Examining the Sources of Racial Bias in Potentially Capital Cases

Abstract

While prior research has uncovered racial disparities in the administration of death sentences, little attention has been devoted to earlier stages in the capital punishment processes. To understand the locus of racial bias within death penalty institutions, this study examines the entry of homicide cases into Los Angeles County’s criminal justice system during a 5-year period. This two-part analysis seeks to answer the following research questions: (1) Does victim/defendant race influence homicide clearance and death penalty charging decisions? and (2) if so, does the likelihood of clearance mediate the effect of victim race on death penalty charges? Logistic regressions indicate that cases involving Latino victims are less likely to be cleared. Moreover, cases with Black and Latino victims are less likely to be prosecuted with a death penalty–eligible charge. Racial disparities accumulate across these stages, with clearance patterns influencing subsequent death penalty charging decisions. Results underscore the cumulative nature of racial within criminal justice institutions. By linking police and prosecution outcomes, these findings also highlight the interrelationship between criminal justice agencies.

Keywords

prosecutorial discretion cumulative racial disadvantage racial disparity capital punishment death penalty homicide clearance

In Furman v. Georgia (1972), the Supreme Court found the death penalty to be unconstitutional due to its arbitrary and capricious application. In response, states implemented new death penalty laws intended to guide jury decision-making and define death penalty–eligible offenses (Carter, Kreitzberg, & Howe, 2012). Despite these efforts to limit juror and prosecutorial discretion, racial disparities persist today. Cases involving White victims, especially those with minority defendants, are more likely to be prosecuted capitally and/or receive a death sentence (for a review, see Baldus & Woodworth, 2003; Baldus, Woodworth, & Pulaski, 1990; U.S. General Accountability Office [USGAO], 1990). However, given that much of the literature focuses on the final stages of capital prosecution, it is unclear whether racial disparities result from the failure of “modern” death penalty laws to guide juror and prosecutorial discretion or from earlier biases in the criminal justice system (Baldus, Woodworth, Zuckerman, Weiner, & Grosso, 2009; Sorensen & Wallace, 1999).

Studies focusing on death-sentencing patterns offer invaluable insights about these outcomes, but a thorough understanding of racial bias within death penalty institutions requires looking at the chain of events leading up to these final decision-making points (Pierce & Radelet, 2005; Radelet & Pierce, 1985). Arrest and charging decisions represent two critical junctures in the death penalty process as they determine which cases are eligible for capital prosecution (Pierce & Radelet, 2005; Radelet & Pierce, 2009). Racial disparities at these stages could shrink the pool of death-eligible¹ cases involving Black and Latino² victims by making cases involving White victims appear more serious than they actually are, thereby obscuring racial disparities at later stages in the process (Pierce & Radelet, 2005; Radelet & Pierce, 1985, 2009). Despite the relevance of these presentencing stages, relatively few studies have investigated death penalty charging practices, especially the relationship between arrest and charging outcomes. This gap in the literature is particularly problematic because policy reforms aimed at ameliorating racial disparities within death penalty institutions must first identify the source of these patterns.

The current study helps to fill this gap in the literature by examining homicide arrest and charging patterns in Los Angeles (LA) County, California. This two-part analysis seeks to answer the following research questions. Do racial characteristics influence arrest decisions and prosecutors’ filing of death penalty–eligible charges? Does victim race have an indirect effect on prosecutors’ death penalty charging practices, operating through homicide arrests? Findings indicate that cases involving minority victims are less likely to result in an arrest, which in turn, is negatively associated with the odds of a death penalty–eligible charge. As a result, defendants accused of killing minority victims are less likely to be charged with a death penalty–eligible charge. Moreover, cases involving a White defendant and White victim are more likely to involve a death penalty–eligible charge than other victim-by-defendant racial combinations. These results suggest that prior research may underestimate the extent to which race influences capital punishment decision-making by ignoring the racialization of death eligibility and the relationship between arrests and charging patterns.

Literature Review

Racial bias in the American death penalty system has most commonly been identified at the postcharging stages. Homicides involving White victims are more likely to be prosecuted capitally and/or receive a death sentence than those with minority victims (Baldus & Woodworth, 2003; Baldus et al., 1990; USGAO, 1990). And while Black defendants accused of killing White victims are punished more harshly than other defendant by victim racial configurations, victim race primarily accounts for these findings—defendants who kill Whites receive harsher punishment (Baldus & Woodworth, 2003; Baldus et al., 1990; USGAO, 1990). In contrast, comparatively little attention has been devoted to earlier stages of the death penalty process and particularly the point at which the decision to bring a capital charge is weighed. Most studies examine prosecutors’ decision to seek the death penalty and/or jurors’ tendencies to impose a death sentence among a sample of death-eligible defendants, taking for granted the entry of cases into the capital punishment system (Kaplan, Ganschow, Angioli, & Tabin, 2009; Radelet & Pierce, 1985, 2009). In its 1990 report, the USGAO criticized studies for focusing on the final stages of capital prosecution, largely ignoring earlier biases in the system. Similarly, while sentencing scholars have assessed presentencing disparities in a variety of noncapital settings (Kutateladze, Lynn, & Liang, 2012), the issue has received little attention from death penalty researchers.

Racial Bias, Prosecutorial Discretion, and Cumulative Disadvantage

Focal concerns theory is frequently used to help explain racial disparities within the criminal justice system (for a review, see Ulmer, 2012). Steffensmeier, Ulmer, and Kramer (1998, p. 767) note that officials attempt to gauge defendants’ culpability when making decisions, but because legal actors often have incomplete or partial information they tend to invoke racial stereotypes as a kind of “perceptual shorthand”. Stereotypes linking racial/ethnic minorities to criminality may lead prosecutors and judges to use race as a proxy for less easily observable focal concerns (e.g., culpability, future dangerousness, etc.). Studies of homicide charging decisions generally provide support for the focal concerns perspective insofar as race shapes case outcomes. Prior research finds that defendants accused of killing White victims are more likely to be charged with first-degree murder than those accused of killing minority victims (Bowers, 1983; Bowers & Pierce, 1980; Lee, 2007; Radelet, 1981; Weiss, Berk, & Lee, 1996; Weiss, Berk, Li, & Farrell-Ross, 1999). In addition, cases involving minority offenders and White victims are more likely to include an accompanying felony charge, and this “upgrading” process increases the risk of a capital conviction (Radelet & Pierce, 1985).

Focal concerns theory describes the formation of racial bias, while the cumulative disadvantage perspective sheds light on the accumulation of racial bias across multiple stages. Zatz (1987, 2000) argues that racial disparities take two forms: subtle biases which operate through case processing mechanisms such as bail or plea agreements (i.e., indirect effects) and overt bias arising from more intentional forms of discrimination (i.e., direct effects). Early decisions in the criminal justice system (e.g., arrest) also influence later ones (e.g., charging/sentencing severity) by shaping the pool of potential defendants at each stage (Kutateladze, Andiloro, Johnson, & Spohn, 2014; Kutateladze et al., 2012; Spohn, 2000; Zatz, 1987, 2000). In line with the cumulative disadvantage perspective, research suggests that racial disparities at the charging stage may have implications for death-sentencing outcomes. The proportion of death-eligible homicides with White victims and/or Black defendants increases as cases advance through the death penalty system (Baldus et al., 1990; Bowers, 1983; Bowers & Pierce, 1980; Paternoster, Brame, Bacon, & Ditchfield, 2004; Sorensen & Wallace, 1999). Multivariate analyses suggest that these patterns cannot be explained away by racial differences in offense severity (Sorensen & Wallace, 1999). Research on nonhomicidal crimes indicates that racial disparities compound as cases move through the court system, underscoring the accumulation of race effects across multiple decision-making points (Kutateladze et al., 2014; Schlesinger, 2008; Stolzenberg, D’Alessio, & Eitle, 2013; Sutton, 2013).

On the other end of the criminal justice system, racial disparities can arise before a case even enters into the courts due to differential police enforcement. In the death penalty context, policing patterns can influence punishment outcomes in subtle, but profound ways, by determining the pool of prosecutable cases and availability/quality of evidence presented to the District Attorney’s (DA’s) office (Bright, 1994; Pierce & Radelet, 2005; Songer & Unah, 2006). For example, lower arrest rates among homicides with minority victims could decrease the proportion of prosecutable cases involving Black and Latino victims, simultaneously increasing the proportion of cases with White victims (Pierce & Radelet, 2005). Racial disparities at the arrest stage may also lead to the underdevelopment of evidence in minority victim cases, making the cases of defendants who kill White victims appear more serious than they actually are (Bright, 1994; Pierce & Radelet, 2005; Songer & Unah, 2006).

Indeed, prior research indicates that a disproportionate number of unsolved homicides involve minority victims. While cases with characteristics linked to greater evidence are more likely to be solved (e.g., firearm weapon, residential crime scene), homicides involving minority victims are less likely to result in an arrest even after controlling for such factors (for a review, see Riedel, 2008). Homicide arrest rates also vary at the neighborhood level. Homicides occurring in predominately Black and Latino communities are less likely to result in an arrest even after controlling for victim and offense characteristics (Litwin, 2004; Litwin & Xu, 2007; Petersen, forthcoming; Regoeczi & Jarvis, 2013). Low homicide clearance rates in minority areas could stem from the fact that residents may be reluctant to speak with the police because of legal cynicism or strained police–community relations more generally (Puckett & Lundman, 2003). Taken together, these patterns of racial disparity produce a “separate and unequal” system of homicide policing (LaFree, Baumer, & O’Brien, 2010, p. 94), which may have implications for the criminal justice processing of homicide cases once they enter into the court system.

Organizational Dynamics and the Police–Prosecution Nexus

The police–prosecution relationship can also be conceptualized as a loosely coupled system. Weick (1982, p. 380) defines a loosely coupled system as one in which “elements are responsive, but retain evidence of separateness and identity … [the] elements affect each other … indirectly (rather than directly) ….” In other words, loosely coupled systems are responsive to each other but still maintain their independence and identity (Hagan, Hewitt, & Alwin, 1979; Orton & Weick, 1990; Weick, 1976). This implies that loosely coupled systems share few predictors in common and, when they do, their effects differ (Glassman, 1973). In contrast, components in a tightly coupled system depend on each other to function, leaving little room for autonomy and interpretation (Alarid, Sims, & Ruiz, 2011). Based on these definitions, the police–prosecution relationship may represent a loosely coupled system wherein each agency has autonomy to make their own decisions based on information they deem relevant, while still cooperating to advance the commonly shared goal of an eventual criminal conviction.

Several scholars have characterized the criminal justice system as loosely coupled, with each subsystem having relative autonomy to create their own brand of justice (Hagan, 1973, 1989; Hagan et al., 1979; Jackson, Webster, & Hagan, 1982; Reiss, 1973; Sampson & Laub, 1993). In their seminal study, Hagan, Hewitt, and Alwin (1979) find that while probation reports and sentencing decisions are tightly linked on paper, in actuality the autonomy afforded to judges allows them to largely disregard the recommendations of probation officers. Subsequent work has replicated these general findings, demonstrating that although criminal justice agencies share common goals and functions, they are often autonomous and thus rely on different decision-making criterion (Alarid et al., 2011; Bishop, Leiber, & Johnson, 2010; Jackson et al., 1982; Leiber & Jamieson, 1995). The decoupling of criminal justice organizations has also been linked to racial disparities. When criminal justice agencies are loosely coupled, the diversity of organizational goals and interests allows extralegal factors to more easily influence case outcomes by augmenting the discretionary power of legal actors (Bishop et al., 2010; Leiber & Jamieson, 1995).

The Present Research: Conceptualizing Entry into the Death Penalty System

Postcharging studies implicitly assume that prosecutors file death penalty charges when appropriate. Yet, given the organizational pressures placed on securing convictions (Albonetti, 1986, 1987; Landes, 1971; Rasmusen, Raghav, & Ramseyer, 2009), prosecutors may socially construct homicide cases in ways that align with these professional goals. As Radelet and Pierce (1985) note:

… Sentencing studies that take the prosecutor’s case descriptions and the formal charges as objective and unbiased reflections of the seriousness of a crime can lead to the underestimation of race effects on sentencing whenever race has affected earlier processing decisions. To understand the full effects of race (and other variables), the presentencing and precharging decisions that affect the prosecutor’s construction of a case must be examined. (p. 616, emphasis added)

In other words, if prosecutors file death penalty charges in a racially discriminatory manner, postcharging estimates might not fully capture the amount of racial bias within the death penalty system (Baldus, Woodworth, Zuckerman, et al., 2009; Sorensen & Wallace, 1999). Thus, understanding the extent to which postcharging studies may underestimate race effects first requires gauging the influence of race on arrest and charging decisions.

However, little is currently known about death penalty charging practices. Much of the research on the topic has been methodological in nature, focusing on the development of Bayesian or counterfactual techniques for analyzing death penalty data, rather than understanding the role of racial factors (e.g., Berk, Boger, & Weiss, 1993; Weiss et al., 1996, 1999). And even when researchers do explicitly focus on victim/defendant race effects, they often include samples of a few hundred defendants, raising concerns about statistical power (USGAO, 1990). In contrast, this study analyzes the charging outcomes of over 5,000 defendants and includes a wider range of covariates, thereby providing more a comprehensive picture of death penalty charging practices. Moreover, this study includes a large number of Latino victims and defendants, a group which is often neglected in death penalty and sentencing research (Baldus & Woodworth, 2003; Baldus, Woodworth, & Weiner, 2009; Kutateladze et al., 2014; Spohn, 2000; Zatz, 2000).

The present research also examines the police–prosecution nexus. Despite a robust literature on the racialization of homicide arrests (Riedel, 2008), no study has directly examined the relationship between homicide arrest and prosecution outcomes. It has been argued that the police play a critical role in shaping death penalty outcomes by determining which cases enter into the system and the evidence brought to prosecutors, but this argument has not been empirically tested (Bright, 1994; Pierce & Radelet, 2005; Songer & Unah, 2006). By analyzing the full universe of homicides during a 5-year period from the arrest to charging stages, this analysis sheds light on cumulative race effects and the organizational dynamics of the police–prosecutor relationship.

Theoretical expectations

While the present analysis is not solely focused on theory testing, the aforementioned theories can help to guide the interpretation of my findings. According to focal concerns theory, race serves as a proxy for less readily observable indicators such as “social status,” “innocence,” or “dangerousness” in the absence of detailed victim/defendant information (Steffensmeier et al., 1998). As such, focal concerns theory predicts that homicide cases involving minority victims are less likely to result in an arrest or the filing of a death penalty–eligible charge because of stereotypes regarding Black/Latino criminality. I hypothesize that these same racial stereotypes will lead officials to punish minority defendants more harshly than White defendants. It is also expected that death penalty charging rates will be higher in crimes involving White victims and minority offenders given prior research indicating that, in some contexts, cases with White victims and Black defendants receive harsher punishments than other victim–offender racial dyads (Baldus et al., 1990; Baldus & Woodworth, 2003; USGAO, 1990). According to cumulative disadvantage theory, victim race will have an indirect effect on death penalty charging practices, with racial disparities compounding across stages of the death penalty system. In other words, victim race will have an indirect effect on prosecutorial charging practices through its prior effect on homicide arrest patterns. From an organizational perspective, the magnitude and significance of covariates are expected to vary across police and prosecutorial outcomes if they represent a loosely coupled system (Glassman, 1973). On the other hand, similarities in arrest and charging outcomes would indicate that the system is tightly coupled.

Data and Methods

This two-part analysis focuses on willful homicides that occurred in LA County between 1990 and 1994.³ Part 1 examines the clearance outcomes of all willful homicide incidents in the sample; accidental, vehicular, and justifiable homicides were excluded. Part 2 investigates prosecutors’ death penalty charging decisions among defendants in the sample of cleared homicides (murder and voluntary manslaughter cases), using the estimated likelihood of clearance from Part 1 (i.e., hazard rate) as a predictor. Recognizing that racial disparities may compound as cases move through the criminal justice system, the hazard rate is employed to model this funneling process (Berk, 1983; Bushway, Johnson, & Slocum, 2007; Zatz, 1987).

Data from local and state criminal justice agencies were combined to construct a comprehensive data set. Information on homicide incidents came from local law enforcement agencies, California Department of Justice (CADOJ), California Vital Statistics, and Coroner’s Office. Data on homicide court cases originate from the LA County Superior Court and DA’s Office. Victim information (e.g., name, age and race) and incident characteristics (e.g., date and location) found in court files was used to merge defendants and victims across these various data sources. These data were analyzed because they include the full universe of homicides during the study period, enabling me to examine cumulative race effects by tracking homicides through multiple stages of the criminal justice system (see the online appendix for the full list of variables and their coding schemes).

Research Setting: LA County’s Sociolegal Landscape During the Early 1990s

This study focuses on the years 1990 through 1994 for pragmatic and substantive reasons. First, data on both homicide victims and defendants were only made available to me for the years 1990–1994, thus necessitating a focus on this period. Second, the high homicide rates of the early 1990s serve to increase the sample size. In LA County and elsewhere, homicide rates were at historical high levels in the early 1990s (Blumstein & Wallman, 2006), allowing for a robust analysis of criminal justice processing of homicide cases by augmenting the sample size. For example, there were 1,768 homicides in LA County in 1990 versus 600 in 2012 (CADOJ, 2015). Moreover, not only does LA County have one of the largest prosecutorial agencies in the United States (LA County District Attorney’s Office, 2015), but it also accounts for the majority of death sentences and unsolved homicides in California (CADOJ, 2010; California Department of Corrections and Rehabilitation, 2015).

Entry into LA County’s death penalty system begins with an arrest. Once a homicide is “cleared by arrest,” it is eligible for prosecution by the DA’s office. In California, only first-degree murders that involve a statutorily defined aggravating circumstances enumerated in Penal Code (PC) §190.2 qualify for the death penalty (Kreitzberg, 2008). These aggravating circumstances, known as “special circumstances” in California, include factors such as multiple murder or felony murder that make a case more heinous. Almost 90% of first-degree murder cases factually qualify for the death penalty in California under one or more special circumstance (California Commission on the Fair Administration of Justice [CCFAJ], 2008; Shatz & Rivkind, 1997), rendering the state’s death-eligibility criterion even broader than the one deemed unconstitutional in Furman v. Georgia (Shatz & Rivkind, 1997). California’s exceedingly expansive death penalty statute affords prosecutors considerable charging power, increasing the potential for interjurisdictional and racial disparities (Petersen & Lynch, 2013). Despite the state’s broad death eligibility, it appears that a few special circumstances, especially felony murder and multiple murder, account for the majority of death-eligible homicides (Kreitzberg, 2008; Shatz, 2007).

Dependent Variables: Homicide Clearance and Death Penalty Charge

Dependent variable in Part 1

The first dependent variable focuses on one of the most important turning points in the investigation and prosecution of a homicide—whether the case is cleared by arrest or exceptionally cleared. Based on the CADOJ definition, and in line with prior research (Riedel, 2008), clearance status is measured dichotomously (1 = at least one suspect arrested or identified and 0 = no arrests made or suspects not identified). According to the CADOJ (2010, p. 50), a case is cleared by arrest when “at least one person is arrested, charged with the commission of an offense, and turned over to a court for prosecution.” The CADOJ notes that cases can also be “‘cleared exceptionally’ for crime reporting purposes when an investigation has definitely established the identity of an offender, enough information exists to support an arrest, and the exact location of an offender is known but, for some reason, law enforcement cannot take the offender into custody.”

Dependent variables in Part 2

The second set of dependent variables includes a binary and ordinal measure of special circumstance filings as defined under PC §190.2. For the binary measure, defendants charged with at least one of the special circumstances enumerated in PC §190.2 are coded as 1, otherwise the defendant is coded as 0 (1 = at least one special circumstance and 0 = no special circumstance). This binary measure taps into a critical point in California’s death penalty process—the decision to file charges that make a case death penalty eligible. For the ordinal variable, I measure the number of special circumstances filed (0 = no special circumstance, 1 = one special circumstance, and 2 = two or more special circumstances).⁴ Although only one special circumstance is required for death eligibility in California under PC §190.2, the filing of additional special circumstances is important because it is associated with a higher probability of capital prosecution (Petersen & Lynch, 2013; Yarvis, 2000). Prosecutors often seek to maximize conviction rates when making charging decisions (Albonetti, 1986, 1987; Landes, 1971; Rasmusen et al., 2009), and thus the filing of multiple special circumstances increase the likelihood of advancing to a capital trial (Petersen & Lynch, 2013; Yarvis, 2000). While much of the prior capital punishment research examines prosecutors’ decision to seek the death penalty, I analyze the filing of death-eligible charges, given the paucity of work on this topic and given the focus of this study on earlier decision-making points in the death penalty system.

Predictor Variables

Predictors of homicide arrests in Part 1

Victim racial groups were coded using a series of dummy variables: Latino, Black, and White (reference).⁵ Prior research suggests that victim demographics can play a major role in homicide case outcomes, and thus I control for a host of victim characteristics (Baldus & Woodworth, 2003; Baldus et al., 1990; Phillips, 2009a, 2009b). Victim gender, marital status, and citizenship were dichotomously coded, while victim age was measured in years. Victim social status was measured based on their education level, using a series of dummy variables. Like prior research on homicide arrests, suspect demographics are excluded given that this information is not known for unsolved homicides (Riedel, 2008).

Models also adjust for incident characteristics and contextual factors. Some homicides may be harder to solve than others due to the circumstances surrounding the incident and thus it is important to consider such factors (Riedel, 2008). For example, residential homicides usually contain greater evidence because the suspect is typically at the crime scene when police arrive (Riedel, 2008). As such, a series of categorical variables measuring crime characteristics were added, including the number of victims, crime scene location, precipitating circumstance, weapon, incident day, and victim–offender relationship. Dummy variables for years 1990–1994 were included to control for annual effects (Lee, 2005; Litwin, 2004; Litwin & Xu, 2007). In light of research finding jurisdictional variation in homicide arrest patterns, I control for the police agency investigating the case (Borg & Parker, 2001; LaFree et al., 2010). In particular, I include dummy variables comparing the two largest law enforcement agencies in LA County—the LA Police Department (LAPD) and LA Sheriff’s Department (LASD)—to all other police departments in the county (e.g., Long Beach Police, Torrance Police, etc.). Finally, racial/ethnic characteristics of the crime scene community were included, given prior research highlighting neighborhood disparities in terms of homicide arrest rates (Litwin, 2004; Litwin & Xu, 2007; Petersen, forthcoming; Regoeczi & Jarvis, 2013). Racial composition in the crime scene community was measured as the percentage of Black and Latino residents per census tract.

Predictors of death penalty charges in Part 2

Given the cumulative nature of homicide case processing, models predicting death penalty charges in Part 2 utilize many of the aforementioned variables from Part 1. In addition to the covariates listed above, Part 2 includes defendant demographics and characteristics associated with the court case. These factors are used in Part 2, but not Part 1, because defendant demographics and court case characteristics are by definition only available when the courts initiate a criminal case. Defendant race is coded using dummy variables: Latino, Black, and White (reference).⁶ Defendant gender is dichotomously coded, whereas defendant age is continuous. The number of prior felony convictions was logarithmically transformed to capture its potential diminishing effect.

In addition, Part 2 controls for court case characteristics. A dummy variable captures whether the case or homicide charges resulted in a dismissal since homicide cases that were eventually dismissed may be less likely to involve a death-eligible charge. Prosecutors may also be more likely to offer a charge/sentence reduction in codefendant cases in exchange for evidence implicating another defendant (CCFAJ, 2008), and thus I control for the presence of multiple defendants. I measure the number of criminal counts charged, which was logarithmically transformed. Felony murder (e.g., robbery murder, rape murder, etc.) and multiple murder are among the most commonly filed death-eligible charges in California and elsewhere (Acker & Lanier, 1993; Kreitzberg, 2008; Peterson & Bailey, 1991; Petersen & Lynch, 2013). Therefore, binary variables capture whether the case involved multiple victims and a contemporaneous felony. Mitigating and aggravating factors identified by Baldus et al. (1990, pp. 526–535) were dichotomously coded based on whether they were present in the case file. Given the rarity of some of these aggravators/mitigators, they were used to construct factor scores via principal components factor analysis rather than included as separate indicators. Aggravating factors include case characteristics that make the crime more heinous like torture or dismemberment, while mitigating factors such as mental impairment during the crime serve to diminish the defendant’s criminal culpability.

Analysis Strategy

A two-part modeling strategy was employed.⁷ Part 1 estimates the odds of clearance, which is then used as a covariate in Part 2 predicting the odds of a death penalty–eligible charge. This two-part approach has been commonly utilized in the sentencing literature and simulates many of the key features of a Heckman selection model, including the ability to control for selection bias and model entry into the sample (Demuth, 2003; Keil & Vito, 1990; Leiber & Fox, 2005; Leiber & Mack, 2003; Rodriguez, 2010; Steen, Engen, & Gainey, 2005; Steffensmeier & Demuth, 2001; Ulmer & Johnson, 2004). It is unlikely that victims/defendants within the same incident/case are statistically independent, and thus clustered standard errors (SEs) were employed at the incident level for Part 1 and case level for Part 2. Given the large number of coefficients, the interpretation of specific estimates from the various models is largely limited to victim/defendant race variables as those are most germane to the study’s focus.

Like most data sets, this one has some missing data. While missingness levels vary across variables, the average amount of missingness is less than 3%. Moreover, key variables—victim and defendant race—are missing in less than 1% of the cases. Chained multiple imputation was used in STATA version 13 via the “mi impute chained” command. Ten imputed data sets were constructed as this amount is sufficient to introduce random error into the process (Schafer, 1999; Wang & Mears, 2010a, 2010b).

Part 1 includes a logistic regression model predicting the likelihood of arrest among the full universe of homicides. After running the model, the predicted probability of arrest was saved and used to calculate the hazard rate by dividing the probability density function over the cumulative distribution function.⁸ In Part 2 among the sample of cleared homicides, the hazard rate is used as a predictor of death penalty charges, allowing me to model this funneling process. Models predicting homicide arrests (Part 1) and charging decisions (Part 2) include many of the same covariates, with a few notable exceptions. Several key variables were purposefully excluded from Part 2 to fulfill the exclusion restriction requirement (Bushway et al., 2007).⁹ Based on the relevant literatures, exclusion restrictions for this study include incident day and crime scene location (Baldus & Woodworth, 2009; Riedel, 2008).¹⁰

Results

Summary Statistics

Summary measures reveal a dramatic funneling process at the front end, with only 46% of homicides resulting in an arrest (see Table 1). Among cleared cases, special circumstances were charged in 20% of cases (see Table 1). Blacks (35%) and Latinos (50%) comprise the majority of all victims, but only 70% of victims in death-eligible cases. In solved cases, nearly all victims (86%) and offenders (94%) are young men, the majority of whom range from 25 to 28 years old. Offense severity and prior criminal history vary based on death eligibility: Compared to the total pool of defendants, cases with a special circumstance are more serious in terms of aggravating factors, number of victims, contemporaneous felony, and number of counts.

Table 1.

Summary Statistics for LA County Homicides.

	All Homicide Incidents		Cleared Cases		Death-Eligible Cases
Variables	M	SD	M	SD	M	SD
Dependent variables
Homicide cleared	0.46	0.50
Special circumstances (yes/no)			0.20	0.40
Number of special circumstances (ordinal)			0.26	0.56	1.31	0.46
Victim demographics (Parts 1 and 2)
Victim race: Black	0.35	0.48	0.36	0.48	0.32	0.47
Victim race: Latino	0.50	0.50	0.48	0.50	0.38	0.49
Victim race: White	0.15	0.36	0.16	0.37	0.30	0.46
Victim gender: male	0.86	0.35	0.85	0.36	0.77	0.42
Victim age	28.77	13.57	28.62	13.27	33.72	15.03
Victim edu: non–high school grade	0.32	0.47	0.34	0.47	0.39	0.49
Victim edu: college grade	0.12	0.33	0.14	0.34	0.22	0.41
Victim married/widowed	0.23	0.42	0.24	0.43	0.38	0.48
Victim legal resident	0.74	0.44	0.74	0.44	0.77	0.42
Social contextual factors (Parts 1 and 2)
Incident year 1991	0.20	0.40	0.22	0.41	0.23	0.42
Incident year 1992	0.21	0.41	0.22	0.41	0.23	0.42
Incident year 1993	0.21	0.41	0.19	0.40	0.18	0.38
Incident year 1994	0.18	0.39	0.16	0.37	0.16	0.37
LAPD case	0.54	0.50	0.63	0.48	0.61	0.49
LASD case	0.26	0.44	0.21	0.41	0.21	0.41
% Latino in neighborhood	0.51	0.26
% Black in neighborhood	0.24	0.26
Crime characteristics (Part 1)
Multiple victims	0.12	0.32
Firearm offense	0.78	0.41
Location: public place	0.66	0.47
Location: residence	0.25	0.43
Relationship: stranger	0.36	0.48
Contemporaneous felony	0.23	0.42
Weekend homicide	0.38	0.48
Case characteristics (Part 2)
Factor score: mitigators			0	1	−0.1	0.96
Factor score: aggravators			0.00	1.00	0.43	1.83
Multiple victims			0.09	0.28	0.32	0.47
Contemporaneous felony			0.49	0.50	0.89	0.31
Multiple defendants			0.40	0.49	0.58	0.49
Log (# of counts)			0.75	0.25	0.87	0.45
Case/charges dismissed			0.20	0.40	0.17	0.38
Firearm weapon			0.75	0.43	0.76	0.43
Relationship: stranger			0.19	0.40	0.33	0.47
Relationship: unclear			0.30	0.46	0.31	0.46
Defendant demographics (Part 2)
Defendant race: Black			0.41	0.49	0.48	0.50
Defendant race: Latino			0.46	0.50	0.33	0.47
Defendant race: White			0.13	0.33	0.19	0.39
Defendant gender: male			0.94	0.23	0.95	0.23
Log (# of prior felony convictions)			0.25	0.47	0.27	0.51
Defendant age			25.59	8.66	24.54	7.10

Note. Listwise deleted sample. M = mean; SD = standard deviation; LAPD = LA Police Department; LASD = LA Sheriff Department. Reference groups: White victim, female victim, high school, single marital status, nonlegal resident victim, single victim case, nonfirearm weapon, other locations, nonstranger relationship, nonfelony, weekday incident, 1990 incident year, other city police agencies, single victim case, nonfelony, single defendant case, case/charges not dismissed, nonfirearm weapon, nonstranger relationship, White defendant race, and female defendant.

Regression Estimates

Part 1: Predicting the likelihood of homicide clearance

Compared to cases involving White victims, the odds of clearance are 26% lower for Latino victims (see Table 2). For Blacks, however, victim race is not significant (see Table 2). According to Model 1, a 1-unit change in the percentage of Black residents corresponds to a 35% reduction in the odds of clearance, while the odds of clearance decrease by 21% as the Latino population increases by 1%. Like prior research, cases with incident characteristics typically associated with greater evidence are more likely to be cleared (Riedel, 2008). For example, cases with a stranger victim–offender relationship and incidents occurring in public places or residences are more likely to be solved. In contrast, the contemporaneous felony or multiple victim variables were not significant. Social contextual factors influence policing behaviors as well. Homicides occurring after 1992 and those handled by smaller city police departments are less likely to be cleared.

Table 2.

Logistic Regression Predicting the Likelihood of Clearance for LA County Homicides.

Predictor Variables	Model 1 Odds Ratio (SE)
Victim characteristics
Victim race: Black^a	0.91 (0.08)
Victim race: Latino^a	0.79*** (0.07)
Victim edu: non–high school grade^b	1.05 (0.06)
Victim edu: college grade^b	1.10 (0.08)
Victim age squared	1.00* (0.00)
Victim age	1.00 (0.01)
Victim married/widowed^c	1.10 (0.07)
Victim gender: male^d	0.88* (0.06)
Victim legal resident^e	0.99 (0.05)
Crime characteristics
Multiple victims^f	1.00 (0.09)
Firearm offense^g	0.66*** (0.04)
Location: public place^h	1.34*** (0.15)
Location: residence^h	1.73*** (0.20)
Victim–suspect relationship: strangerⁱ	0.69*** (0.04)
Circumstance: felony^j	0.93 (0.06)
Weekend homicide^k	1.01 (0.05)
Social contextual factors
Incident year 1991^l	1.03 (0.07)
Incident year 1992^l	0.91 (0.07)
Incident year 1993^l	0.87** (0.06)
Incident year 1994^l	0.78*** (0.06)
LAPD case^m	2.10*** (0.13)
LASD case^m	0.99 (0.07)
% Latino in neighborhood	0.79** (0.09)
% Black in neighborhood	0.65*** (0.08)

Note. N = 9,137. SE = standard error; LAPD = LA Police Department; LASD = LA Sheriff Department. Exponentiated coefficients with SEs in parentheses. SEs are clustered by homicide incident using STATA’s “vce(cluster)” command.

Reference groups: ^aWhite victim. ^bHigh school. ^cSingle marital status. ^dFemale victim. ^eNonlegal resident victim. ^fSingle victim case. ^gNonfirearm weapon. ^h“Other” locations. ⁱNonstranger relationship. ^jNonfelony. ^kWeekday incident. ^l1990 incident year. ^mOther city police agencies.

*p < .1. **p < .05. ***p < .01.

Part 2: Predicting the likelihood of death penalty charges

Regression estimates displayed in Table 3 speak to the impact of race on prosecutorial decision-making. Compared to cases with White victims, the odds of a death penalty charge are 62–65% lower for cases with Black victims and 47–49% lower for cases with Latino victims (Models 2 and 3). In addition, victim race indirectly shapes charging practices, as evidenced by the positive effect of the hazard rate (odds ratio = 4.80, p < .01 in Model 2; odds ratio = 4.73, p < .01 in Model 3). Cases involving minority victims are less likely to be cleared (Model 1), which, in turn, decreases the odds of a death penalty charge (Models 2 and 3). While victim race is implicated in clearance and charging outcomes, there are no differences among Black and White defendants in terms of death penalty charges. However, Latino defendants are 36–39% less likely to be charged with a death-eligible crime than White defendants.

Table 3.

Logistic and Ordered-Logistic Regressions Predicting the Likelihood of Death Penalty–Eligible Charges in LA County Homicide Cases.

	Logistic, Model 2	Ordered-Logistic, Model 3
Predictor Variables	Odds Ratio (SE)	Odds Ratio (SE)
Hazard rate: odds of clearance from Model 1	4.80** (3.65)	4.73** (3.40)
Victim demographics
Victim race: Black^a	0.35*** (0.08)	0.38*** (0.07)
Victim race: Latino^a	0.53*** (0.10)	0.51*** (0.10)
Victim gender: male^b	0.62*** (0.10)	0.57*** (0.09)
Victim age	1.10*** (0.02)	1.09*** (0.02)
Victim age square	1.00*** (0.00)	1.00*** (0.00)
Victim edu: non–high school grade^c	1.04 (0.15)	1.01 (0.14)
Victim edu: college grade^c	1.29 (0.32)	1.18 (0.27)
Victim married/widowed^d	1.26 (0.19)	1.24 (0.17)
Victim legal resident^e	0.84 (0.13)	0.88 (0.12)
Defendant demographics
Defendant race: Black^f	1.04 (0.24)	1.01 (0.22)
Defendant race: Latino^f	0.61** (0.13)	0.64** (0.14)
Defendant gender: male^g	0.98 (0.23)	0.95 (0.21)
Log (# of prior felony convictions)	1.14 (0.13)	1.19 (0.14)
Defendant age	1.03 (0.04)	1.04 (0.04)
Defendant age squared	1.00 (0.00)	1.00 (0.00)
Case characteristics
Factor score: mitigators	0.81*** (0.05)	0.82*** (0.05)
Factor score: aggravators	1.66*** (0.13)	1.49*** (0.10)
Multiple victims^h	13.05*** (3.97)	14.63*** (3.61)
Contemporaneous felonyⁱ	9.34*** (1.45)	8.40*** (1.29)
Multiple defendants^j	1.33** (0.18)	1.40*** (0.17)
Log (# of counts)	1.28 (0.30)	1.22 (0.25)
Case/charges dismissed^k	0.64** (0.13)	0.62** (0.12)
Firearm weapon^l	1.03 (0.17)	1.03 (0.16)
Relationship: stranger^m	2.24*** (0.34)	2.15*** (0.32)
Relationship: unclear^m	1.16 (0.19)	1.12 (0.17)
Social contextual factors
Incident year 1991ⁿ	1.29 (0.23)	1.32 (0.22)
Incident year 1992ⁿ	1.26 (0.28)	1.33 (0.28)
Incident year 1993ⁿ	1.04 (0.19)	1.08 (0.19)
Incident year 1994ⁿ	1.23 (0.24)	1.30 (0.24)
LAPD case^o	1.13 (0.19)	1.09 (0.18)
LASD case^o	0.93 (0.17)	1.03 (0.19)

Note. N = 5,012. SE = standard error; LAPD = LA Police Department; LASD = LA Sheriff Department. Exponentiated coefficients with SEs in parentheses. SEs are clustered by case number using STATA’s “vce(cluster)” command.

Reference groups: ^aWhite victim. ^bFemale victim. ^cHigh school. ^dSingle marital status. ^eNonlegal resident victim. ^fWhite defendant race. ^gFemale defendant. ^hSingle victim case. ⁱNonfelony case. ^jSingle defendant case. ^kCase/charges not dismissed. ^lNonfirearm weapon. ^mNonstranger relationship. ⁿ1990 incident year. ^oOther city police agencies.

*p < .1. **p < .05. ***p < .01.

In addition, regression estimates in Table 3 uncover the significance of several nonracial predictors. Defendants accused of killing victims who are female or older are more likely to be charged with a death-eligible offense. In contrast to Model 1 predicting the likelihood of homicide clearance, stranger victim–offender relationship and multiple victims have a positive effect, but the presence of a firearm is not significant. Multivictim cases are 13–14 times more likely to be prosecuted with a death-eligible charge than single-victim cases and defendants charged with a contemporaneous felony are 8–9 times more likely to be charged with a special circumstance.

In light of research highlighting the interactive effects of victim and defendant race, Models 4 and 5 examine various victim-defendant racial interactions, with White-on-White crimes serving as the reference group (Baldus & Woodworth, 2003; Baldus et al., 1990; USGAO, 1990). In the interest of parsimony, Table 4 only lists these racial interaction terms since many of the other variables have similar effects (see the online appendix for full models). Compared to cases with White victims and White defendants, the odds of a death-eligible charge are 63–67% lower for Black victim/Latino defendant cases, 57–61% lower for Black victim/Black defendant cases, and 63–65% lower for Latino victim/Latino defendant cases (see Models 4 and 5 in Table 4). However, contrary to my expectations, cases with White victims and Black defendants are not statistically more likely to be prosecuted with a special circumstance.

Table 4.

Victim and Defendant Racial Interactions Predicting the Likelihood of Death Penalty–Eligible Charges in LA County Homicide Cases.

	Logistic, Model 4	Ordered-Logistic, Model 5
Racial Interaction Terms	Odds Ratio (SE)	Odds Ratio (SE)
Black victim and Latino defendant^a	0.37*** (0.13)	0.33*** (0.11)
Black victim and Black defendant^a	0.39*** (0.09)	0.43*** (0.10)
Black victim and White defendant^a	1.14 (0.51)	1.11 (0.45)
Latino victim and Latino defendant^a	0.35*** (0.08)	0.37*** (0.09)
Latino victim and Black defendant^a	0.75 (0.21)	0.70 (0.18)
Latino victim and White defendant^a	0.75 (0.33)	0.90 (0.44)
White victim and Latino defendant^a	0.76 (0.22)	0.96 (0.26)
White victim and Black defendant^a	1.54 (0.50)	1.26 (0.35)

Note. N = 5,012. SE = standard error. Exponentiated coefficients with SEs in parentheses. SEs are clustered by case number using STATA’s “vce(cluster)” command. Models 4-5 control for all of the same variables as Models 2-3, but these covariates are excluded for visual simplicity.

^aReference group for all interaction terms is White victim and White defendant.

*p < .1. **p < .05. ***p < .01.

Figure 1, which displays the adjusted probabilities for the various victim–defendant racial dyads from Model 4, speaks to these patterns. For cases with Black or Latino defendants, the predicted probability of a death-eligible charge is higher for White victims than it is for Black or Latino victims (see Figure 1). However, the predicted probability of a death-eligible charge is fairly high for White victims regardless of defendant race, which may help to explain the null findings for the dyads “White victim and Latino defendant” and “White victim and Black defendant.” In other words, Figure 1 shows that White victims have a high predicted probability of a death-eligible charge across the various defendant racial groups, and thus these patterns may reflect the fact that White victim homicides are generally treated more punitively.

Figure 1.

Predicted Probability of a Death-Eligible Charge by Victim and Defendant Race.

Discussion and Conclusion

Despite the implementation of modern death penalty laws, racial disparities persist in the post-Furman era. The breadth of California’s death eligibility affords prosecutors considerable latitude at the charging stage (Petersen & Lynch, 2013; Shatz, 2007), producing racially patterned death penalty charges. Consistent with prior research focusing on prosecutors’ decision to seek the death penalty, victim race plays a larger role in death penalty decision-making than defendant race (Baldus & Woodworth, 2003; Baldus et al., 1990; USGAO, 1990). Even after controlling for a host of case characteristics, defendants accused of killing White victims are more likely to be charged with a death-eligible offense than those accused of killing minority victims.

The focal concerns perspective can help shed light on these findings. Victim race effects are consistent with prior research highlighting the influence of officials’ stereotypes on criminal justice processes in southern California (Frohmann, 1991, 1997; Herbert, 1996). Although regression models cannot directly speak to these processes due to the lack of individual-level data on prosecutors’ stereotypes, victim race findings are generally consistent with the expectations of focal concerns theory regarding racial stereotypes (Steffensmeier et al., 1998). In contrast, defendant race alone has little influence on death penalty charging decisions (except for Latino defendants). Moreover, cases involving a White victim and Black defendant were are not statistically more likely to receive a death penalty–eligible charge than those with a White victim and defendant.¹¹

Defendant race may influence death penalty charging decisions in other ways. To the extent that police and prosecutors more aggressively investigate White victim homicides, that would mean there is greater evidence available in such cases. And because most homicides in LA County and elsewhere are intraracial, police and prosecutors might have more evidence implicating White defendants (Jacobs & Wood, 1999; Messner & South, 1992; Petersen, 2016; Petersen & Ward, 2015).¹² In other words, these findings may simply reflect the fact that most homicides are intraracial, and as such cases with a White victim generally progress further in the criminal justice system (for a similar discussion, see Spohn, 2000). The prevalence of intraracial homicides could also help to explain the curious finding that Latino defendants were treated more leniently than White defendants. That is, since cases with Latino victims are generally treated less punitively and Latino victims are more likely to have a Latino victim, this may account for the relative leniency of Latino defendants as compared to White defendants.

Precharging decisions matter too, underscoring the police–prosecution nexus. In line with prior research, incident characteristics shape homicide clearance outcomes, but they cannot explain away victim race effects (Riedel, 2008). Cases with Latino victims and those occurring in predominately minority neighborhoods are less likely to be cleared, which, in turn, decreases their likelihood of receiving a death penalty charge. In this regard, racially disparate death penalty charges are partly a product of lower clearance rates among minority victim cases and those crimes occurring in minority communities. This pattern underscores the cumulative nature of racial bias within the death penalty institutions, beginning with the crime scene investigation (Baldus, Woodworth, Zuckerman, et al., 2009; Radelet & Pierce, 1985, 2009). Given the close working relationship between detectives and prosecutors, the police may help to construct death eligibility by underdeveloping of cases with minority victims (Bright, 1994; Pierce & Radelet, 2005; Songer & Unah, 2006).

Results also speak to the (dis)connect between criminal justice institutions. The differential effect of several variables across clearance and charging decisions speaks to the decoupling of these criminal justice agencies. Crime characteristics associated with evidence (e.g., location and weapon) have a greater impact on clearance outcomes, while offense severity measures (e.g., mitigators, aggravators and number of victims) are more influential for charging decisions. And although victim racial characteristics influence both clearance and charging decisions, they play a bigger role at the charging stage. These divergent effects suggest that police and prosecutors respond to different organizational demands and thus may be regarded as a loosely coupled system (Glassman, 1973). According to organizational theory, this incongruity allows victim race and other extralegal factors to creep into the decision-making process by augmenting the discretion of criminal justice actors (Bishop et al., 2010; Leiber & Jamieson, 1995).

The present study extends the death penalty literature in important ways. While prior research has found racial differences in homicide clearance rates (Riedel, 2008), scholars have not examined the police–prosecution nexus. This study explicitly modeled the funneling of cases into the death penalty system and, in doing so, helps to bridge the literatures on homicide arrests and court processing by highlighting the influence of policing practices on prosecutorial charging decisions. Findings indicate that the pool of Latino victims winnows as homicides enter into the criminal justice system, potentially obscuring victim race effects at later stages (Radelet & Pierce, 1985, 2009). In this regard, existing sentencing studies may actually underestimate the amount of racial bias within death penalty institutions by paying insufficient attention to the racial homogenization of cases at earlier stages in the process. These insights suggest that multistage analyses offer a more holistic assessment of American capital punishment, attending to the multiple, and often subtle, ways in which race matters (Kutateladze et al., 2014).

Like any study, however, the contributions of this research are bounded by its shortcomings. Although evidence can play a key role in both arrest and prosecution decisions, I was unable to control for the strength of evidence, relying instead on proxy measures—for example, location, victim–offender relationship, murder weapon, and circumstance (Baldus & Woodworth, 2009; Baldus, Woodworth, Zuckerman, et al., 2009; Riedel, 2008). Moreover, the hazard rate helps to control for the availability/strength of evidence at the charging stage by explicitly modeling selection into the system (Bushway et al., 2007; Leiber & Fox, 2005). A second consideration is the study’s limited geographical focus on a single California County. While the study is constrained to one county, the policy and theoretical relevancy of LA County warrants the analysis of this jurisdiction. Death penalty practices are fundamentally local (Liebman & Clarke, 2011) and thus focusing on a single jurisdiction provides a more nuanced account of prosecutorial decision-making. Moreover, LA County’s homicide and death-sentencing rates in the early 1990s were comparable to other large urban jurisdictions, thereby increasing the generalizability of model estimates (Bureau of Justice Statistics, 2014; U.S. Census, 1990).¹³

Future research should examine the relationship between death penalty charging practices and later outcomes in the process. While regression models reveal victim race effects, it is unclear whether these effects translate into racial disparities in terms of capital trial rates. Given the paucity of research on the initial stages of the death penalty system, this study focused on clearance and charging decisions, but future research should examine later stages in the process. Cumulative disadvantage theory predicts that the victim race effects observed here would compound as cases advance through the death penalty system to the trial stage, yet this postulate requires empirical assessment and validation.

Notwithstanding these limitations, the results of this study have a number of implications. Given that the public often views modern capital punishment as a valuation of the victim, such patterns could reinforce stereotypes about “worthy” victims (Baldus & Woodworth, 2003; Baldus, Woodworth, Zuckerman, et al., 2009). More broadly, racial bias in handling of homicide cases can leave minority communities feeling marginalized and undermine the public’s confidence in the criminal justice system as a whole (CCFAJ, 2008; Riedel, 2008). The present study also extends prior research to a broader range of institutional processes and racial/ethnic groups. Research on cumulative race effects and loosely coupled systems has mainly focused on defendants, and thus the findings presented here shed new light on victim-based processes. Moreover, relatively little attention has been devoted to Latinos in the death penalty context as well as the sentencing literature more generally (Baldus & Woodworth, 2003; Baldus et al., 2009; Kutateladze et al., 2014; Spohn, 2000; Zatz, 2000). The inclusion of Latinos in this study offers novel insights into the criminal justice processing of this growing and increasingly criminalized group (Chavez, 2013).

These findings also contribute to ongoing capital punishment debates. In 2012, Proposition 34—a measure that sought to replace California’s death penalty with life-without-parole—was narrowly defeated (Elias, 2012). Cost issues were central to the Proposition 34 campaign given the financial toll that capital cases exact on county budgets, hampering homicide investigations by requiring the diversion of law enforcement resources to the DA’s office (Rupp, 2002). Recognizing these trade-offs, Proposition 34 sought to reallocate death penalty funds to help solve more murders (California Legislative Analyst Office, 2012). By examining the linkage between police and prosecutorial responses to homicide, this study sheds light on how the reallocation of death penalty funds toward homicide investigations could not only help to improve the quality of homicide investigations but may also aid in reducing racial biases at the charging stage.

In light of these policy debates, assessing the police–prosecution nexus is especially important in California. The CCFAJ (2008, p. 4) report characterized California’s death penalty system as “broken” in terms of its economic costs and the quality of justice it affords, outlining two potential remedies: (1) increased funding for capital litigation and (2) narrowing the number of death-eligible offenses. This study suggests that single-stage reforms may not sufficiently improve the quality of justice afforded to capital defendants in California, as they would not likely deal with the interconnectedness of criminal justice and the cumulative nature of racial inequality. Instead, policy reforms should take a more holistic approach, addressing multiple stages of the death penalty process and their interconnectedness.

In summary, this study finds that cases involving Latino victims are less likely to be cleared. Moreover, homicides with Black and Latino victims are less likely to result in a death penalty–eligible offense. These racially disparate charging practices are partially explained by lower clearance rates among Latino victims, underscoring the cumulative effects of victim race. This study contributes to the death penalty literature by shedding new light on the formation of racial bias in potentially capital cases and its mutation across multiple decision-making points. In LA County, race permeates the policing and prosecution of homicides, shaping the life course of a case from the crime scene investigation onward, ultimately influencing the filing of death penalty–eligible charges.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Acknowledgements

The author thanks the anonymous reviewers for their constructive feedback. I would also wish to thank Mona Lynch, Ann Hironaka, Carroll Seron, Shauhin Talesh, Anjuli Verma, Anna Raup-Kounovsky, Levi LaChappelle, and James Pratt for their insightful comments on earlier drafts of this paper. Special thanks to Dick O’Reilly, Doug Smith, and Sandra Poindexter for helping me access the data.

Supplemental Material

The online appendices are available at

References

Acker

Lanier

(1993). Aggravating circumstances and capital punishment law: Rhetoric or real reforms. Criminal Law Bulletin, 29, 467–501.

Alarid

L. F.

Sims

B. A.

Ruiz

(2011). Juvenile probation and police partnerships as loosely coupled systems: A qualitative analysis. Youth Violence and Juvenile Justice, 9, 79–95.

Albonetti

C. A.

(1986). Criminality, prosecutorial screening, and uncertainty: Toward a theory of discretionary decision making in felony case processings. Criminology, 24, 623–644.

Albonetti

C. A.

(1987). Prosecutorial discretion: The effects of uncertainty. Law & Society Review, 21, 291.

Baldus

D. C.

Woodworth

(2003). Race discrimination and the death penalty: An empirical and legal overview. In Bohm

R. M.

Lanier

C. S.

(Eds.), America’s experiment with capital punishment: Reflections on the past, present, and future of the ultimate penal sanction (pp. 470–501). Durham, NC: Carolina Academic Press.

Baldus

D. C.

Woodworth

Zuckerman

Weiner

Grosso

(2009). Empirical studies of race and geographic discrimination in the administration of the death penalty: A primer on the key methodological issues. In Lanier

C. S.

Acker

J. R.

(Eds.), The future of America’s death penalty: An agenda for the next generation of capital punishment research (pp. 153–197). Durham, NC: Carolina Academic Press.

Baldus

D. C.

Woodworth

Pulaski

(1990). Equal justice and the death penalty: A legal and empirical analysis. Lebanon, NH: Upne.

Baldus

D. C.

Woodworth

Weiner

(2009). Perspectives, approaches, and future directions in death penalty proportionality studies. In Lanier

C. S.

Bowers

W. J.

Acker

J. R.

(Eds.), The future of America’s death penalty: An agenda for the next generation of capital punishment research (pp. 135–151). Durham, NC: Carolina Academic Press.

Baumer

E. P.

Messner

S. F.

Felson

R. B.

(2000). The role of victim characteristics in the disposition of murder cases. Justice Quarterly, 17, 281–307.

10.

Beckett

Nyrop

Pfingst

(2006). Race, Drugs, and Policing: Understanding Disparities in Drug Delivery Arrests. Criminology, 44(1), 105–137.

11.

Berk

(1983). An introduction to sample selection bias in sociological data. American Sociological Review, 48, 386–398.

12.

Berk

Boger

Weiss

(1993). Chance and the death penalty. Law & Society Review, 27, 89.

13.

Bishop

D. M.

Leiber

Johnson

(2010). Contexts of decision making in the juvenile justice system: An organizational approach to understanding minority overrepresentation. Youth Violence and Juvenile Justice, 8, 213–233.

14.

Black

(1976). The behavior of law. New York, NY: Academic Press.

15.

Blumstein

Wallman

(2006). The crime drop in America. Cambridge, MA: Cambridge University Press.

16.

Borg

M. J.

Parker

K. F.

(2001). Mobilizing law in urban areas: The social structure of homicide clearance rates. Law and Society Review, 35, 435–466.

17.

Bowers

W. J.

(1983). The pervasiveness of arbitrariness and discrimination under post-“Furman” capital statutes. Journal of Criminal Law and Criminology, 74, 1067–1100.

18.

Bowers

W. J.

Pierce

G. L.

(1980). Arbitrariness and discrimination under post-Furman capital statutes. Crime & Delinquency, 26, 563–632.

19.

Bright

S. B.

(1994). Discrimination, death and denial: The tolerance of racial discrimination in infliction of the death penalty. Santa Clara Law Review, 35, 433.

20.

Bureau of Justice Statistics. (2014). Homicide trends in the United States, 1980–2008. Retrieved from http://www.bjs.gov/index.cfm?ty=pbdetail&iid=2221

21.

Bushway

Johnson

B. D.

Slocum

L. A.

(2007). Is the magic still there? The use of the Heckman two-step correction for selection bias in criminology. Journal of Quantitative Criminology, 23, 151–178.

22.

California Commission on the Fair Administration of Justice. (2008). Official recommendations on the fair administration of the death penalty in California. Retrieved from http://www.ccfaj.org/documents/reports/dp/official/FINAL%20REPORT%20DEATH%20PENALTY.pdf

23.

California Department of Corrections and Rehabilitation. (2015). Death row tracking system: Condemned inmate list (secure). Retrieved from http://www.cdcr.ca.gov/capital_punishment/docs/condemnedinmatelistsecure.pdf

24.

California Department of Justice. (2010). Homicide in California, 2010. Retrieved from http://oag.ca.gov/sites/all/files/agweb/pdfs/cjsc/publications/homicide/hm10/preface.pdf

25.

California Legislative Analyst Office. (2012). Proposition 34. Death penalty. Initiative statute. Retrieved from http://vig.cdn.sos.ca.gov/2012/general/pdf/34-title-summ-analysis.pdf

26.

Carter

L. E.

Kreitzberg

Howe

(2012). Understanding capital punishment law. New York, NY: LexisNexis.

27.

Chavez

(2013). The Latino threat: Constructing immigrants, citizens, and the nation. Stanford, CA: Stanford University Press.

28.

Demuth

(2003). Racial and ethnic differences in pretrial release decisions and outcomes: A comparison of Hispanic, Black, and White felony arrestees. Criminology, 41, 873–908.

29.

Elias

(2012). Calif. voters retain death penalty despite costs. Retrieved from http://www.huffingtonpost.com/huff-wires/20121107/us-prop-34-death-penalty/

30.

Fox

J. A.

Swatt

M. L.

(2009). Multiple imputation of the supplementary homicide reports, 1976–2005. Journal of Quantitative Criminology, 25, 51–77.

31.

Frohmann

(1991). Discrediting victims’ allegations of sexual assault: Prosecutorial accounts of case rejections. Social Problems, 38, 213–226.

32.

Frohmann

(1997). Convictability and discordant locales: Reproducing race, class, and gender ideologies in prosecutorial decision making. Law & Society Review, 31, 531.

33.

Glassman

R. B.

(1973). Persistence and loose coupling in living systems. Behavioral Science, 18, 83–98.

34.

Grosso

C. M.

O’Brien

Taylor

Woodworth

(2014). Race discrimination and the death penalty: An empirical and legal overview. In Acker

Bohm

Lanier

(Eds.), America’s experiment with capital punishment: Reflections on the past, present, and future of the ultimate penal sanction (pp. 525–576). Durham, NC: Carolina Academic Press.

35.

Hagan

(1973). Extra-legal attributes and criminal sentencing: An assessment of a sociological viewpoint. Law & Society Review, 8, 357.

36.

Hagan

(1989). Why is there so little criminal justice theory? Neglected macro- and micro-level links between organization and power. Journal of Research in Crime and Delinquency, 26, 116–135.

37.

Hagan

Hewitt

J. D.

Alwin

D. F.

(1979). Ceremonial justice: Crime and punishment in a loosely coupled system. Social Forces, 58, 506–527.

38.

Herbert

(1996). The normative ordering of police territoriality: Making and marking space with the Los Angeles Police Department. Annals of the Association of American Geographers, 86, 567–582.

39.

Jackson

M. A.

Webster

C. D.

Hagan

J. L.

(1982). Probation outcome: Is it necessary to fulfill the conditions. Canadian Journal of Criminology, 24, 267.

40.

Jacobs

Wood

(1999). Interracial conflict and interracial homicide: Do political and economic rivalries explain white killings of blacks or black killings of whites? American Journal of Sociology, 105, 157–190.

41.

Kaplan

Ganschow

Angioli

Tabin

(2009). Public but opaque: The problems of tracking homicide charging in a California county. Justice Policy Journal, 6, 1–29.

42.

Keil

T. J.

Vito

G. F.

(1990). Race and the death penalty in Kentucky murder trials: An analysis of post-Gregg outcomes. Justice Quarterly, 7, 189–207.

43.

Kreitzberg

(2008). A review of special circumstances in California death penalty cases. Retrieved from http://www.ccfaj.org/documents/reports/dp/expert/Kreitzberg.pdf

44.

Kutateladze

Andiloro

N. R.

Johnson

B. D.

Spohn

C. C.

(2014). Cumulative disadvantage: Examining racial and ethnic disparity in prosecution and sentencing. Criminology, 52, 514–551.

45.

Kutateladze

Lynn

Liang

(2012). Do race and ethnicity matter in prosecution? A review of empirical studies. Vera Institute of Justice. Retrieved from http://www.vera.org/pubs/do-race-and-ethnicity-matter-prosecution-review-empirical-studies

46.

LA County District Attorney’s Office. (2015). Office Overview. Retrieved from http://da.co.la.ca.us/about/office-overview

47.

LaFree

Baumer

E. P.

O’Brien

(2010). Still separate and unequal? A city-level analysis of the black–white gap in homicide arrests since 1960. American Sociological Review, 75, 75–100.

48.

Landes

W. M.

(1971). An economic analysis of the courts. Journal of Law and Economics, 14, 61–107.

49.

Langan

P. A.

Brown

J. M.

(1997). Felony sentences in state courts, 1994. US Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from http://bjs.gov/content/pub/pdf/fssc94.pdf

50.

Lee

(2005). The value of life in death: Multiple regression and event history analyses of homicide clearance in Los Angeles County. Journal of Criminal Justice, 33, 527–534.

51.

Lee

(2007). Hispanics and the death penalty: Discriminatory charging practices in San Joaquin County, California. Journal of Criminal Justice, 35, 17–27.

52.

Leiber

M. J.

Fox

K. C.

(2005). Race and the impact of detention on juvenile justice decision making. Crime & Delinquency, 51, 470–497.

53.

Leiber

M. J.

Jamieson

K. M.

(1995). Race and decision making within juvenile justice: The importance of context. Journal of Quantitative Criminology, 11, 363–384.

54.

Leiber

M. J.

Mack

K. Y.

(2003). The individual and joint effects of race, gender, and family status on juvenile justice decision-making. Journal of Research in Crime and Delinquency, 40, 34–70.

55.

Liebman

J. S.

Clarke

(2011). Minority practice, majority’s burden: The death penalty today. Ohio State Journal of Criminal Law, 9, 255.

56.

Litwin

K. J.

(2004). A multilevel multivariate analysis of factors affecting homicide clearances. Journal of Research in Crime and Delinquency, 41, 327–351.

57.

Litwin

K. J.

(2007). The dynamic nature of homicide clearances a multilevel model comparison of three time periods. Homicide Studies, 11, 94–114.

58.

Messner

S. F.

South

S. J.

(1992). Interracial homicide: A macrostructural-opportunity perspective. Sociological Forum, 7, 517–536.

59.

Orton

J. D.

Weick

K. E.

(1990). Loosely coupled systems: A reconceptualization. Academy of Management Review, 15, 203–223.

60.

Paternoster

Brame

Bacon

Ditchfield

(2004). Justice by geography and race: The administration of the death penalty in Maryland, 1978–1999. University of Maryland Law Journal of Race, Religion, Gender and Class, 4, 1.

61.

Petersen

(2016). Neighborhood context and media constructions of murder: A multilevel analysis of homicide newspaper coverage in Los Angeles County. Homicide Studies, 20(1) 25–52.

62.

Petersen

(forthcoming). Neighbourhood context and unsolved murders: The social ecology of homicide investigations. Policing and Society, 1–21. DOI: 10.1080/10439463.2015.1063629.

63.

Petersen

Lynch

(2013). Prosecutorial discretion, hidden costs, and the death penalty: The case of Los Angeles County. Journal of Criminal Law & Criminology, 102, 1233.

64.

Petersen

Ward

(2015). The transmission of historical racial violence: Lynching, civil rights-era terror, and contemporary interracial homicide. Race and Justice: An International Journal, 5, 114–143.

65.

Peterson

R. D.

Bailey

W. C.

(1991). Felony murder and capital punishment: An examination of the deterrence question. Criminology, 29, 367.

66.

Phillips

(2009a). Legal disparities in the capital of capital punishment. The Journal of Criminal Law and Criminology, 99, 717–756.

67.

Phillips

(2009b). Status disparities in the capital of capital punishment. Law & Society Review, 43, 807–838.

68.

Pierce

Radelet

(2005). Impact of legally inappropriate factors on death sentencing for California homicides, 1990–1999. The Santa Clara Law Review, 46, 1.

69.

Puckett

J. L.

Lundman

R. J.

(2003). Factors affecting homicide clearances: Multivariate analysis of a more complete conceptual framework. Journal of Research in Crime and Delinquency, 40, 171–193.

70.

Radelet

(1981). Racial characteristics and the imposition of the death penalty. American Sociological Review, 46, 918–927.

71.

Radelet

Pierce

(1985). Race and prosecutorial discretion in homicide cases. Law and Society Review, 19, 587–621.

72.

Radelet

Pierce

(2009). Racial and ethnic disparities in resolving. In Lanier

C. S.

Acker

J. R.

(Eds.), America’s experiment with capital punishment: Reflections on the past, present, and future of the ultimate penal sanction (pp. 113–134). Durham, NC: Carolina Academic Press.

73.

Rasmusen

Raghav

Ramseyer

(2009). Convictions versus conviction rates: The prosecutor’s choice. American Law and Economics Review, 11, 47–78.

74.

Regoeczi

W. C.

Jarvis

J. P.

(2013). Beyond the social production of homicide rates: Extending social disorganization theory to explain homicide case outcomes. Justice Quarterly, 30, 983–1014.

75.

Reiss

A. J.

(1973). The police and the public (Vol. 39). New Haven, CT: Yale University Press.

76.

Riedel

(2008). Homicide arrest clearances: A review of the literature. Sociology Compass, 2, 1145–1164.

77.

Rodriguez

(2010). The cumulative effect of race and ethnicity in juvenile court outcomes and why preadjudication detention matters. Journal of Research in Crime and Delinquency, 47, 391–413.

78.

Rupp

(2002). Death penalty prosecutorial charging decisions and county budgetary restrictions: Is the death penalty arbitrarily applied based on county funding. Fordham Law Review, 71, 2735.

79.

Russell-Brown

(1998). The color of crime: Racial hoaxes, white fear, black protectionism, police harassment, and other macroaggressions. New York, NY: NYU Press.

80.

Sampson

R. J.

Laub

J. H.

(1993). Structural variations in juvenile court processing: Inequality, the underclass, and social control. Law and Society Review, 27, 285–311.

81.

Schafer

J. L.

(1999). Multiple imputation: A primer. Statistical Methods in Medical Research, 8, 3–15.

82.

Schlesinger

(2008). The cumulative effects of racial disparities in criminal processing. The Advocate, 13, 22–34.

83.

Shatz

S. F.

(2007). Eighth amendment, the death penalty, and ordinary robbery–burglary murderers: A California case study. The Florida Law Review, 59, 719.

84.

Shatz

S. F.

Rivkind

(1997). California death penalty scheme: Requiem for Furman. The New York University Law Review, 72, 1283.

85.

Songer

M. J.

Unah

(2006). Effect of race, gender, and location on prosecutorial decisions to seek the death penalty in South Carolina. The South Carolina Law Review, 58, 161.

86.

Sorensen

Wallace

(1999). Prosecutorial discretion in seeking death: An analysis of racial disparity in the pretrial stages of case processing in a Midwestern county. Justice Quarterly, 16, 559–578.

87.

Spohn

(2000). Thirty years of sentencing reform: The quest for a racially neutral sentencing process. Criminal Justice, 3, 427–501.

88.

Spohn

Sample

L. L.

(2008). The dangerous drug offender in federal court: Intersections of race, ethnicity, and culpability. Crime & Delinquency, 59, 3–31.

89.

Steen

Engen

R. L.

Gainey

R. R.

(2005). Images of danger and culpability: Racial stereotyping, case processing, and criminal sentencing. Criminology, 43, 435–468.

90.

Steffensmeier

Demuth

(2001). Ethnicity and judges’ sentencing decisions: Hispanic–Black–White comparisons. Criminology, 39, 145–178.

91.

Steffensmeier

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, black, and male. Criminology, 36, 763–798.

92.

Stolzenberg

D’Alessio

S. J.

Eitle

(2013). Race and cumulative discrimination in the prosecution of criminal defendants. Race and Justice, 3, 275–299.

93.

Sutton

J. R.

(2013). Structural bias in the sentencing of felony defendants. Social Science Research, 42, 1207–1221.

94.

Ulmer

J. T.

(2012). Recent developments and new directions in sentencing research. Justice Quarterly, 29, 1–40.

95.

Ulmer

J. T.

Johnson

(2004). Sentencing in context: A multilevel analysis. Criminology, 42, 137–178.

96.

U.S. Census. (1990). Summary tape file 1, prepared by the Bureau of the Census. Washington, DC: Bureau of Census.

97.

U.S. General Accountability Office. (1990). Death penalty sentencing: Research indicates pattern of racial disparities. Government Accountability Office, Washington, D.C. Retrieved from http://www.gao.gov/assets/220/212180.pdf

98.

Wang

Mears

D. P.

(2010a). A multilevel test of minority threat effects on sentencing. Journal of Quantitative Criminology, 26, 191–215.

99.

Wang

Mears

D. P.

(2010b). Examining the direct and interactive effects of changes in racial and ethnic threat on sentencing decisions. Journal of Research in Crime and Delinquency, 47, 522–557.

100.

Weick

K. E.

(1976). Educational organizations as loosely coupled systems. Administrative Science Quarterly, 21, 1–19.

101.

Weick

K. E.

(1982). Administering education in loosely coupled schools. Phi Delta Kappan, 63, 673–676.

102.

Weiss

Berk

Lee

(1996). Assessing the capriciousness of death penalty charging. Law and Society Review, 30, 607–626.

103.

Weiss

Berk

Farrell-Ross

(1999). Death penalty charging in Los Angeles County: An illustrative data analysis using skeptical priors. Sociological Methods & Research, 28, 91–115.

104.

Yarvis

R. M.

(2000). Homicide, psychopathology, prosecutorial and jury discretion and the death penalty. Criminal Behaviour and Mental Health, 10, 256–268.

105.

Zatz

M. S.

(1987). The changing forms of racial/ethnic biases in sentencing. Journal of Research in Crime and Delinquency, 24, 69–92.

106.

Zatz

M. S.

(2000). The convergence of race, ethnicity, gender, and class on court decision making: Looking toward the 21st century. Criminal Justice, 3, 503–552.

107.

Furman v. Georgia, 408 U.S. 238, 92 S. Ct. 2726, 33 L. Ed. 2d 346 (1972).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.07 MB

0.00 MB