.4%	18%	16.5%^c

^a San Diego Regional Task Force on Homelessness (2020).

^b National Institute of Mental Health (2022).

^c Substance Abuse and Mental Health Services Administration (2021). National Survey on Drug Use and Health. Calculated for all individuals in the US aged 12 years old or older.

The racial disparities we see on the lots reflect a pattern seen across San Diego as well as the country (National Alliance to End Homelessness, 2020; San Diego Regional Task Force on Homelessness, 2022a) and point to the connections between structural racism and housing insecurity (Lurie et al., 2015). Yet while Black, American Indian/Alaska Native, and Native Hawaiian/Pacific Islander individuals are overrepresented within JFS SPP as compared to the general population, there is also an underrepresentation of Black/African American individuals in the JFS SPP program when compared to the total number of Black/African American individuals experiencing homelessness in the region. This aligns with the findings of Giamarino, Blumenberg, et al. (2022) that more white persons reside in their vehicles than in other unsheltered situations, highlighting that the connection between structural racism and resources extends into who is and is not in possession of a vehicle before or during their experience of homelessness, particularly for Black individuals, who have the lowest rates of vehicle ownership in the nation (National Equity Atlas, 2023). Indeed, according to the latest Point-in-Time Count, 19% of individuals experiencing unsheltered homelessness within the San Diego CoC identified as Black, compared to 14.2% of JFS SPP clients who identified as Black (San Diego Regional Task Force on Homelessness, 2023).

The current and former clients we interviewed represented a broad range of experiences, backgrounds, identities, and personalities. Yet a set of themes consistently emerged around the circumstances, conditions, and causes that precipitated people’s loss of housing and led to their subsequent experience with vehicular homelessness. In general, the stories shared by clients were consistently those of crises or “shocks” that overwhelmed the economic and personal capacities of the individual or family. These findings converge with a recent study on vehicular dwelling undertaken in Oakland, California (Pruss et al., 2022), and also complement the findings of a recent statewide survey of persons experiencing homelessness in California (Kushel & Moore, 2023). The factors most commonly cited as being the immediate cause of an individual’s homelessness include the loss of a job, a medical crisis, or loss of a significant person in an individual’s life due to a death, divorce, or domestic violence. In some cases, a cluster of these “push factors” occurred simultaneously. As Zoe,²¹ a 54-year-old biracial former client recounted for us, diminishing health resulted in her losing her job:

Because of the repercussions of liver failure, I have what they call brain fog. And I was just exhausted all the time. I couldn’t do anything because my body was failing and so I had to leave my job. Or actually my job would leave me because I wasn’t performing.

No longer able to work, Zoe lost her housing and experienced homelessness for the first time in her life. The majority of JFS SPP clients are, like Zoe, from San Diego. This finding, that homelessness has largely local origins, has been replicated both regionally (San Diego Regional Task Force on Homelessness, 2023) and statewide (Kushel & Moore, 2023).

Experiencing a crisis does not necessarily lead to the loss of one’s housing. Most people are able to weather traumatic life events if they have a buffer of savings, family, and/or broader social support that helps them to absorb the “shock” and bounce back. In contrast, the clients of JFS SPP whom we spoke with consistently described a lack of an economic and/or a social safety net. Reflecting the larger structural reasons for homelessness in the nation (Elliott & Krivo, 1991; Wolch et al., 1988), clients of JFS SPP find themselves experiencing deep economic insecurity at the time of their housing crisis. Additionally, clients either lack a social network, or the communal and familial ties that they do have are not able to keep them housed, a common trend among those experiencing homelessness (Burt, 1992). This is not for lack of desire to help; oftentimes, friends and family lack the financial resources to assist their loved ones in staying housed. We additionally learned through our interviews that clients often keep their situation a secret from family and friends, either out of feelings of shame or of not wanting to burden loved ones.

Not only were economic precarity and a lack of social ties key factors in explaining pathways into vehicular homelessness, but so too were regional housing costs. Close to a fifth (17.8%) of San Diegans live at or below the federal poverty line (Aurand et al., 2021), and there is a severe shortage of housing that is affordable to this group. According to the National Equity Atlas, 56% of households across the San Diego region were housing cost-burdened in 2019. According to HUD, households are cost-burdened if they spend more than 30% of their income on housing costs. The statistics were even more sobering when parsed out by race and ethnicity, as 61% of Latinx and 64% of Black households were found to be housing cost-burdened. While San Diego lacks housing in general to meet demand, its specific lack of affordable housing for low and very low-income residents is most pressing. Meanwhile, a dearth of affordable housing for moderate-income residents puts stress on the whole system, often prompting a pattern of “down renting” where households that could potentially afford higher rents wind up competing for lower income units (Corporation for Supportive Housing, 2019).

History of trauma was one other important factor contributing to homelessness that arose in many of our interviews. A significant number of clients and former clients we interviewed revealed life trajectories populated with experiences of abuse, neglect, and/or significant privation, both in natal homes and in foster care. For some, this early abuse was compounded by experiences of domestic violence in adulthood and/or the anxieties and challenges of post-traumatic stress. In addition to the ways that traumatic experiences can be precipitating factors for individuals experiencing homelessness (Substance Abuse and Mental Health Services Administration, 2022; Wiewel & Hernandez, 2021), they are also often an outcome of living unhoused. Clients spoke of living in constant anxiety while sleeping in their cars before entering JFS SPP, both over their physical safety and over fears of rejection and stigma.

Boosting Upward Cycles: Safe Parking Benefits Those Who Can Work, Use Social Ties, or Qualify for Assistance by Offering a Safe and Free Space While They Marshal Their Resources

Analysis of HMIS data reveals that, between February 1, 2019, and March 31, 2021, 40% of client households who had enrolled in JFS SPP between February 1, 2019, and November 30, 2020, and who had a known exit destination had positive exits to either permanent or temporary housing (30% and 9.9%, respectively; for a complete review of client exits see Online Appendix Table A1).²² A recent evaluation of San Diego’s temporary bridge shelters suggests that JFS SPP’s rates of positive exits into either permanent or temporary housing are significantly higher than the positive exit rates from San Diego’s shelter system. According to the study, the positive exit rate for the shelters under investigation was 9% (8% into permanent and 1% into temporary housing) (Focus Strategies, 2018), making JFS SPP’s overall positive exit rate four times higher than the overall positive exit rate for some of San Diego’s larger shelters. Client demographics for users of the temporary bridge shelters were very similar to those of JFS SPP clients in terms of household type, age, racial identity, and gender identity. However, it may be that differences in other client characteristics (i.e., differences in material resources such as vehicular ownership) account for these differences in positive exit rates rather than variances in programmatic approach.²³ For instance, while we cannot determine how many shelter clients owned vehicles, we can see that rates of disability for this group (physical and mental) were much higher than was seen on the JFS SPP lots.

One of the statistically significant associations discovered in our analysis was the connection between income and exits (see Table 5 for all significant associations). As a head of household’s monthly income increases by $100, the relative odds of having a negative exit from the program (versus no exit) decrease slightly, by 0.04%. While it is intuitive to suppse that an increase in income will prevent one from exiting JFS SPP into sheltered or unsheltered homelessness, through our interviews we discovered that differences in income varied significantly by age. Many younger clients we interviewed came to experience vehicular homelessness due to the loss of a job. Younger former clients we talked with would come into JFS SPP, find a new job while staying on the lots, and then use that money to move into housing. Oftentimes their incomes after finding new employment were significantly higher than those of many older clients living on fixed incomes due to age and/or ability. Many of these younger former clients additionally exited the program without notifying their case managers. Thus, while their exits were marked as to “destinations unknown,” many of the younger former clients we interviewed actually had positive exits from the program.

Table 4. Comparison of racial composition of Jewish Family Service of San Diego Safe Parking Program clients to other populations.

	Jewish Family Service clients	Individuals experiencing homelessness in San Diego Continuum of Care^a	General San Diego County population^b	Individuals experiencing homelessness in the nation^c	General United States population^b
White	52.5%	61.5%	43.8%	50.0%	59.3%
Hispanic	20.4%	31.1%	34.8%	24.1%	18.9%
Black	14.2%	24.5%	5.6%	37.3%	13.6%
American Indian/Alaskan Native	3.2%	1.9%	1.4%	3.4%	1.3%
Asian	3%	1.8%	12.9%	1.4%	6.1%
Native Hawaiian/Pacific Islander	1.5%	1.3%	0.6%	1.8%	0.3%
Multiple races	2.5%	3.9%	4.9%	6.1%	2.9%

Note. While our team collapsed racial and ethnic identity for our analysis, most established sources on local and national demographic information use separate race and ethnicity questions. Additionally, because of the problems inherent in disentangling race from ethnicity, we in all likelihood have an undercount of individuals who identify as Hispanic or Latino/a/x.

^a San Diego Regional Task Force on Homelessness (2022b). It is important to note that this data source asked separate questions for race and ethnicity. As a result, because “white, non-Hispanic” is not disaggregated, there may be an overrepresentation of who identifies as white in this count.

^b US Census Bureau (2020). Data were available on white, non-Hispanics.

^c de Sousa et al. (2022).

Table 5. Multinomial logistic regression results.

	Street homeless or emergency shelter		Permanent or temporary housing
Predictors	Odds ratios	p	Odds ratios	p
Intercept	1.4	0.742	3.23	0.183
Age (per 10 years)	1.01	0.926	0.9	0.334
Income/month ($100s)	0.96	0.016	0.99	0.33
Disability	0.89	0.744	1.01	0.985
Physical disability	1.63	0.138	1.41	0.282
Times homeless	1.1	0.16	1.07	0.295
Weeks in program	1	0.54	0.98	0.013
Gender (female)	1.88	0.046	2.14	0.012
Race (non-white)	0.87	0.649	1.08	0.789
Household without children	1.11	0.815	0.68	0.389
Household with children	11.81	0.024	8.26	0.049
Veteran	1.32	0.604	2.65	0.035
Mental health condition	0.89	0.722	0.9	0.746
Chronic illness	1.93	0.034	1.28	0.422
Substance abuse	1.01	0.994	1.74	0.522
Observations	362
R^2/R^2 adjusted	0.195/0.193

One such example comes from the story shared by a relatively young white client named Jim. Forty-eight years old, Jim came to JFS SPP unemployed. However, over the 37 days he spent at the lot, Jim got a job, saved money for a deposit, and was able to move into a studio apartment. For Jim and other younger clients, JFS SPP was invaluable: providing them with a safe and free space to park while they worked on increasing their income. No longer fearing for their safety on the streets, individuals like Jim could feel rested enough to look for and keep their jobs. Not needing to pay for a hotel or motel while figuring out their next move, many younger clients with the ability to boost their financial resources could then save for future security deposits on new apartments.

Gender was another significant factor in rates of positive exits. Women were two times more likely to have a positive exit from the program (versus no exit). Discussions with former clients point to the possibility that women tend to have greater social resources to draw from compared to men using JFS SPP. Indeed, many of the former male clients were able to rehouse on their own by increasing their income. Many of the women we interviewed did so as well, but they utilized an additional strategy of social connection, drawing on friends and family for assistance. One example of this trend is the story we heard from Ashley, a 60-year-old Black former client. Ashley had been living on $1,200 a month in income from benefits. With such a low income, she would have been unable to exit to permanent or temporary housing. However, she was able to leave JFS SPP after 117 days due to her strong social network. As Ashley shared:

My rabbi not only had kept me in a Motel 6, I mean paid every week, so I knew I had a roof over my head. She paid my incoming expenses and paid my rent toward, you know, to get me into housing. And then do you want to hear a little miracle? A week and a half later and she said you want a free car? I had been praying for a free car. I’m serious. I had been praying for a free car and I get this text message saying, “do you want a free car?” So, I just typed back yes. It turns out someone [in my faith circle] they were giving it away for free and buying a new car. Wow. So, I no longer had to rent a car.

For Ashley, deep social ties to members of her religious community not only helped her into housing, but enabled her to eliminate the car payments she was making and consequently boost her savings.

In addition to fostering upward cycles for those who can work or who have resources through social ties, JFS SPP benefits individuals who are more likely to receive housing assistance, including veterans and families with children (Balagot, 2019). A veteran was 2.65 times as likely to have a positive exit from the program (versus no exit). Households with children were eight times more likely to have a positive exit (versus no exit). Through our interviews with former clients, we saw that all veterans who exited the program received housing assistance specifically set aside for veterans. Similarly, many of the families with children headed by single mothers whom we interviewed exited JFS SPP to a transitional housing program designated specifically for this demographic.

Preventing Downward Spirals: Safe Parking Benefits Clients With Fewer Resources and Provides an Alternative to Outreach and Shelters

Through their descriptions of the strengths of JFS SPP, clients highlighted the benefits of this model beyond housing placements and often explicitly or implicitly stated their preference for this intervention over other services. A primary strength of JFS SPP that emerged across all of our interviews was that JFS SPP gave clients a sense of safety in a time of upheaval. The availability of a safe, sanctioned, fenced, overnight parking lot patrolled by a security guard enabled clients to avoid both ticketing by the police and potential vehicle break-ins. As Shania, a 40-year-old client who identified as Black, told us: “It just feels safe. You can sleep knowing that you’re not going to be harmed or get arrested or anything like that.” Additionally, many clients noted that JFS SPP was a welcome alternative to the shelter system, which many found to be dangerous. As Shirley, an 80-year-old white client who had used shelter options in the past, told us: “My case manager asked me if I wanted to go to a shelter and I said no. They are all mostly downtown and are dangerous! My grandson stayed at one downtown and he got beat up and had his ID stolen.” Thus, while JFS SPP was perceived by all clients as safer than sleeping on city streets, it was also perceived as safer than local emergency shelters, where interviewees felt that their property and physical safety were constantly in jeopardy.

It is important to note that we asked former clients about their experiences with other homelessness services, specifically emergency shelters. Approximately half of these individuals had used emergency shelters in the past and compared those experiences to their experiences on the JFS SPP lots. Yet while we rely on these interviews specifically to document a preference for JFS SPP over emergency shelter options, it is important to highlight that this finding may be the product of self-selection, as we were only able to interview individuals who had used both shelters and safe parking but who had chosen and were currently using the latter.

Other strengths identified by clients can be used to compare JFS SPP to street outreach efforts and demonstrate the specific utility of safe parking lots. One particular strength highlighted by many JFS SPP clients was the program’s ability to provide basic needs such as showers, restrooms, dinners, and Wi-Fi. JFS SPP clients highly valued this access to basic necessities. Many shared that not having to attend to core needs including their physical safety and access to toilets, showers, and food enabled them to focus on other important tasks such as improving their physical and mental health, securing or maintaining employment, and finding housing. Having all of these resources located at one site where clients can safely spend the night made their lives easier than if they were reliant on street outreach services alone. While street outreach efforts can point individuals to available showers, food distribution centers, and toilets, such resources may be more dispersed across the region and require transportation to access them. Indeed, as many clients told us, “Being homeless is a job.” Before entering JFS SPP, many clients recounted long hours spent running around locating bathrooms, using showers at their gyms (if they were lucky enough to have gym memberships), and trying to locate food pantries. Although none of the former clients explicitly brought up street outreach in their interviews, their descriptions of life before the lots revealed the difficulty of securing all of one’s basic needs when resources were not located in one place. While emergency shelters offer similar centralized resources, as mentioned above, many individuals found JFS SPP to be a safer alternative.

Numerous clients also emphasized the importance of JFS SPP’s pet-friendly policy, framing it as a basic need that allowed them to keep their companions. This is in contrast to most emergency shelters, which prohibit pets. Not only did JFS SPP allow pets, but case managers would often assist clients in accessing services for their animals. As Julia, a 41-year-old biracial client told us, her case manager helped her secure affordable veterinarian assistance for her dog before he passed away:

My case manager…she gave me her number and I was able to call her one time when I needed proof that I was homeless. I was able to call her up one time and she was able to communicate with the vet that I was trying to get my dog help with the vet or whatever…she was able to talk to the vet and get the pricing for homeless dogs. My dog ended up dying anyway, but she was able to do that for me and I was grateful to her for that.

Not only does JFS SPP allow pets, it actively assists clients in caring for their animals, who have become important companions. For these reasons, JFS SPP was praised by many pet-owning clients who had previously run into barriers using shelters in the region that would not allow them to bring their animals into those programs.

Social connections were an additional benefit of JFS SPP identified by clients and staff alike. Friendly interactions with staff generated a sense of belonging, community, and comfort. These positive social interactions on the lots were often contrasted with the isolation individuals had felt in emergency shelters due to negative interactions with staff members. As Peter, a 63-year-old white client, told us about entering JFS SPP:

It was great. I went there [JFS SPP] with my friend, we’ve been friends for 20 years, and they welcomed us with open arms. They were really nice and toured us around the program and told us about the rules. And the food was wonderful. Down there, they made you feel like a person, not like the scourge of the earth.

From Peter’s accounts, positive interactions with JFS SPP staff helped buoy him while he waited for his Housing Choice Voucher, which did eventually arrive. Peter, who is now in permanent housing, had come to the lots from a shelter. He made it clear that his experiences with the staff at the shelter and at JFS SPP contrasted sharply. While the former treated him disrespectfully, informing him that if he did not follow the shelter rules he would be “kicked out with no place to go,” the latter welcomed him with open arms.

Shedding light on what may account for perceived differences in staff interactions is the fact that many clients highlighted that the rules they were meant to follow in emergency shelters often left them feeling infantilized (such as mandatory chores). By contrast, at JFS SPP, clients highlighted that, while there were certain rules they needed to follow, they seemed fair. As Cathy, a 56-year-old white client, told us:

…the big thing is that I felt like they [JFS SPP staff] cared about us. I felt like they gave us respect and dignity and…even though they have rules, the rules weren’t so overbearing that we felt like people were trying to make us slip up on rules… like when there’s rules and they [staff at a program] go out of their way to find out if you’re breaking them. That’s what I find [happens] a lot in the [shelters].

From Cathy’s perspective, the rules inherent in the shelter system create an opportunity for staff to police clients’ behaviors. This results in some staff at shelters trying to “spice things up because they’re just going to either make trouble or because they’re bullies.” While rules at shelters may result in certain staff members using them to assert authority, it may simply be the presence of more restrictive policies that make interactions with shelter staff feel less friendly. By contrast, at JFS SPP, fewer rules may create a better dynamic between clients and staff, boosting client morale and sense of belonging.

JFS SPP particularly benefits clients who are unable to work due to age or ability, who lack social ties, or who are not offered housing assistance by providing a safety net, one often preferred to emergency shelters. Individuals who remain on the lots longer than others continue to feel safer using the lots than they would using emergency shelters.²⁴ When Shirley shared her story with our team, it encapsulated many of these themes. Retired and living on a limited fixed income of $1,700 a month from savings and Social Security, Shirley earns considerably more than other older clients, many of whom are living on Supplemental Security Income alone. Yet, on this income, Shirley does not feel that she can move into housing without assistance given the high rents in San Diego. While Shirley lamented that she has not been prioritized for housing assistance through coordinated entry,²⁵ she further bemoaned the fact that she has no social ties that are able to help her. Her family is not an option. In fact, as Shirley told us, the reason she came to experience homelessness for the first time in her life, at the age of 76, was because her granddaughter, whom she had been living with, physically assaulted her. While her granddaughter was serving time in jail for this attack, her boyfriend kicked Shirley out of the house. As mentioned above, Shirley declined her case manager’s suggestion of using a traditional emergency shelter because “they are mostly downtown and dangerous.” As such, she has been at JFS SPP for almost three years.

Despite the shelter aversion we found in our interviews with former clients, some clients are more likely to exit to shelters. Women, households with children, and individuals with chronic health conditions all have higher relative odds of exiting to street homelessness or an emergency shelter from JFS SPP compared to not exiting the program. Through our interviews, we discovered that these populations were moving into shelters as opposed to leaving JFS SPP for street homelessness. While women and households with children may exit to shelters because they find shelters to be safer or more conducive to the promotion of good health (i.e., women may consider shelters safer than safe parking programs because of the significant sexual and physical violence they face, and families may consider the health benefits of sleeping inside a building to be more significant than the privacy they get at a safe parking lot), persons with chronic health conditions may be driven to leave the lots for shelters due to their health. However, with Shirley, as with other older adults, it may be that fear and age are what keep her away from emergency shelters despite her gender and deteriorating health, as such shelters are often perceived to be dangerous by seniors (Serving Seniors, 2021). Indeed, most individuals who expressed that the shelters they had stayed in prior to JFS SPP were dangerous were over the age of 60.

Expanded Hours and Resources Would Benefit All JFS SPP Clients

Notwithstanding the strengths of JFS SPP identified above, many themes emerged in our interviews that highlighted ways in which the program could be improved to benefit all clients. Mandates around entering and exiting the lots were consistently identified as a source of aggravation and a limitation of the program. For clients working unconventional shifts that started mid-afternoon and ended late at night, getting onto the lot by 21:00 pm was an impossibility. As Steve, a 54-year-old white client told us:

The restrictions were very difficult because they had an entrance time like you had to be in there by six o’clock or eight o’clock or something and I work nights…I am not entitled to work protection laws. [My job] can make me work whenever they want, however many hours, and I think the people running the program, they couldn’t fathom that…. They wanted my work schedule. I don’t have a work schedule. I may work eight in the morning until five one day and I could come in from 12 to nine the next day.

While clients could seek an exception to this rule at some of the lots, as became the case for Steve, the need to be on the lots by a certain time may have dissuaded people with nontraditional work hours from using the program. Additionally, the requirement that clients leave by 07:00 was a burden for many people, especially those with children, older adults, individuals without employment, and/or people dealing with a disability. These clients had to find other places they could go to in the early morning, when most public places, like libraries, were not yet open. They also had to cover the added expense of costly gas to drive to off-site locations during the day and return to the lot at night. Many clients in this position described their morning routine as one of driving to the closest nearby park, where their children could play or where they could sit and rest in their cars or on park benches.

Beyond expanded access to the lots, many clients and former clients we spoke with expressed a desire that JFS SPP increase their capacity to provide clients with showers. Many interviewees bemoaned the fact that showers were only offered a few times a week. Parents noted that the 15-minute limit on showers made it very difficult for them to bathe their children properly. Clients with physical disabilities additionally lamented the 15-minute limit on showers. As Charlie, a 55-year-old Pacific Islander former client, told us: “I am thankful that [JFS SPP] had a handicap shower that I could utilize or even somebody that was in a wheelchair could utilize. Because showering, that is challenging in and of itself, especially when you only have 15 minutes to take a shower.” Thus, from Charlie’s perspective, being able to shower for longer would allow all clients, regardless of their ability status, to feel that they could adequately maintain their hygiene.

Beyond expanded hours and access to showers, clients expressed a desire to meet with case managers more frequently. Case managers were not assigned to the clients we spoke with until one to two weeks post-entry to the program. For Amanda, a 47-year-old white single mother of three, this general policy left her in an impossible bind. She had just gotten off of the list for a Housing Choice Voucher and desperately needed to talk with someone about navigating affordable housing options and landlords who might or might not accept her application. Yet because she had just entered JFS SPP, she was told she had to wait 14 days until she could speak to her case manager. And while case managers are expected to hold bi-monthly meetings with the individuals on their caseloads, some clients and former clients noted that they were only able to meet with their case managers twice, despite being enrolled in the program for months. Not only did clients want increased engagement with their case managers, they wanted consistently knowledgeable assistance. Knowledge of resources and procedures on the part of case managers and staff seemed to increase levels of motivation and engagement among clients. While many case managers and staff were praised for their insights into the service ecosystem, clients shared that they sometimes received conflicting information from staff. Veronica, a 50-year-old Latina client, shared that one staff member gave her incorrect information. She and her husband were told that they could get work done on their vehicle and that the cost would be covered by JFS SPP. When she asked her case manager about this offer, she was informed that this was not true. The lack of consistency in staff knowledge meant that Veronica and her husband’s hopes were raised and then dashed, causing great frustration.

Furthermore, not all interactions with staff were positive, and the negative interactions mentioned in interviews had significant deleterious effects on clients, causing negative spirals rather than positive movement toward becoming stably rehoused. While the establishment of trust between a client and a provider served to bolster, motivate, and connect clients to resources, by contrast, mistrust resulted in feelings of hopelessness and slowed the process of being connected to services. As Ashley shared with us, negative interactions with staff erode trust and disconnect clients from providers (and thus resources). Such interactions can also be interpreted by clients of color as arising from bias. As Ashley said:

If you are too smart for your own good, then you will be complained about as aggressive or told that you are being hostile. In fact, I actually got kind of a veiled verbal threat from [one staff member]. She was so nasty to me and I’m not a nasty person. I know how to use my words, though. And I can use my words well. I don’t have to curse at you and scream and I told her that I knew what was what and so then you know…it’s terrible because I’m a woman of color. So, what’s the first thing that gets done to a woman of color when she speaks truth to power? She’s labeled as an angry Black woman. So, what I got was someone from the head department come to me and basically tell me that I needed to get a different attitude or I was going to be removed from the program.

Discussion, Conclusion, and Policy Implications

Safe parking programs are playing an increasingly important role in the homelessness servicescape. This will likely continue as the growth trend in vehicular homelessness continues, reflecting structural pathways into homelessness for those with limited financial and social resources, especially older adults living on fixed incomes yet still in possession of a vehicle. We have identified three key findings from the research described here: (a) safe parking programs bolster upward cycles for those with the capacity to work, draw on social ties, or access housing assistance; (b) safe parking programs prevent downward spirals for those with fewer resources by acting as safety net, are preferred as a safety net by clients who have used both safe parking programs and emergency shelters over the latter, and provide greater access to resources than is possible through street outreach efforts; and (c) expanded access to lots, case management, and showers would benefit all clients of safe parking programs. These key findings are at the root of two policy recommendations described below. In addition to these recommendations, we also encourage more research on vehicular homelessness in general and safe parking programs in particular until HUD mandates universal data collection efforts.

Policy Recommendation 1: Expand Services, Operations, and Capacities at Safe Parking Lots

Extrapolating from the findings that older JFS SPP clients and others with fewer resources utilize the lots for extended periods of time, we recommend that safe parking program providers consider how they will meet the needs of such clients. Throughout our interviews we learned that many individuals desired more contact with case managers. Increasing the number of case managers would result in greater support for individuals on low, fixed incomes, helping them to identify new income streams from untapped benefits programs as well as support determining eligibility for various housing programs, including permanent supportive housing or rapid rehousing. Furthermore, given that individuals who lack income, vouchers, or social ties with resources are likely to remain on the SPP lots for extended periods of time, and given that the lots offer a place of physical, emotional, and psychological safety for all clients, we strongly recommend that safe parking programs seek funding to operate 24 hours a day, seven days a week. Continuously open lots allow clients with nowhere to go during the day—including families with children, senior citizens, and persons with physical disabilities—to stay in a safe space where they can easily connect with their case managers, save money on gas and, in some instances, have access to free Wi-Fi. Twenty-four-hour lots would also benefit clients with nontraditional work schedules as they would no longer need to worry about entering and exiting the program.

Another recommendation is that safe parking programs invest in their ability to continuously train staff. Regular training is a way to keep staff up-to-date about available resources so that clients do not receive conflicting information. Training that encompasses justice, equity, diversity, and inclusion (JEDI) will be critical to fostering cultural humility and improving conditions for all clients, but particularly clients of color (San Diego Regional Task Force on Homelessness, 2022b). Finally, we encourage safe parking programs to dismantle other barriers to utilization when possible, including limits on oversize vehicles. Current research demonstrates that people living in RVs are not only excluded from most safe parking programs, but they specifically fear the loss of their vehicle—a fear that prevents them from seeking out services (McElwain et al., 2021).

Policy Recommendation 2: Increase Funding and Standardize Benchmarks for Safe Parking Programs by Creating an Official Federal Classification

While our first policy recommendation focuses on the safe parking program level, our remaining recommendations call for action at the city, county, and federal levels. Interest in safe parking programs is growing, and our research demonstrates the positive impact such programs can have on individuals as they grapple with a housing crisis. Our data show that positive exits into temporary or permanent housing from JFS SPP lots are twice as high as those from emergency shelters in the same CoC: a finding that may reflect differences in disability rates and abilities to increase earned income across these groups. For those unable to exit to housing in the short term, including many older adults and persons with physical disabilities, safe parking programs offer a safe space. Yet despite the benefits of JFS SPP, their capacity to serve individuals and families residing in their vehicles falls significantly short of the need in San Diego County. Recently, elected officials in several of the smaller jurisdictions in San Diego County introduced proposals to consider establishing safe parking programs. Similar discussions are happening in other parts of the state and elsewhere in the country. Securing adequate resources for these programs may be challenging since they are not currently eligible for HUD funding. However, legislation has recently been proposed by three California lawmakers that would change this exclusion. The Safe Parking Programs Act, H.R. 2956 and S. 3788 (Section 301),²⁶ were introduced in 2021 and 2022, respectively, and would require HUD to provide grants of up to $5 million to local governments to start or expand safe parking programs.²⁷ These additional funds could be used to operate 24/7 lots and to provide ongoing training to staff. We recommend federal-level support from HUD to provide these funds to local jurisdictions and that such support occur in tandem with official classification of safe parking programs, so that federal funding can be earmarked for these initiatives and their effectiveness systematically tracked. We also recommend that elected officials at the city and county level continue to advocate for the creation of safe parking programs, even as we await the availability of increased federal funding.

Beyond our formal findings, we argue that understanding the driving forces of vehicular homelessness should prompt policymakers and elected officials to engage in proactive strategies that will reduce the number of people falling into homelessness in the first place. The provision of shallow rental subsidies is one such strategy. San Diego County recently piloted a program providing shallow rental subsidies of $500 to low-income senior citizens for up to 18 months (Acevedo, 2022), a decision that was based on recent research that many older San Diegans only require a few hundred extra dollars a month to remain housed (Serving Seniors, 2021). In addition to programs that address economic precarity, we should consider ways to increase social ties for individuals experiencing housing insecurity or homelessness. One program that has been in existence since the 1980s, Bridge of Hope,²⁸ connects persons exiting homelessness with multiple sponsors who commit to sharing time, knowledge, and connections, thus expanding the social support network of individuals exiting homelessness. Upscaling such efforts could help many individuals who either lack social networks or have social ties that are unable to help them financially. Finally, efforts should be made to identify, preserve, and connect individuals to affordable housing options that currently exist, albeit in limited numbers, while advocating for and awaiting new construction. Such efforts can include the preservation of single room occupancy hotels (SROs) and the conversion of motels into affordable units, as is being advocated for in the City of San Diego (Layne, 2022; Shaw, 2021). Other city and county governments can take up similar initiatives beyond San Diego. Beyond immediate efforts to increase affordable options, structural changes must occur at the federal level to ensure a long-lasting increase in affordable housing. These efforts can include increasing funds for Housing Choice Vouchers, public housing projects, and Community Development Block Grants.

Research Recommendation: More Collaborative Research Must Be Conducted on Vehicular Homelessness Until HUD Mandates Better Data Collection

Beyond our policy recommendations, we urge researchers to make further inquiries into vehicular homelessness in general and safe parking programs in particular. A major strength of our work is that we are an independent research team conducting an evaluation of a safe parking program with the support of a willing community partner. Thus, we could engage in structured data collection and objective analysis of both quantitative and qualitative data, in contrast to outcome data that is based on agencies’ self-reporting to CoCs. Numerous beneficial outcomes have arisen from our evaluation, including the successful opening of a 24/7 JFS SPP lot based on our recommendation to the San Diego City Council (City News Service, 2022); better data collection on JFS SPP client exits; positive student learning outcomes from the courses related to this project; and the creation of a community of practice that encompasses safe parking practitioners, former clients, and researchers (Lewis et al., 2022). Given these positive outcomes, we recommend that similar research collaborations on vehicular homelessness and safe parking be undertaken between academic institutions and service providers. Such research collaborations can serve as an interim measure for collecting CoC-level data on vehicular homelessness before it is mandated by HUD and can additionally help foster a better understanding of the effectiveness of different safe parking programs—as they vary in size, amenities, and rules—until safe parking programs are officially categorized and given universal benchmarks for success. The availability of such data would facilitate our further understanding of vehicular habitation and the uniqueness of safe parking clients. Research collaborations of this nature can additionally continue to evaluate how safe parking compares to other programs for those experiencing homelessness.

Acknowledgments

We thank our community partner, JFS SPP, for their willing collaboration and facilitation support over three years of data collection. The breadth and depth of this research would not have been possible without the rich contributions of students in our accompanying courses on homelessness. Todd Levinson provided valuable research assistance in the early phases of the project. We also thank Awanti Acharya, Jennifer Nations, and Nathaniel Gonzalez for their help with the quantitative analysis for this work. Thank you to the three anonymous reviewers for their thorough and helpful comments and suggestions. Finally, we are grateful to the more than 300 individuals who shared, through surveys, interviews, and listening sessions, details about their lives, the challenges they are facing, and the insights they have gained through their experiences of vehicular homelessness in Southern California. It is our deepest hope that our findings will contribute to effective policies and helpful services for people across the country experiencing homelessness and housing precarity.

Disclosure Statement

The authors report that there are no competing interests to declare.

Additional information

Funding

This work was funded by a generous gift from philanthropists Hanna and Mark Gleiberman.

Notes on contributors

Leslie R. Lewis

Leslie R. Lewis, PhD, MPH, is a Continuing Lecturer in the Department of Urban Studies and Planning at UC San Diego, and the Director of Education and Community Engagement at the Homelessness Hub. Her research interests include health equity, homelessness, and democratizing research.

Mirle Rabinowitz Bussell

Mirle Rabinowitz Bussell, PhD, is an associate teaching professor in the Department of Urban Studies and Planning at UC San Diego and is the faculty director of the Homelessness Hub. Her research interests include affordable housing, homelessness, and place-based philanthropy.

Stacey Livingstone

Stacey Livingstone is a postdoctoral fellow at the Homelessness Hub in the Department of Urban Studies and Planning at UC San Diego. Her research interests include homelessness and community based action research.

Notes

1 A Continuum of Care is a regional planning body that coordinates services for people experiencing homelessness (National Alliance to End Homelessness, 2010).

2 The Department of Housing and Urban Development classifies individuals residing within their vehicles as “unsheltered,” along with individuals who reside in tents or constructed shelters located in such areas as public streets or parks (de Sousa et al., 2022).

3 Safe parking programs generally draw funding from city and county program funds, private donors, and Department of Housing and Urban Development Homeless Prevention Funds (Homelessness Policy Research Institute, 2018).

4 The Department of Housing and Urban Development’s destination classifications can be viewed in their System Performance Measure 7 tool, at https://www.hudexchange.info/resource/4966/system-performance-measure-7-destination-classification/#:∼:text=For%20street%20outreach%20projects%2C%20nearly,to%20some%20form%20of%20shelter.

5 The Point-in-Time Count is conducted annually across all Continuums of Care in the United States. It captures the number of individuals experiencing sheltered and unsheltered homelessness (de Sousa et al., 2022).

6 It is important to note that, while the milder weather in California makes vehicular residency a possibility, it does not draw people who are experiencing homelessness into California. The vast majority of people experiencing homelessness in California lost their housing while living in the state (Kushel & Moore, 2023).

7 2-1-1 is a local nonprofit in San Diego that connects individuals to resources including shelters, housing assistance, and food assistance.

8 Jewish Family Service of San Diego Safe Parking Program will soon open two new lots, increasing their capacity. When we conducted our study, one of the lots had to reduce capacity to additionally operate an isolation tent for COVID-19-positive individuals. During this time of reduced capacity, the Jewish Family Service of San Diego Safe Parking Program served an average of 183 vehicles per month.

9 While the benefits of a community-initiated, student-engaged research model have been well documented (Greenberg et al., 2020), we will leave a review of our own collaboration and its myriad benefits and beneficiaries for other writings and publications.

10 An exit from the program refers to a client’s destination upon leaving the Jewish Family Service of San Diego Safe Parking Program.

11 A household can be any one of the following: (1) a single adult, (2) adults but no children, or (3) adults and children.

12 In the first year of this study, 70% of Jewish Family Service of San Diego Safe Parking Program clients exited to unknown destinations. Upon informing Jewish Family Service of this statistic, this number dropped significantly to 28.6% as the program collected better exit data.

13 Moves into an institution such as a jail or a psychiatric hospital (which were few in our sample) were dropped from the data set.

14 A review of the provider self-reports of program success compiled in McElwain et al.’s (2021) survey indicates that many safe parking programs consider exits into temporary or permanent housing to be positive. This reveals a strategy of denoting safe parking programs’ success rates in a similar fashion to the denoting of success rates for street outreach efforts, that, following the Department of Housing and Urban Development’s strategy for determining program success, consider essentially all moves from unsheltered homelessness to be positive exits, including those made into emergency shelters, safe havens, and temporary housing arrangements.

15 In the first year of the study, only the three original sites were visited (the fourth site in Encinitas was not yet in operation). In the second year of the study, all four Jewish Family Service of San Diego Safe Parking Program sites were visited. In the third year of the study, only the three original lots were again visited due to time constraints. Approximately 25 to 50% of clients using a particular lot on a night we were conducting interviews participated in the study.

16 Homeless Management Information System questionnaires only collect data on experiences of homelessness over the prior three years; it is possible that some individuals had episodes of homelessness prior to this.

17 While some of the low substance use rates may be attributed to the requirement that clients do not use drugs on the lots, it is important to highlight that the Jewish Family Service of San Diego Safe Parking Program is a low-barrier service. Clients are not asked about their substance use nor are they screened for substance use as a precondition for using the program. As such, it is possible that this rule does not preclude the majority of individuals living out of their vehicles who may additionally live with substance abuse issues from using the program.

18 Persons experiencing homelessness face obstacles to utilizing healthcare, which make it more difficult for diagnoses to occur (Baggett et al., 2010).

19 According to the Department of Housing and Urban Development, most families experiencing homelessness across the United States are considered “sheltered,” staying in emergency shelters, transitional housing programs, or safe havens. The Department of Housing and Urban Development considers a family household to be one with at least one adult over the age of 18 and with at least one child under the age of 18 (de Sousa et al., 2022).

20 Only six Jewish Family Service of San Diego Safe Parking Program clients for which we had Homeless Management Information System data had a transgender or nonbinary gender identity.

21 All persons we interviewed are referred to by pseudonyms to maintain confidentiality. We created random pseudonyms based on the suggestions of Lahman et al. (2015) that preserve the cultural significance of names when appropriate. All demographic information of interest to us in this study (i.e., income, age, gender, race) are included in descriptions of our interview participants unless they did not provide certain information to us.

22 We used a time frame of at least four months for Jewish Family Service of San Diego Safe Parking Program clients to exit from the program.

23 Homeless Management Information System data does not track vehicular ownership. Furthermore, the study on shelters we use here for comparison of positive exit rates combined vehicular homelessness—either before or after using a shelter—with all other forms of unsheltered homelessness (Focus Strategies, 2018). As such, we do not have a good sense of how many shelter clients own vehicles.

24 Many clients we interviewed who cannot rehouse on their own due to limited income or who are not prioritized for housing assistance remained on the lots and were interviewed by our team periodically over the three-year study. Indeed, as we saw in our Homeless Management Information System analysis, for every additional week in the program, a head of household’s relative odds of having a positive exit (versus no exit) decreased by 2%.

25 Coordinated Entry Systems match individuals experiencing homelessness with housing assistance based on availability and need (HUD, 2017).

26 The full Senate bill is available at: https://www.govinfo.gov/content/pkg/BILLS-117s3788is/pdf/BILLS-117s3788is.pdf.

27 An overview of the proposed legislation is available at: https://endhomelessness.org/legislation/safe-parking-programs-act-h-r-2965/?gclid=Cj0KCQiA_bieBhDSARIsADU4zLfwZkdBXlW70c_oHOWvIX9iuCdxxUj4Pa9M17yXr2IDb12biHxrH2IaAn8WEALw_wcB.

28 Bridge of Hope offers a faith-based model of connecting volunteers with their unhoused neighbors. Volunteers undergo training and then offer tangible and emotional support and encouragement as they engage with families who have exited homelessness: https://bridgeofhopeinc.org/our-model/.