Debtors’ Blocks: How Monetary Sanctions Make Between-neighborhood Racial and Economic Inequalities Worse

Abstract

Although recent scholarship has enumerated many individual-level consequences of criminal legal citations and sentences involving fines and fees, we know surprisingly little about the structural consequences of monetary sanctions or legal financial obligations (LFOs). We use social disorganization and critical race theories to examine neighborhood-level associations between and among LFO sentence amounts, poverty, and racial and ethnic demographics. Using longitudinal data from the Washington State Administrative Office of the Courts, and the American Community Survey, we find LFOs are more burdensome in high-poverty communities and of color, and that per-capita rates of LFOs sentenced are associated with increased future poverty rates across all neighborhoods.

INTRODUCTION

Poverty and racial inequality are pernicious and persistent dimensions of American life. People who are poor and BIPOC (Black, Indigenous, and people of color) have worse education outcomes, shorter life expectancies, higher rates of felony conviction and incarceration, and lower rates of social mobility than their affluent and White counterparts (Sampson and Wilson 1995; Wildeman and Western 2010; Williams 2014). Furthermore, poverty, inequality, and the accumulation of disadvantage that accompany them are concentrated in resource-poor neighborhoods¹ where residents’ access to educational resources, labor markets, and financial institutions that could improve their lives is often limited (Kawachi, Kennedy, and Wilkinson 1999; Sampson 2009). Economic change (W. J. Wilson 1987), historic and contemporary racism (Bonilla-Silva 1997; Golash-Boza 2016), institutional economic practices (Massey, Eggers, and Denton 1990; Oliver and Shapiro 1995), and cultural assumptions and norms (Massey and Denton 1993) have helped create racially segregated communities and neighborhoods. Subsequently, the isolation of poor and differently racialized neighborhoods (i.e., neighborhoods with a larger share of BIPOC residents) exacerbates these disadvantages, locking communities into dynamics that reproduce poverty and inequality (Massey 2007; Pattillo-McCoy 1999).

One omnipresent institution that has invaded the lives and communities of BIPOC and people who are poor is the American criminal legal system. Black, Native American, and Latino people have disproportionately borne surveillance, convictions, and incarceration in the era of mass conviction. Incarceration has emerged as a salient experience in the life course for young Black men, one in six Latino men born in 2001 can expect to go to prison, and Native Americans are re-incarcerated for parole violations at nearly triple the rate of White people (Hartney and Vuong 2009; Mauer 2011; Pettit and Western 2004).

The system of monetary sanctions, or legal financial obligations (LFOs), is a growing part of the criminal legal system (Harris 2016). Borne from similar social, political, and economic forces that generated the prison boom (see Gilmore 2007), this sentencing scheme allows judges to impose financial penalties on people processed and convicted (Martin et al. 2018). For people too poor to pay, the sentence is transformed into penal debt (Atkinson 2017). We examine the relationship between the state-imposed system of monetary sanctions and the reproduction of neighborhood-level poverty and inequality. We use critical race theory (CRT) to build on and critique a popular framework in studies of residential inequality—social disorganization theory. In doing so, we demonstrate how CRT highlights and attenuates social disorganization theory’s failure to reckon with systemic racism, and show how LFOs, a deliberate state intervention, perpetuate racial and economic inequalities at the neighborhood level.

BACKGROUND

The Price of “Justice”: The System of Monetary Sanctions

The manifest functions of the system of monetary sanctions are to hold people accountable for criminal behaviors, recoup legal-related costs, and punish (Harris 2016; Kirk, Fernandes, and Friedman 2020). However, emergent research highlights how jurisdictional LFO policies negatively affect people’s ability to secure safe and affordable housing, educational access, stable employment, and livable wages (Cadigan and Kirk 2020; Pattillo and Kirk 2020; Shannon et al. 2020). The intransigence of these inequalities is explained, in part, by the spatial overlap of poverty and LFOs. LFOs include fines, fees, costs, and restitution sentenced to people by the legal system as a result of both minor legal contact, such as traffic citations, and serious offenses, such as felony convictions (Kohler-Hausmann 2018; Natapoff 2018). While most LFOs are imposed at the time of citation or sentencing, individuals can also find themselves “on the hook” for costs that accrue prior to adjudication or determination of verdict, and costs connected with sentences and subsequent legal processing (Harris 2016; Harris, Smith, and Obara 2019). As with police contact, arrest, incarceration, and supervision, communities with higher rates of poverty and BIPOC residents (Barnes et al. 2015; Gaston 2018; Phelps 2017; Roberts 2004) are disproportionately affected by LFOs. Black and Latino defendants receive harsher financial penalties at sentencing than do White defendants (Harris, Evans, and Beckett 2011; Ruback 2004), and there is reason to believe post-sentencing LFO outcomes differ by race and ethnicity due to pervasive racial and economic inequalities.

Although recent scholarship has enumerated many individual-level consequences of LFOs, we know surprisingly little about their structural consequences. Our analyses focus on average cash amounts sentenced per tract (or neighborhood) resident, which we refer to as the neighborhood’s “LFO burden.” We use this term to emphasize the way LFO debt, although sentenced to individuals, puts economic pressure on the neighborhood as a whole. Harris (2016) argues the system of monetary sanctions is “two-tiered”: Financially stable individuals are able to pay LFOs quickly, whereas these sanctions constitute a permanent punishment for the poor that creates and perpetuates inequality. Given higher-than-average rates of poverty in many BIPOC communities (Lopez and Cohn 2011; Rankin and Quane 2000), it fits with our knowledge regarding racialized social systems that these groups are far more likely to experience post-sentencing hardships—both financial and social—than are White Americans. Therefore, we examine the extent to which LFOs create and perpetuate structural inequalities across neighborhoods in Washington State. Specifically, we argue LFOs exacerbate existing neighborhood-level racial and economic inequalities and perpetuate poverty across neighborhoods.

The spatial distribution of LFOs is entwined with the realities of economic disadvantage, racial segregation, and systemic racism within the criminal legal system. Individual characteristics like income, race, and ethnicity are associated with place of residence (Massey and Denton 1988; Pattillo-McCoy 1999; Sampson 2012), legal contact (Alexander 2010), and duration of LFO debt (Harris, Evans, and Beckett 2010; Harris et al. 2011). Therefore, we may reasonably expect LFOs and their consequences are disproportionately concentrated in poor, BIPOC neighborhoods. Scholarship on spatially concentrated disadvantage and social disorganization finds that per-capita rates of debt, crime, and legal entanglement are unequally distributed among neighborhoods (Clear et al. 2003; Kubrin et al. 2011) and often serve as barriers to success among community residents (W. J. Wilson 1996; W. J. Wilson and Aponte 1985). These barriers are particularly disruptive in majority non-White neighborhoods, where residents also grapple with systemic racism and xenophobia (Krivo, Peterson, and Kuhl 2009; Sampson 2012). Such conditions—barriers to success and the exacerbation oftheir effects in BIPOC communities—are (re)produced by racialized social systems (Bonilla-Silva 1997; Golash-Boza 2016; Ray 2019), implicating racism in both present realities of unequal contact with the legal system and the historical processes which produced the present (Seamster 2015). That a state intervention is implicated in the reproduction of racialized poverty is alarming, but not altogether surprising given the repeated finding that America’s criminal legal system favors White and affluent people in its dispensation of leniencies (Pettit and Western 2004; Sutherland 1949). Taken together, existing literature strongly suggests monetary sanction policies could be both a cause and consequence of racial inequalities at the individual and structural level.

The Consequences of Legal Debt

We suggest high per-capita rates of outstanding LFO debt, as opposed to other types of financial debt, are uniquely injurious because these financial penalties are involuntarily acquired and do not lead to desired services or material gain (Pattillo and Kirk 2020). While there is evidence that debt from other sources, such as student loans, mortgages, or leases on cars, negatively affect neighborhood and individual well-being (Duca and Rosenthal 1993; Krishnan and Wang 2018), such debt also assists people in gaining and transmitting wealth (Seamster and Charron-Chénier 2017). In contrast, LFO debt exacerbates existing economic problems such as difficulty affording childcare, housing, and food, and generates a new slate of issues, including increased costs from interest, collection fees, and continual court supervision (Cadigan and Kirk 2020; Friedman and Pattillo 2019). Furthermore, LFO debt can remove people from community labor pools by preventing their criminal records from being expunged (Pager 2003), and can result in driver’s license suspension (Kohler-Hausmann 2018), extended supervision (Shannon et al. 2020), disruptive court reporting requirements (Cadigan and Kirk 2020), and incarceration (Fernandes et al. 2019; Martin et al. 2018).

All of the above listed outcomes for people who carry LFO debt are in addition to the consequences of both criminal legal contact and debt in general. These obstacles to reform and upward mobility are further complicated when the person carrying LFO debt is also saddled with a criminal and/or carceral record (Miller 2021). Criminal convictions can prevent people from accessing loans for housing, vehicles, or entrepreneurial ventures (Duca and Rosenthal 1993; Krishnan and Wang 2018). This exclusion from legitimate credit markets has generated and amplified existing and emerging forms of structural disadvantage in poor and BIPOC spaces (Kubrin et al. 2011; Massey and Denton 1993; Oliver and Shapiro 1995). We propose the system of monetary sanctions is one such mechanism of exclusion. Given LFOs are situated at the intersection of debt, crime, and legal contact, it stands to reason the sentencing and payment of LFOs follows similar spatial and associative, particularly regarding how they map on to poverty and racial and ethnic compositions.

Disorganized by Design: Social Disorganization and CRTs

One enduring explanation for the persistence of neighborhood-level concentrated disadvantage is social disorganization. Social disorganization is defined as an absence of community effort or resources to adequately cope with undesirable conditions (Kornhauser 1978). Scholars who use this framework focus on why certain communities are locked into dynamics that reproduce disadvantage at individual and structural levels. Social disorganization has always been concerned with how and why the racial and ethnic composition of neighborhoods maps onto crime, poverty, and other forms of disadvantage and its originators were careful to specify that “ … one must beware of attaching causal significance to race or nativity” (Shaw and McKay 1942:155). However, many early theorists nonetheless pathologized high crime and poverty rates in BIPOC communities: attributing them to social and cultural deficits (Moynihan 1965; R. A. Wilson 1971) and spotlighting what they viewed as a lack of effort toward community improvement (Lander 1954). Although many social disorganization theorists now acknowledge legacies of systemic racism and xenophobia as precursors to patterns of disadvantage (Massey and Denton 1993; Sampson 2012; W. J. Wilson and Aponte 1985), the theory itself falls short of fully accounting for deliberate racism within the very institutions recommended as mechanisms of upward mobility and does not extend to the inner workings and motivations of state policy and intervention. CRT, however, suggests the spatial distribution of community resources and the ensuing “disorganization” in resource-poor communities are by design. Subsequently, while we build on research and findings borne from social disorganization theory, we turn to CRT to ground our interpretation of the findings. We use this theory to holistically examine whether state intervention in the form of LFOs contributes to a system of spatial oppression that perpetuates racial and economic inequalities (see McKittrick 2006).

CRT aims to centralize race, racism, and inter-sectionality with other forms of subordination; center experiential knowledge and transdisciplinary perspectives; and to challenge dominant public and academic ideologies while committing to social justice (Solorzano 1998; Solorzano and Delgado Bernal 2001). Scholars informed by CRT seek to better understand how racism consistently generates differences in group and individual experiences. In regard to systems of punishment, CRT suggests the criminal legal system is implicitly racialized and designed to propagate White people’s values, norms, and privileges through the unequal distribution of punishment, social control, and resources (Davis 2001; Ray 2019), and stresses the importance of dissecting intersecting systems of oppression (Collins 2009; Crenshaw 1991). The system of monetary sanctions, we argue, is one such system of oppression.

The LFO burden carried disproportionately by people who are BIPOC is evidence of a fully functional legal system in that it punishes differently racialized citizens and expropriates economic resources from their communities to maintain the very systems responsible for their subjugation, surveillance, and control (see Davis 1981, 2013). These “race-class subjugated communities”² are the subject of intense social control, where “inferior political positions flow from both insufficient governmental attention and too much government oversight, interference, and predation” (Soss and Weaver 2017:2). Although it is hard to empirically illustrate how “colorblind” institutional interventions reproduce racialized inequality (Bonilla-Silva 2013), our analyses allow us to see how this punishment triggers and reproduces inequalities. We therefore frame the system of monetary sanctions as a racialized and classed institutional intervention, a hidden but omnipresent form of social control that exacerbates disadvantages experienced in BIPOC and impoverished communities, and an inevitable consequence of a racialized criminal legal system. In this way, the concentrated disadvantage long-highlighted by social disorganization theorists is not an accident, it is not a cultural byproduct, and it is not due to lack of institutional force. Rather, it is the inevitable consequence of a system explicitly designed to subjugate and surveil marginalized populations. The disorganization is by design.

To understand this contemporary iteration of social control, the system of monetary sanctions, we must consider the context of the U.S. criminal legal system, borne of a racialized history of social control and the marginalization of African Americans, beginning with the enslavement of people of African descent in the early 1600s (Muhammad 2010; Washington [1900] 1959). This early form of social control was supported through the creation of slave codes, which allowed patrols to stop and capture African Americans who had been enslaved and return them to enslavers. Following the ratification of the Thirteenth Amendment, which ended legalized slavery except in cases of “punishment for crime whereof the party shall have been duly convicted,” slave codes were adapted into racial or “Black codes.” Black codes’ criminalization of the daily activities of African Americans, and enforced limitations on entrance into labor contracts and property ownership ushered in America’s first iteration of mass conviction in the nineteenth century (Blackmon 2009). When Black codes faded from legal coda around the turn of the century, Jim Crow laws took their place in deliberately assuring African Americans’ exclusion from structural opportunities for advancement and disproportionate representation among those entangled with the criminal legal system (Alexander 2010). The structure and culture of this distinctively American system is borne of these historic realities, and the persistent and disproportionate social control of BIPOC—especially African Americans—is at the foundation of American institutions. To paraphrase Patricia Hill Collins (2009), the shape of domination has changed, but the raced and classed outcomes of this system of social control remain consistent across time and space.

The contemporary criminal legal system continues this legacy with the onset of a second wave of mass conviction and incarceration starting in the 1970s. The system of monetary sanctions expanded out of necessity when local jurisdictions were unable to shoulder financial burdens associated with the 1990s prison boom (Harris 2016). This expansion disproportionately affected BIPOC populations, ensuring entire segments of society are limited in their physical and financial freedoms (Pattillo and Kirk 2021). LFOs effectively lock people and their communities in a cycle of prolonged legal entanglement, poverty, and disadvantage, aiding in the development of race-class subjugated communities. In this context, LFOs represent not just a pathway toward the reproduction of neighborhood inequalities, but a crucial part of the contemporary racialized social system, and a continuation of racially targeted social control.

CURRENT STUDY

Our analysis builds on Harris et al.’s (2010) model (Figure 1) of how LFOs perpetuate poverty, race and ethnic disparity, and class inequality, by analyzing the link between long-term legal debt imposed at conviction and community instability and poverty.

Figure 1.

Impact of penal expansion on poverty and inequality.

Source. Adapted from Figure 3.0 of Harris, Evans, and Beckett (2010).

Harris et al. (2010) theorize legal debt has negative consequences for individuals’ housing, education, and employment and suggest “monetary sanctions create additional mechanisms by which criminal conviction contributes to the reproduction of poverty and inequality” (p. 1789). However, they do not directly show how LFOs affect neighborhood-level poverty. Our analyses examine how LFO burden is related to neighborhood poverty and extend the Harris et al. (2010) model to include structural consequences of LFOs. We examine this sentencing practice not only as part of the criminal legal system, but as part and parcel of a larger American structure that perpetuates poverty and racial inequality. After presenting our findings, we discuss the relationship between neighborhood debt burden and racial and economic inequality, and review the potential mechanisms connecting LFOs to neighborhood poverty.

We use insight from social disorganization theory to discuss our expectations regarding the spatial distribution of LFOs, and use CRT to examine their role in exacerbating racial and economic inequalities. Our analyses interrogate the relationship between penal debt and its spatial distribution, and shed light on the economic, racial, and ethnic characteristics of LFO-burdened neighborhoods. Importantly, we analyze the relationship between LFO burden and neighborhood poverty over time, and argue the system of monetary sanctions is itself a driver of economic, racial, and ethnic inequalities at the neighborhood level.

We first describe the spatial distribution of LFO burden in Washington State, and the characteristics of neighborhoods with high LFO burden. Prior literature on concentrated disadvantage suggests “things go together” (Sampson 2012), in that many spatially concentrated disadvantages—such as high rates of debt per capita (Ratcliffe et al. 2014)—are interrelated and self-perpetuating. Furthermore, an abundance of criminological literature on neighborhoods demonstrates legal contact is concentrated in poor, non-White neighborhoods (Clear 2009; Peterson, Krivo, and Harris 2000). Therefore, it is likely that the distribution of LFOs closely aligns with other forms of neighborhood disadvantage, and is not distributed equally across geographic areas or neighborhood demographics. We focus our analysis on average cash amounts sentenced per tract resident, referred to as the tract or neighborhood’s “LFO burden.”

We then examine LFO burden’s role in reproducing between-neighborhood poverty and inequality. Our framing suggests the economic conditions within race-class subjugated neighborhoods, coupled with the high concentrations of LFO burden, will decline over time, and that this decline will be especially pronounced in neighborhoods with large populations of BIPOC residents. Residents in these neighborhoods, in addition to experiencing LFO burden, must navigate racism within legal systems, labor markets, and housing markets (Ray 2019). As such, we expect LFOs burden associated with future shares of residents in poverty and the impact of LFOs on poverty will differ according to the racial and ethnic composition of neighborhoods. In particular, we anticipate predominately BIPOC neighborhoods’ economic prospects are constrained more severely by LFO burden than are predominately White neighborhoods’ economic prospects.

DATA AND ANALYTIC APPROACH

We link data from two sources for these analyses. Case-level data on monetary sanction sentencing come from the Washington State Administrative Office of the Court (AOC). It includes information about LFOs charged between 2008 and 2014 in Washington State, as well as addresses collected by charging law enforcement officers and/or city prosecutors and basic demographic information about justice-involved people. We measure LFO burden as the cash sum of all “amounts charged” to individuals living in a particular tract in a particular year, divided by the American Community Survey (ACS) estimate of the population of that tract in that year. Notably, this measure does not capture how long it takes residents to pay their LFOs in full. Because the distribution of LFO burden is skewed, we take the log of the LFO burden in our analyses. In addition to LFO burden, we use AOC data to calculate trace-level counts of felony charges and a rates of homicides per 100,000 residents. We then combined geocoded AOC data with ACS five-year tract-level estimates, matching at the ACS mid-year, per the U.S. Census Bureau’s recommendations.³ We use ACS estimates of tracts’ racial and ethnic composition, shares of residents in poverty, shares of housing built after 2010, and shares of residents who are college-educated, have long commutes, or have recently moved into the tract.

In total, AOC data contain 5.6 million cases from 2008 to 2014, 3.2 million of which include an address. We excluded cases with addresses outside of Washington State, cases with no address listed, and cases with addresses that could not be traced to a particular tract. In total, 2,581,175 cases were geocoded and assigned to 1,444 tracts across a seven-year period. Accounting for lags, left us with 8,664 observations. We conducted robustness checks to ensure the LFOs without addresses would not substantially influence our results and discuss the implications for these missing cases in more detail in the discussion.

Analytic Approach

We describe the racial and socioeconomic distribution of LFO burden by regressing annual per-capita LFO charges in each tract on the percent poverty and a neighborhood typology to reflect the tract’s racial composition. Following Crowder, Pais, and South (2012), we label tracts where White people are a “supermajority” (i.e., 80 percent or more of the population) predominantly White. For neighborhoods where White people are the majority, but not the supermajority, we label the tract with the word “White” followed by a word designating the second largest group. If no non-White group share is greater than 10 percent, we label the tract “White-Mixed.” Tracts where no groups’ share exceeds 40 percent are labeled “Mixed,” and tracts with a White share less than 50 percent are “Majority non-White.” The resulting eight neighborhood types are shown in Table 1.

Table 1.

Descriptive Statistics by Neighborhood Type.

Neighborhoods (Tract-Year Observations)	Variables Mean over All Tract-years (Standard Deviation)
	Legal Financial Obligation Burden	Share in poverty	Share College Educated	Share with Long commute	Share Housing Built after 2010	Share New Residents	Homicides per 100,000	Felony convictions
All neighborhoods (8,662)	41.10 (42.64)	13.51 (9.91)	0.14 (0.08)	0.25 (0.09)	0.07 (0.09)	0.17 (0.09)	2.64 (11.59)	0.01 (0.15)
Predominantly White (3,420)	34.16 (40.93)	10.52 (7.31)	0.15 (0.07)	0.25 (0.09)	0.05 (0.07)	0.14 (0.07)	2.20 (11.9)	0.01 (0.17)
White-Latino neighborhood (1,935)	52.64 (34.27)	16.61 (9.28)	0.09 (0.05)	0.22 (0.09)	0.08 (0.10)	0.19 (0.09)	2.61 (9.37)	0.02 (0.14)
Latino-White neighborhood (242)	84.52 (42.13)	29.27 (9.64)	0.03 (0.02)	0.16 (0.05)	0.07 (0.07)	0.16 (0.08)	5.09 (12.30)	0.06 (0.30)
White-Mixed neighborhood (1,000)	43.42 (34.94)	13.93 (10.13)	0.13 (0.07)	0.26 (0.09)	0.07 (0.09)	0.19 (0.10)	3.33 (13.65)	0.01 (0.09)
White-Asian neighborhood (1,414)	26.36 (29.86)	10.16 (10.20)	0.21 (0.07)	0.30 (0.08)	0.09 (0.11)	0.19 (0.1)	1.64 (7.95)	0.02 (0.15)
Majority non-White neighborhood (139)	62.38 (65.20)	24.19 (11.11)	0.11 (0.08)	0.2 (0.09)	0.09 (0.11)	0.15 (0.08)	6.55 (26.29)	0.02 (0.19)
White-Black neighborhood (249)	49.73 (43.07)	17.15 (9.59)	0.13 (0.08)	0.26 (0.09)	0.09 (0.1)	0.25 (0.12)	5.42 (15.57)	0.00 (0.06)
Mixed neighborhood (263)	57.41 (96.70)	22.44 (9.71)	0.10 (0.05)	0.28 (0.06)	0.09 (0.12)	0.20 (0.07)	4.26 (11.33)	0.01 (0.09)

We use a neighborhood typology, rather than other measures of neighborhood racial composition for theoretical and methodological reasons. Traditional interpretations of difference across neighborhoods with differently raced populations—including some social disorganization approaches—tend to misattribute differences to individuals living in a particular place. Per CRT, however, these differences may actually be driven by both the tract’s exposure to systemic racism and its spatial racialization, as people judge spaces depending on how they perceive inhabitants of the space (Bonam, Taylor, and Yantis 2017; Bonds and Inwood 2016). Therefore, theoretically, neighborhoods are racialized. We operationalize this spatialized racialization, and its attendant exposure to systemic racism, using our typology. Methodologically, the neighborhood typology allows for easier interpretation of results as an explicit comparison of LFO burden across neighborhoods.

Our data are spatially and temporally structured, so we account for that structure in our modeling. Following Blangiardo et al. (2013), we include Besag York Mollié (BYM2) spatial errors, and include one-year temporal lags for our tract-level dependent variables. We estimate quantities of interest using Integrated Nested Laplace Approximation, or INLA (Bivand, Pebesma, and Gómez-Rubio 2008). We also include a number of tract-level covariates to represent other associative paths connecting LFOs, poverty, and racial composition, including shares of tract residents who completed a college degree as a measure of education; shares of residents who spend more than 20 minutes commuting as driving long distances increases the odds of receiving traffic-related LFOs; the share of housing units built after 2010 as an indication of recent neighborhood development; the number of homicide cases per 100,000 residents as a measure of violent crime; and the number of felony convictions per tract to somewhat disentangle the effects of LFO burden from the mark of a criminal record (Pager 2003). Our first analysis describes associations between the percent of residents living under the poverty line in a tract and the LFO-burden residents in that tract experience. The model is shown in equation (1) where S_i is the spatial BYM2 error, i is the yearly fixed-effect, and X is our matrix of independent variables.

$$\mathit{log}\left({\mathit{\text{LFOburden}}}_{it}\right)\sim \phi \mathit{log}\left({\mathit{\text{LFOburden}}}_{it-1}\right)+i+\beta X_{it}+S_i+{\epsilon }_{it}.$$ (1)

We test our expectations regarding the relationships between LFO burden, poverty, and inequality by extending the model used in our descriptive work to examine the influence of LFO burden on tract percent poverty over time. This model takes poverty as its outcome and LFO burden as the focal independent variable. We also add an interaction between LFO burden and neighborhood type, to see if the effect of LFO burden on tract poverty differs depending on the neighborhood racial composition. This model is shown in equation (2) where S_i is the spatial BYM2 error and Y_i is the yearly fixed-effect, and X is our matrix of independent variables.

$${\mathit{\text{poverty}}}_{it}\sim \phi {\mathit{\text{poverty}}}_{i{t}_{-1}}+i+\beta X_{it}+\mathit{log}\left({\mathit{\text{LFO}}}_t\right)\times {\mathit{\text{NHtype}}}_t+S_i+{\epsilon }_{it}.$$ (2)

Notably, our key substantive findings are robust to various ways of accounting for spatial and temporal effects, including using state-level AR1 temporal errors and spatially lagged dependent variables, or using fixed or random effects for each tract. While we believe models including spatial error structure potentially overcontrol for factors tightly bound to LFOs (such as policing practices), we nonetheless present and discuss results from models both with- and without this structure.

RESULTS

The Spatial, Racial, and Socioeconomic Distribution of LFO Debt

Regressing LFO burden on the share of neighborhood residents in poverty and neighborhoods’ racial typology shows this burden is concentrated in poorer, less-White neighborhoods across Washington state. Figure 2 shows tracts with high LFO burden and larger shares of residents in poverty often overlap in the Seattle metropolitan area, for example.

Figure 2.

Spatial association of poverty and LFO burden, Seattle area.

Note. LFO = legal financial obligation.

The lightest gray areas with no lines have low poverty and low LFO burden, and tend to be concentrated in west Seattle, a wealthy, gentrifying area, and Seattle’s eastside, home of Microsoft and other corporate offices. The south-central band of areas that are both dark gray and crosshatched—and therefore have both high LFO burden and high poverty—aligns with Seattle’s legacy of racialized housing distribution (Taylor 1994): high-poverty, high-LFO areas also tend to have larger BIPOC populations. Regression results shown in Table 2 provide additional context to these maps: Model I, which includes neighborhood typology but not spatio-temporal effects, shows less White tracts tend to have higher LFO burden, even after accounting for poverty. However, only two neighborhood racialization dummies, those for White-Latino neighborhoods and White-Mixed neighborhoods, remained significant after adding spatial errors and lagged log LFO burden. Moreover, the coefficients are much smaller, suggesting the relationship between poverty and LFO burden are associated with the unequal application of criminal legal sanctions across space and time.

Table 2.

Logged LFO Burden Regressed on Share of Residents in Poverty.

	Model I	Model II
Intercept	3.86a (0.03)	2.90a (0.07)
Focal variables
log (LFO burden)t−1	—	0.11a (0.01)
Share in poverty	0.015a (0.001)	0.01a (0.001)
White-Latino neighborhood	0.22a (0.02)	0.04a (0.02)
Latino-White neighborhood	0.25a (0.04)	0.03 (0.06)
White-Mixed neighborhood	0.16a (0.02)	0.04a (0.02)
White-Asian neighborhood	0.02 (0.02)	0.03 (0.02)
Majority non-White neighborhood	0.10a (0.05)	0.12 (0.06)
White-Black neighborhood	0.19a (0.04)	0.06 (0.04)
Mixed neighborhood	0.16a (0.04)	0.04 (0.04)
Control variables
Share college educated	−4.31a (0.10)	−1.27a (0.19)
Share with long commute	0.07 (0.08)	0.60a (0.14)
Share housing built after 2010	−0.22a (0.07)	−0.25a (0.06)
Share new residents	−0.64a (0.08)	0.05 (0.1)
Homicides per 100,000	0.01a (0.001)	0.004a (0.00)
Felony convictions	0.09a (0.04)	0.05 (0.03)
Deviance information criterion	14,259.2102	8,162.529045

Note. Standard errors in parentheses. LFO = legal financial obligation.

^aDenotes the 95 percent credible interval does not include zero.

Model II shows neighborhoods with larger BIPOC populations have higher LFO burdens, even when accounting for spatio-temporal errors. This model estimates a 10 percentage-point increase in the poverty rate is associated with an almost five dollar expected increase in per-capita LFO burden at the median.

LFO Burden across Time

Our empirical expectations regarding LFO burden and the reproduction of economic and racial inequality are strongly supported. We find LFO burden is associated with higher levels of poverty one year on. Table 3 displays Models III to VI, which regress the share of residents in poverty on logged LFO burden.

Table 3.

Share of Residents in Poverty Regressed on Logged LFO Burden.

	Model III	Model IV	Model V	Model VI
Intercept	6.12a (0.61)	11.44a (0.76)	2.03a (0.32)	2.86a (0.38)
Focal variables
Share in povertyt−1	—	—	0.86a (0.006)	0.853a (0.01)
log(LFO burden)	2.19a (0.13)	0.66a (0.19)	0.33a (0.06)	0.09 (0.09)
White-Latino neighborhood	0.70a (0.20)	−16.82a (1.14)	0.23a (0.10)	−2.32a (0.51)
Latino-White neighborhood	10.31a (0.47)	−7.36 (3.96)	1.41a (0.26)	1.44 (1.82)
White-Mixed neighborhood	0.44 (0.27)	−6.55a (1.25)	0.09 (0.11)	−1.57a (0.55)
White-Asian neighborhood	1.33a (0.23)	4.04a (0.96)	0.15 (0.12)	−0.11 (0.44)
Majority non-White neighborhood	10.45a (0.58)	−2.29 (2.57)	1.22a (0.28)	−2.05 (1.20)
White-Black neighborhood	0.83 (0.45)	−18.26a (1.92)	−0.30 (0.21)	−2.19a (0.86)
Mixed neighborhood	7.35a (0.44)a	2.50 (2.78)	0.30 (0.23)	1.87 (1.24)
Control variables
Share college educated	−30.31a (1.32)	−33.70a (1.30)	−7.99a (0.84)	−8.20a (0.83)
Share with long commute	−19.98a (0.92)	−19.45a (0.90)	−3.89a (0.60)	−3.95a (0.59)
Share housing built after 2010	−3.00a (0.88)	−1.66 (0.86)	2.80a (0.38)	2.87a (0.38)
Share new residents	46.24a (0.90)	46.55a (0.88)	3.46a (0.48)	3.82a (0.48)
Homicides per 100,000	0.02a (0.01)	0.02a (0.01)	0.004 (0.003)	0.004 (0.003)
Felony convictions	−0.26 (0.46)	−0.14 (0.45)	−0.24 (0.19)	−0.23 (0.19)
LFO burden × Neighborhood (interactions)
log(LFO burden) × White-Latino	—	4.74a (0.30)	—	0.69a (0.14)
log(LFO burden) × Latino-White	—	4.35a (0.91)	—	0.06 (0.42)
log(LFO burden) × White-Mixed	—	2.04a (0.35)	—	0.48a (0.15)
log(LFO burden) × White-Asian	—	−1.00a (0.30)	—	0.06 (0.14)
log(LFO burden) × Majority non-White	—	3.52a (0.67)	—	0.91a (0.31)
log(LFO burden) × White-Black	—	5.37a (0.52)	—	0.53a (0.23)
log(LFO burden) × Mixed	—	1.40 (0.72)	—	−0.38 (0.32)
Deviance information criterion	57,495.55015	57,090.738	41,992.17	41,965.33912

Note. Standard errors in parentheses. LFO = legal financial obligation.

^aDenotes the 95 percent credible interval does not include zero.

We find a positive association between LFO burden and poverty that maintains its significance with reduced magnitude in models including spatio-temporal error structure (as seen in Models V and VI), additional covariates, and the neighborhood interaction. Notably, while the coefficient for LFO burden in Model VI is not significant, the interaction improves model fit, and three interaction coefficients are significant. In Model V, the short-term effect of LFOs on poverty, net of the tract’s history of poverty, spatial autocorrelation, and other covariates, seems small: a one standard deviation increase in logged LFO burden, which would be an increase of $35.00 per person at the median, would only be associated with a short-term increase in poverty of a third of a percentage point. In a tract with 4,500 people (roughly our sample average), that places 15 additional people below the poverty line. However, the long-term effect is much greater: estimating using β/(1 – φ) implies a 2.29 percent increase in the percent of people below poverty, or just over 100 people in a 4,500-person tract.⁴

Moreover, adding interactions shows the association between LFO burden and future poverty differs depending on neighborhood racialization. Figure 3 shows these differences using counterfactuals to examine what happens to hypothetical neighborhoods of each neighborhood type across the study period. Counterfactuals were estimated by using our model to predict posterior distributions for all of our observed tract-years using their original values of LFO burden, holding LFO burden to its median for all tracts as a baseline, and increasing LFO burden by one standard deviation. We drew 1,000 samples from the joint posterior distribution and subtracted each of our counterfactual conditions: the baseline and the one standard deviation increase, from the posterior draws for the observed LFO-burden values. We plot the median differences and 95 percent credible intervals for each neighborhood type in Figure 3.

Figure 3.

Estimated change in the share of residents in poverty from one standard deviation increase in LFO burden.

Note. LFO = legal financial obligation.

Solid black lines and dark confidence intervals show Model VI’s expectations based on a one standard deviation increase while dashed gray lines and light gray intervals show Model V’s expectations based on LFO burden held to their median value of $32.00 per capita. This plot shows a disturbing reality: While growth in poverty in predominantly White neighborhoods is little affected by LFO burden, other neighborhood types have vastly different outcomes depending on their exposure to LFO burden over time. White-Latino, White-Mixed, Majority non-White, and White-Black tracts have predicted increases in poverty approaching two percentage points or greater, and slopes indicating that the true long-term effect of that increase would be greater. The decreasing poverty for the baseline of median LFO burden also reflects the fact that LFO burden in these tracts tends to be high. Results for predominantly White, Latino-White, and White-Asian neighborhoods suggest LFO burden is not an important driver of poverty in those spaces. Results for mixed neighborhoods reflect the unique status of neighborhoods with relatively high poverty, but comparably low LFO burden.

DISCUSSION

Our analyses suggest LFOs sentenced per capita are associated with increased future shares of residents in poverty, and the strength of this association varies according to the racial and ethnic composition of neighborhoods. We find neighborhoods with higher levels of LFO burden in a particular year tend to have higher poverty over time, and this association is stronger in less-White neighborhoods. These results suggest, as we argue, that the system of monetary sanctions increases within-neighborhood poverty and exacerbates existing racial inequalities across neighborhoods. In using CRT to build on social disorganization literatures, we highlight the fraught relationship certain spaces have with government institutions.

While our results are striking, it is possible they are not due to LFOs in particular, but criminal legal contact in general. We believe we capture most of that indirect effect, but we do not dispute that some of our associations link to other consequences of criminal legal involvement, such as incarceration and coercive mobility, stigmatization, or legal cynicism. However, we counter LFOs represent a direct and pernicious example of this relationship at the neighborhood level, and outline four mechanisms specific to LFOs that demonstrate their importance in the reproduction of inequality: (1) the direct extraction of economic resources from neighborhoods, (2) surveillance, (3) courtesy stigma, and (4) neighborhood residents’ civic engagement.

First, LFOs represent a direct extraction of economic resources from neighborhoods, which weakens or subverts institutional bonds among neighborhood residents. As with other types of debt, when people are unable to pay LFOs in full, unpaid amounts often accrue interest, collection fees, and annual fees at high rates—placing enormous economic strain on people who are poor. LFOs “tether” individuals to the criminal legal system with almost no benefit to individuals or neighborhoods (Harris 2016). LFO debt cannot be transferred, discharged with bankruptcy, or leveraged to acquire goods or services (Atkinson 2017). Essentially, the criminal legal system forces people to purchase their freedom, and people who are poor face coercive financialization, weakening their status and citizenship rights until debts are paid (Pattillo and Kirk 2021). Penal debt prevents people from investing funds into local economies and can force them to dip into already limited neighborhood resources to survive. This can result in the broad and gradual deterioration of resident- and community-resource ties, as residents may increasingly find social networks and institutions are unable to meet their needs despite their stated purpose. In this way, penal debt is diffused throughout the neighborhood. Furthermore, while some collections, such as restitution, may be recirculated within neighborhoods with higher penal debt burden, the overall operating costs of the system of monetary sanctions means it barely pays for itself (Harris 2016; Martin and Fowle 2020). In sum, the system of monetary sanctions extracts resources from neighborhoods, straining social network ties, weakening institutional bonds, and, ultimately, preventing neighborhoods from building structural resources to support upward mobility (see Kawachi et al. 1999). Our results indicate this extraction, and its consequences are more severe in neighborhoods with larger populations of BIPOC residents.

The second mechanism linking LFOs to neighborhood disadvantage is the continued and increased surveillance and social control of LFO-burdened neighborhoods. People entangled in the system of monetary sanctions are subjected to increased surveillance (Brayne 2017; A. Goffman 2014). Continual court supervision, probation, address reporting requirements, warrant issuances, and regular court appearances are all examples of surveillance and social control triggered by nonpayment (Cadigan and Kirk 2020; Katzenstein and Waller 2015; Natapoff 2018). In general, residents of BIPOC and poor neighborhoods are more likely than are residents of White, affluent neighborhoods to experience surveillance (Gaston 2018). In addition, the use of LFOs and related surveillance has expanded substantially over the past 15 years, with one primary function to “keep debtors in an almost constant state of surveillance” (Harris 2016:76). In many cases, this surveillance extends well beyond the completion of other formal sentences (Kohler-Hausmann 2013). Subsequently, residents in heavily surveilled communities avoid contact with institutions that could conceivably contribute to their surveillance, even in scenarios where those institutions could otherwise alleviate financial or medical problems (Brayne 2014; A. Goffman 2014). This system avoidance may explain why the economic prospects of neighborhoods with heavy LFO burdens deteriorate over time, even where institutional or structural assistance is available. Furthermore, local government entities’ reliance on LFOs as a source of revenue (Edwards 2020; O’Neill, Smith, and Kennedy 2022) has increased the scope of police responsibilities to include debt imposition and collection (Brett 2020). Not only does this increase police surveillance in LFO-burdened neighborhoods, it increases cynicism regarding the efficacy and fairness of the system of monetary sanctions (Pattillo and Kirk 2020; Shannon et al. 2020). Heavy surveillance of neighborhood residents can contribute to within-neighborhood social isolation, widening the chasm between neighborhood residents and resources needed for economic mobility.

The third mechanism linking LFOs to neighborhood poverty is courtesy stigma (E. Goffman 1963), which illustrates how LFO burden weakens neighborhood ties and increases social isolation within neighborhoods (see W. J. Wilson 1987, 1996). Stigma, shame, and exclusion resulting from criminal legal contact is not limited to entangled individuals. Rather, it extends to those suspected of criminal legal involvement regardless of actual criminal record (Pager 2003). Harris et al. (2011), for example, find people in counties with large shares of Black and Latino residents receive harsher financial penalties at sentencing for racially and ethnically coded offenses, regardless of defendant race, than do residents in counties with higher White populations. This finding, they argue, indicates the increased magnitude of LFO penalties that flow from racialization affect not only Black and Latino populations, but anyone spatially associated with these populations. Similarly, we argue stigmas associated with debt and legal contact are associated with neighborhoods’ LFO burden in general, rather than just individual debtors in particular. Not only does stigma negatively influence interactional decisions of institutional actors within the criminal legal system, it also negatively affects the dispositions of other institutional actors such as hiring managers and welfare disbursement officers (Pager 2003; Seefeldt 2016). In this way, LFO debt is positioned as both a cause and consequence of neighborhood disadvantage in general, and race-linked courtesy stigma in particular. Our results regarding average per-capita debt amounts in White versus non-White neighborhoods are supportive of the role of LFOs in the perpetuation of between-neighborhood inequality.

Finally, our fourth mechanism addresses the strain legal debt places on the social and citizenship status of residents through social disengagement and disenfranchisement (Harris 2016; Sebastian, Lang, and Short 2020). Social disorganization theory suggests neighborhoods benefit from widespread, shared values, norms, and goals, but the power of these shared ideations is weakened considerably by systems of felon disenfranchisement and voter suppression—both of which disproportionately affect the ability of neighborhoods with high concentrations of people unable to serve as fully engaged citizens (Manza and Uggen 2008). As a result, the political capital in these neighborhoods is so limited as to prevent or dissuade residents from petitioning for structural interventions from government institutions that may serve to improve their well-being. LFOs are a mechanism of disenfranchisement in already politically weakened neighborhoods. Furthermore, the criminal legal system inhibits poverty-preventing structural change or the generation of “spatial advantage” (Sharkey 2013:115). Our results build on these findings by illustrating the strength of associations between LFO burden and future poverty.

As with any study, our methodological approach has its limitations. While our modeling strategy is designed to address possible endogeneity between past LFO burden, poverty, and unobserved spatial patterns to try to isolate the effect of LFOs, it is possible some of our observed associations are a reflection of the powerful intersection of neighborhood processes that (re)produce disadvantage. Further work should address these issues by collecting additional data on possible endogenous causal paths and by using explicitly causal methods such as difference in difference or regression discontinuity. A further limitation of this study is the large number of LFO cases without geocodable address data provided by the AOC. Our robustness checks indicate that although cases without an address are broadly similar to those with addresses in that they cover many of the same types of charges, dates, counties, and court types, there are a few notable differences. Geocodable cases included in analyses tend to have higher total LFOs ordered than do un-geocodable cases (i.e., missing cases). Although both geocoded and missing cases include a substantial number of “paid” cases, missing cases were more likely to have been paid. This is likely because people who quickly paid their LFOs were less likely to offer or be asked to supply their addresses. Furthermore, it is possible that many of those LFOs were charged to people with unstable or no housing, in which case, our models may be underestimating the link between poverty and LFO burden.

In addition to methodological concerns, we are interested in better understanding the mechanisms through which the system of monetary sanctions reproduces the inequalities highlighted in this research, and for whom. The racial and ethnic composition of Washington State is not representative of the United States as a whole, and there are no majority-Black census tracts represented in these analyses. This prevents us from drawing conclusions specific to Black Americans and majority-Black neighborhoods, despite ample evidence that these individuals and locations are uniquely disadvantaged (Clear 2009; W. J. Wilson 2010). Furthermore, although we suggest weakened institutional bonds and social isolation are major drivers of the inequalities highlighted in this research, we are unable to test these mechanisms directly.

Our analysis sets up many research questions: How might the spatial distribution of LFOs differ across and within cities, counties, and states outside of Washington State? Which policies related to monetary sanctions are driving these spatial inequalities? How do LFOs influence residential mobility patterns among neighborhoods? We propose the next step in research on LFOs and the perpetuation of racial inequality is to expand it across states to see how state-level statutes differ in their effects on LFO burden across cities and counties, comparing large metropolitan cities to explore whether patterns of neighborhood LFO burden remain consistent with our findings. Finally, we hope to further expand our knowledge of LFO burden past urban contexts and into rural spaces, where jail incarceration rates continue to stagnate or increase even as prison incarceration rates decrease nationwide (Kang-Brown and Subramaniam 2017).

CONCLUSION

Our finding that LFO burden (re)produces poverty over time is alarming. The system of monetary sanctions appears to contribute to the reproduction of spatial oppression (McKittrick 2006) by reproducing racial differences in the accrual of household wealth and community resources (see Oliver and Shapiro 1995). The system of monetary sanctions is a “perfect” stratifying mechanism employed by a contemporary American institution. It is a perfect mechanism for reproducing inequality in that it disproportionately over-polices and arrests people in impoverished communities and those with higher rates of BIPOC residents and marking them with citations, convictions, and LFOs. This punishment scheme is a perfect system of stratification in that already economically disadvantaged and racially marginalized neighborhoods carry penal debt that does not go away until paid in full. As a result, the impoverished literally pay more than do the affluent for the same crimes. This burden is not just borne by people who have experienced criminal legal contact, but by their communities through punishments designed by state and local policy makers. As such, the system of monetary sanctions labels, stigmatizes, financially burdens, and imposes further legal consequences on poor and BIPOC populations, allowing perpetual state surveillance, intervention, and control over race-class subjugated communities. Our research lays the groundwork for future work to examine community-level mechanisms linking LFO burden to the exacerbation of poverty by highlighting how the system of monetary sanctions is part of a contemporary iteration of a racialized social system based in state policies of social control that disproportionately target African Americans and other BIPOC.

This pernicious perfection makes this system a clear target for policy change. The spatial distribution of penal debt allows us to see how and where the state chooses to employ policy decisions that reproduce poverty and inequality. Policymakers have begun to enact changes to decrease the disparate effect LFOs have on people who are poor, their families, and communities. Jurisdictions across the country have eliminated court’s authority to impose fines and court costs against juveniles, while others have ended drivers’ license suspensions for unpaid court fines, fees, and parking tickets including. Furthermore, state policy makers have begun to revise monetary sanction statutes to include provisions mandating ability to pay hearings, legally defining indigence, and excluding certain categories of people from fiscal penalties.

Despite these incremental changes, our findings call into question the extent to which the system of monetary sanctions can ever be just, given the racist origins of the criminal legal system, and the present-day state of racial and economic inequality. If the system of monetary sanctions is functioning as intended, inasmuch as it disproportionately targets BIPOC and contributes to the cycles of disadvantage within BIPOC communities, is it worth saving? Scholars have long argued for curtailing the reach of the criminal legal system, if not abolishing it entirely (see Davis 2013; Davis and Rodriguez 2003). In this current moment of state-sponsored violence directed at BIPOC in general and Black people in particular, and the health crisis of COVID-19, it seems prudent for local and state officials to curb this extraction of wealth from already marginalized and precariously situated communities. If policy makers can garner the will to change, they can directly improve life circumstances for individuals, families, and communities.

Using a critical race lens, we acknowledge past and existing racial disparities in neighborhood formation and in contact and processing within the criminal legal system. We focus our analysis on this system and the institutional punishment of LFOs, highlighting where social control, the criminal legal system, power, race, and poverty intersect. Empirically, we demonstrate how LFOs isolate, repress, surveil, and further facilitate the marginalization of race-class subjugated communities. Both the race and class composition of the neighborhoods studied, coupled with criminal legal intervention, highlights the cumulative disadvantages borne by race-class subjugated communities and the “distinctive positions in relations of power and oppression” (Soss and Weaver 2017:4). The study of the system of monetary sanctions highlights how the state, via the criminal legal system, perpetuates inequities and further marginalizes BIPOC and impoverished neighborhoods—not just on the individual level but also en masse.

Biographies

AUTHOR BIOGRAPHIES

Kate K. O’Neill is a PhD candidate in sociology at the University of Washington. She holds an MSc in the sociology of crime, control, and globalization from the London School of Economics and Political Science, and MA from the University of Washington in Sociology. Broadly, her research focuses on crime, criminal justice, pathways to crime, and the reproduction of inequalities. Her recent publication in Feminist Criminology on how gender differences in empathic ability perpetuate gender gaps in juvenile delinquency won the Graduate Student Paper Award from the American Sociological Association Section on the Sociology of Emotions.

Ian Kennedy is a graduate student in sociology at the University of Washington. His research focuses on computational social science, and intersections of race, digital platforms, and text analysis. His recent publication in Social Forces explores how racialized discourses in Craigslist rental ads perpetuate housing segregation in Seattle. His work aims to contribute to understandings of how contemporary racism works, in both visible and less visible ways. He is committed to producing useful work beyond scholarly publications, working with groups like the Northwest Justice Project to identify illegal Craigslist ads or with the Election Integrity Partnership to monitor misinformation during the 2020 election.

Alexes Harris is the presidential term professor and professor of sociology at the University of Washington. Her research fundamentally centers on issues of inequality, poverty, and race in United States’s criminal legal systems. Her book, A Pound of Flesh: Monetary Sanctions as a Punishment for the Poor details the ways in which sentenced fines and fees often put an undue burden on disadvantaged populations and place them under even greater supervision of the criminal justice system. She has been appointed to serve on several federal advisory boards, has been inducted into the Washington State Academy of Sciences (2017), and is currently the chair of the Washington State Advisory Committee to the United States Commission on Civil Rights. In 2018, she was acknowledged for her teaching with the University of Washington’s highest teaching honor, the Distinguished Teaching Award.