Race,Ethnicity,and Punishment for Federal Sex Offenses: Changing Composition in Child Pornography and Sex Abuse Cases and Temporal Disparity in Sentencing Over Time

Abstract

In response to several high-profile violent offenses against children over the past two decades, Congress has enacted several pieces of legislation aimed at increasing punishment for those convicted of federal sex offenses. Coinciding with these enhanced penalties was a demographic shift in the composition of those prosecuted for these offenses. In the federal criminal justice system, sex offenses fall into two main categories: child pornography and sexual abuse. The racial and ethnic makeup of individuals charged for federal sex offenses has significantly changed over the past 15 years. The current study utilizes federal sentencing data from the United States Sentencing Commission for the years 2006 to 2017 to explore the nature of punishment for these offenses over time. We also employ multivariate analyses to examine differences in punishment for the two types of offenses and employ disaggregated analyses by offense type to examine temporal change in racial/ethnic disparities in sentence length and departures from the federal sentencing guidelines. Findings demonstrate that convicted individuals who are Black and Hispanic are receiving harsher sentences over time net of controls for other key predictors such as age, sex, criminal history, and presumptive sentence length. Implications for how legislatively enhanced penalties and changing demographic makeup of those convicted for these offenses may have introduced extra-legal disparity into federal sentencing are further discussed.

Keywords

child pornography sexual abuse federal sentencing practices punishment disparity race ethnicity and sentencing

Introduction

Legislation to increase penalties for those convicted of sex offenses has proliferated in the last two decades as the media has intensified their focus on violent offenses against children.¹ Sex offenses, however, account for a very small proportion of all offenses sentenced in federal district court, roughly three and a half percent in 2016 (United States Sentencing Commission [USSC], 2019). In addition, because of the proliferation of the internet and ease of file sharing, Congress has increasingly targeted child pornography offenses for punishment increases citing production and distribution of these materials as akin to abuse (Kaiser & Spohn, 2014; USSC, 2009), and a slate of new and enhanced legislation aimed at curbing child exploitation has been enacted in recent years (Adams & Flynn, 2017; Motivans & Kyckelhahn, 2007).

The determinants of punishment in the federal criminal justice system have been the topic of much research due to the availability of data on federal sentencing and because the federal system relies upon rigid sentencing guidelines to provide sentencing recommendations on the basis of the seriousness of the crime and a defendant’s criminal history (Frase, 2019). Although sentencing guidelines became advisory in 2006 following the Supreme Court decision in U.S. v. Booker (2005), average sentence lengths (Frase, 2019) and extra-legal disparity in sentencing did not change much in subsequent years (Ulmer et al., 2011), though guidelines departures slightly increased in the aftermath of this policy change (Hofer, 2007). The results of the collective body of federal sentencing disparities confirm that legally relevant factors are the primary determinants of punishment (i.e., offense seriousness and defendant criminal history), but that extra-legal defendant characteristics also affect judicial sentencing decisions in certain contexts (Baumer, 2013; Ulmer, 2012). More specifically, this empirical evidence demonstrates that male and minority defendants receive harsher punishments, net of controls for legally relevant factors, than their female and White counterparts (for more thorough reviews of these race, ethnicity, and gender effects, see Baumer, 2013; Bontrager et al., 2013; Doerner & Demuth, 2010; Mitchell, 2005; Spohn, 2017; Ulmer, 2012).

Most of the existing federal sentencing studies, however, have focused on understanding racial/ethnic disparities among all defendants sentenced in federal courts regardless of the type of crime charged (Feldmeyer & Ulmer, 2011; Franklin & Henry, 2020; Johnson & Betsinger, 2009; Light et al., 2014; Mustard, 2001; Steffensmeier & Demuth, 2000; Ulmer et al., 2016) or among specific offense types such as federal immigration (Hartley & Tillyer, 2012) or narcotics offenses (Lynch & Omori, 2018; Ward et al., 2016). In contrast, little research has examined the determinants of sentencing outcomes for those convicted of federal sex offenses; studies examining racial/ethnic disparities that emerge among federal sex offense cases specifically are even more sparse.

This is a notable gap considering that a recent report from the USSC (2019) concluded that (a) both punishment severity and the number of federal sex offending convictions in federal courts have increased in the last two decades, (b) the racial/ethnic composition of those convicted for federal sex offenses differs greatly from those convicted of other offense types, and (c) the racial/ethnic composition of those convicted for federal sex offenses has changed substantially over time. Regarding the latter, the proportion of White and Native Americans charged with sexual abuse has declined in recent years, whereas the proportion of Black and Hispanics has risen substantially (USSC, 2019).

Even so, extant research has not thoroughly examined changes in the levels of sentencing severity for federal sex offenses (e.g., sexual abuse and child pornography cases), and has not assessed whether the changing racial/ethnic composition of individuals charged with federal sex offenses has contributed the racial/ethnic sentencing disparities emerging over time. The current study aims to fill this gap by examining sentence length and sentencing guideline departure outcomes in federal sex offense cases utilizing the U.S. Sentencing Commission data over time; across 12 years (fiscal years 2006–2017) for those convicted of sexual abuse, and across 8 years (fiscal years 2010–2017) for child pornography.²

Considering the changing nature of federal sexual offense cases over this period, we employ multivariate analyses to (a) examine whether punishment for sexual abuse compared with child pornography cases have changed over time, (b) assess the degree of changes in the racial/ethnic composition of individuals convicted for these specific offenses over time, and (c) investigate whether changes in the demographic composition of federal sex offenses has influenced racial and ethnic disparity in punishment over time. In the sections that follow, we review literature on federal sex offenses and disparities in sentencing, noting gaps in the existing research. In doing so, we highlight the contributions of the current study to this body of literature, and then describe our sample, measures, and analytic strategy. We conclude with a discussion of our findings and the implications for public policy.

Federal Sex Offenses

In the federal criminal justice system, most convictions for sex offending are for child pornography and sexual abuse (USSC, 2019). Child pornography offenses comprise possession, receipt, or distribution of child pornography, whereas sexual abuse offenses involve sexual contact or attempted contact with a child or an adult victim, as well as production of child pornography (Motivans & Kyckelhahn, 2007; USSC, 2019).³ Although sex offenses are considered to be of a more serious in nature than other types of crime (Tewksbury, 2014), they encompass a small percentage of federal prosecutions and convictions, though they have increased as a share of the federal caseload since the mid-2000s. In 2006, for example, sex offenses made up only 2.5% (n = 2,039) of those convicted in federal district court (Motivans & Kychelhahn, 2007) but in 2016, they comprised just above 4% (n = 2,633) of those convicted (USSC, 2019).

A number of factors influence the decision to prosecute cases at the federal level, including the seriousness and nature of the offense, such as trafficking of minors for sex, cases where the quantity of pornographic materials reaches a certain threshold, whether a crime entailed the crossing of state lines, and crimes where there are large organized distribution or trafficking networks (Motivans & Kyckelhahn, 2007). In addition, because the federal government holds jurisdiction over most crimes committed on land within Indian Country,⁴ Native Americans have historically been overrepresented among those charged with sex offenses at the federal level (Droske, 2008).

The increase in sex offending cases at the federal level in recent years may stem from federal task forces and other Department of Justice (DOJ) programs that have been established, one of which included Project Safe Childhood, to coordinate all national and international operations involving the sexual exploitation of children by various agencies such as the FBI, the Department of Homeland Security, the U.S. Postal Inspection Service, and the Criminal Division of the DOJ (Motivans & Kyckelhahn, 2007). Due to overlapping subject matter jurisdiction, state agencies investigating these types of offenses might also involve federal agencies to assist them in apprehension and prosecution.

Disparity in Sentencing Practices for Those Convicted of Different Types of Sex Offenses

The discretion afforded to judges in the punishment phase is often cited as a source for disparity in sentencing practices (Albonetti, 1987, 1991; Maddan & Hartley, 2018). The focal concerns theory of sentencing aims to capture the considerations made by judges when determining a sentence for a convicted individual, and therefore can help explain disparity in outcomes (Steffensmeier et al., 1998). Because of limited information and uncertainty during sentencing processes, the theory suggests judges must utilize perceptual shorthands based on their past experiences and characteristics of the crime and defendant when meting out punishment.

More concisely, the theory suggests that three focal concerns guide judicial sentencing decisions: (a) blameworthiness and culpability of the defendant, (b) protection of the community, and (c) practical constraints to sentencing (Steffensmeier et al., 1998). Blameworthiness includes features such as an offender’s culpability and the level of harm done. Protection of the community references the need to protect others from future harm. Practical considerations include a variety of factors ranging from overcrowding of prisons and jails, the cost of incarceration, and the ability of an individual to spend time incarcerated. According to the theory, in the absence of relevant information, judges and other courtroom actors rely on characteristics of the defendants such as their race/ethnicity or features of the crime they are charged with to inform their punishment decisions. In general, the focal concerns theory has found support in prior literature as young–minority–male defendants tend to receive the harshest punishment (Maddan & Hartley, 2018; Spohn & Holleran, 2000; Steffensmeier et al., 2017, 1998) although recent critiques have characterized the focal concerns theory as lacking construct development to better enable explanatory models testing its empirical merit (Lynch, 2019).

One might argue that the decision to prosecute persons charged with a sex offense at the federal level denotes the seriousness of the offense, and that those convicted would be viewed by a judge as very blameworthy and a high risk to the community. In general, persons convicted of sex offenses occupy a unique position in the court of public opinion as they are portrayed as very heinous, and at high risk of reoffense; therefore, incarceration is warranted to protect the community from future victimization (Huebner & Bynum, 2006; Levenson et al., 2007; Pickett et al., 2013; Tewksbury, 2014). For instance, as Pickett and colleagues (2013) suggest,

in the contemporary United States, individuals convicted of committing sexual offenses experience a higher level and intensity of public condemnation, and a greater degree of stigmatization and forfeiture of their civil liberties, than any other type of criminal offender, nonviolent or violent. (p. 730)

Research suggests that judges share similar views toward convicted individuals as those in the general public, and such perceptions of certain defendants as particularly dangerous, culpable, and posing a risk to public safety can influence sentencing decisions (Rachlinski et al., 2009).

There are differences in the types of offenses comprising a federal sex offense, as well as the type of punishments available to a judge at the sentencing phase. For instance, the USSC classifies sexual abuse cases as “contact” offenses (i.e., an offense involving the actual or attempted sexual contact with a victim). Moreover, any sexual abuse offense with a minor is subjected to a federal mandatory minimum sentence (USSC, 2019). Whereas federal child pornography does not focus on anyone who had physical contact with a victim, but rather is constrained to those involved in the distribution, receipt, and possession of such materials (USSC, 2019). Furthermore, research has found that those convicted of contact offenses differ in a number of keyways, including having more access to children, having more antisocial personality traits, and having fewer psychological barriers to contact sex offending such as less victim empathy compared with those convicted of child pornography offenses (Babchishin et al., 2015). Accordingly, individuals convicted of contact sex offenses may be more likely to be viewed as both more blameworthy and as a greater threat to the safety of community members, and therefore warrant more severe punishment. However, as previously noted, over recent years, federal laws have increasingly sought to target and increase punishment for child pornography offenses, and perceptions of these types of crimes have changed, whereby they are viewed as similar to sexual abuse (Kaiser & Spohn, 2014; USSC, 2009). Indeed, the presumptive sentence for a child pornography conviction under the federal sentencing guidelines is, in some cases, 7 times larger than it was before congress began revising these statutes (Hamilton, 2011).⁵

Hessick (2011) argues that conflating internet based- and contact-sex offenses blurs the reality of sexual abuse because children are often abused by someone who is a caretaker or someone who they know, rather than by a stranger, which leads to misperceptions by the public of child sex abuse as a “stranger-danger issue” (p. 2). Despite these misperceptions, lawmakers defend the harsh sentences for those convicted of child pornography due to the harm associated with widespread dissemination of pornographic images of children on the internet (Kaiser & Spohn, 2014). Further complicating sentencing practices for those convicted of federal sex offenses is that the USSC itself has stated that the sentencing ranges for child pornography are too severe, especially for nonproduction offenses (USSC, 2012), and judges have increasingly handed out shorter sentences via downward departures in these cases citing policy disagreements with the guidelines (Kaiser & Spohn, 2014). However, the question of whether over time child pornography cases are treated similarly, more leniently, or more severely than sexual abuse cases has not been examined.

Federal Sentencing, Racial/Ethnic Disparity, and Sex Offenses

Although the seriousness of the offense and the defendant’s prior criminal record continue to be the primary determinants of punishment decisions, extra-legal correlates are also influential. Collectively, the results of sentencing research demonstrate that male and Black and Hispanic defendants receive harsher punishment outcomes compared with their female and White counterparts, net of controls (for reviews of this research, see Baumer, 2013; Bontrager et al., 2013; Doerner & Demuth, 2010; Mitchell, 2005; Spohn, 2017; Ulmer, 2012). The focal concerns theory in particular posits that because of societal stereotypes, minority defendants will be perceived to be more blameworthy, to pose a greater threat to the safety of community members, and be more capable of serving time behind bars, and thus will receive harsher punishments (Steffensmeier et al., 1998). The majority of previous empirical research has interpreted findings of extra-legal disparity as providing support for a focal concerns framework of judicial decision making (Maddan & Hartley, 2018). Notably, very few studies examining racial and ethnic disparity, however, go beyond Black–White–Hispanic comparisons, in fact, only a handful of studies have examined the treatment of Other races compared with Whites, such as Asian defendants (Johnson & Betsinger, 2009) and Native American defendants (Franklin, 2013; Franklin & Henry, 2020; Ulmer & Bradley, 2019).

Although research has investigated racial disparities at the federal level across a range of offense types (see Hartley & Tillyer, 2012; Lynch & Omori, 2018; Ward et al., 2016), very limited research has assessed the punishment of sex offenses in the federal system. To be sure, this is a notable gap in the literature considering that the racial/ethnic composition of those convicted of federal sex offenses is considerably different from many other offense types with Native Americans making up a larger share of those prosecuted for federal sex offenses (Droske, 2008; USSC, 2019). Moreover, the racial/ethnic composition of those convicted of federal sex offenses differs substantially when comparing sexual abuse and child pornography offenses. For instance, those convicted of federal sexual abuse offenses comprise a much more diverse group with nearly half being White, 23.6% Black, 13.3% Hispanic, and 11.7% Native American based on federal statistics from 2016. In comparison, child pornography cases are far more monolithic, with 80.9% White, 3.8% Black, 12.9% Hispanic, and approximately 1% Native American cases (USSC, 2019).

In addition, there is also a dearth of research on how racial/ethnic disparities within classes of federal sex offenses have changed over time. Indeed, recent work exploring sentencing trends across time finds racial and ethnic disparities in both state and federal jurisdictions have declined in recent years (King & Light, 2019). Moreover, research on sentencing of immigration cases in the United States has found that as noncitizens made up larger shares of the federal caseload over time, sentencing for noncitizens became more severe (Light, 2014). Sentencing practices for federal sex offenses are due for an examination of variation in race/ethnic disparities across time considering that variation in the racial/ethnic composition of those convicted of these offenses over time may correspond to differential treatment.

In sum, it remains possible that changes in the demographic makeup of the federal sex offense caseloads would translate to disparities in punishment over time. Even so, no research has examined how racial disparities for sex offenses might have changed over this same period, which is especially salient given that the composition of persons convicted of federal sex offenses has shifted so drastically, and some research has shown that as certain groups increase as a percentage of cases over time, they might be perceived to be more harmful, thus warranting greater punishment severity (Light, 2014).

Current Study

Given trends in the changing composition of those convicted of federal sex offenses in recent years, this study is the first to examine punishment of these individuals over time, as well as assess racial/ethnic disparities for federal sexual abuse and child pornography cases over time. Specifically, drawing on federal sentencing data from the USSC, we examine three research questions:

Research Question 1 (RQ1): Are sentence length and downward departure decisions similar for individuals convicted of federal sexual abuse offenses compared with federal child pornography offenses over time?

Research Question 2 (RQ2): Among those convicted of federal sexual abuse offenses, are racial and ethnic disparities in sentence length and downward departure decisions changing over time?

Research Question 3 (RQ3): Among those convicted of federal child pornography offenses, are racial and ethnic disparities in sentence length and downward departure decisions changing over time?

Data and Method

This study utilizes data from the USSC on defendants sentenced in federal district court for federal sexual abuse and child pornography. The federal criminal court system comprises 94 district courts within all 50 states, the District of Columbia, and U.S. foreign territories. The USSC reports information on all cases, which result in a federal conviction and are subject to sentencing under the federal sentencing guidelines. Similar to previous studies of federal sentencing practices, the scope of the current study was limited to 90 districts (excluding foreign territories). The final analytic sample encompasses a total of 19,594 defendants sentenced in federal district court, including 14,387 child pornography cases sentenced during fiscal years 2010 to 2017 and 5,207 sexual abuse cases sentenced during fiscal years 2006 to 2017.

Dependent Variables

Our first dependent variable is the judicial decision regarding the length of sentence, which is a continuous measure ranging from 0 to 470 months.⁶ This variable was log-transformed to account for the positive skew (Johnson & Betsinger, 2009; Light et al., 2014).⁷ Our second dependent variable, examines a combined measure of receipt of a departure from the guideline sentencing range, which is a dichotomous measure of whether an individual received a judicial downward or substantial assistance departure, where a value of 1 indicates an individual received a downward departure and 0 represents no downward departure.⁸

Independent Variables

Extra-legal independent variables include Race/ethnicity, which is measured using a categorical variable that indicates whether an individual is non-Hispanic White (reference), non-Hispanic Black, Hispanic, or Native American. Sex is coded as a binary indicator of whether an individual who was convicted is male (reference) or female. Age is measured as a categorical variable specifying whether an individual is below 25 years (reference), 25 to 34, 35 to 44, 45 to 54, or 55 and older at the time of sentencing. Citizenship is a binary variable indicating whether the individual convicted is a U.S. citizen (1 = U.S. citizen, 0 = not a U.S. citizen). Educational attainment is a categorical variable measuring an individual’s highest education level: less than high school (reference), high school graduate, some college, and college graduate. Dependents is coded as a dichotomous indicator capturing whether an individual had any financial dependents (1 = has dependents, 0 = no dependents).

Legally relevant predictors include Presumptive sentence length, which measures the minimum number of months of incarceration recommended by the sentencing guidelines. The sentencing guidelines considers the 43-point scale of criminal severity and the 6-point scale of criminal history, while accounting for relevant sentencing adjustments. Consistent with the recommendations of the USSC and prior research, a variable is also included for the 6-point Criminal History Scale (Johnson et al., 2008). Multiple counts is a binary indicator capturing whether an individual was convicted of more than one criminal charge (1 = multiple counts, 0 = single count).

Finally, case processing factors are also included as controls: Guilty plea is a binary variable measuring whether an individual plead guilty (1 = plead guilty, 0 = did not plead guilty). Pretrial detention is a binary variable that measures whether the individual convicted was detained in custody before trial (1 = detained, 0 = not detained). A series of variables are included to account for departures from the recommended sentence. Upward departures measures cases in which sentences occurred above the recommended guidelines range (1 = upward departure, 0 = no upward departure). Substantial assistance departures includes cases where the sentence fell below the recommended range because an individual provided substantial assistance to the government in the prosecution of other federal criminal cases under Federal Rule 5K1.1 (1 = substantial assistance departure, 0 = no substantial assistance departure). Downward departure measures cases where sentences were below the recommended range due to a judicial departure (1 = downward departure, 0 = no downward departure). Within range identifies cases in which an individual was sentenced within the recommended range and serves as the reference category (1 = sentence within range, 0 = sentence not within range). Lastly, Mandatory minimum is a dichotomous measure of whether a mandatory minimum sexual abuse sentence was applicable in the case (1 = mandatory minimum applicable, 0 = no mandatory minimum).

Analytic Approach

To address the first research question, we limit the sample to respondents who were convicted of either a sexual abuse or child pornography offense. We use a dummy variable of offense indicator (1 = sexual abuse, 0 = child pornography) and generate a product term between offense type and time (year) to estimate whether punishment severity for sexual abuse crimes versus child pornography crimes is changing over time (see King, 2019; King & Light, 2019; Testa & Lee, 2020). As previously mentioned, because child pornography cases are only available since 2010, this analysis is limited, from 2010 to 2017 (N = 18,109).

Next, we then disaggregate offense type into two subsamples (sexual abuse cases from 2006–2017 and child pornography cases from 2010 to 2017) to assess whether racial/ethnic disparities in punishment have changed across time. Beginning with a subsample of sexual abuse cases from 2006 to 2017, we interact a categorical variable of race (White, Black, Hispanic, Native American) with time (years 2006–2017) to determine whether racial disparities in punishment outcomes among individuals convicted for sexual abuse offenses have changed across time. Finally, we repeat this procedure using the child pornography subsample for years 2010 to 2017 to assess whether racial/ethnic disparities among child pornography cases have changed over time. Across all samples, ordinary least squares regression is used to estimate the natural logarithm of sentence length (excluding cases with a value of 0 for sentence length), and logistic regression is used to estimate the likelihood of receiving a downward departure given the dichotomous nature of the dependent variable. All models use robust standard errors and include fixed effects for the sentencing district to control for between-district variation in sentencing.

Results

Table 1 reports the descriptive statistics for the analytic sample stratified by offense type for child pornography and sexual abuse cases. Most of the cases are individuals convicted of child pornography offenses and their average mean sentences are slightly higher than those convicted of sexual abuse (128 vs. 122 months). Departure rates, however, are much higher in child pornography cases, 65%, versus 37%, for sexual abuse cases. These statistics highlight that nearly two thirds of those convicted of child pornography receive some type of downward departure from the presumptive guideline sentence, supporting previous research citing judicial, and perhaps prosecutorial disagreement with the guideline sentencing ranges for most individuals convicted of these offenses (Kaiser & Spohn, 2014).

Table 1.

Summary Statistics of Sex Abuse Cases (2006–2017) and Child Pornography Cases (2010–2017) Convicted in U.S. Federal District Court.

Variable	Sex abuse (N = 5,207)		Child pornography(N = 14,387)
Variable	M	SD	M	SD
Dependent variables
Sentence	122.312	104.899	127.987	102.724
Log sentence	4.397	1.048	4.499	1.072
Any departure	0.370	0.483	0.646	0.478
Race/ethnicity
White	0.416	0.493	0.857	0.350
Black	0.180	0.384	0.041	0.198
Hispanic	0.098	0.298	0.095	0.294
Native American	0.306	0.461	0.007	0.082
Male	0.945	0.229	0.985	0.121
Age (years)
Below 25	0.183	0.387	0.084	0.277
25–34	0.322	0.467	0.280	0.449
35–44	0.229	0.420	0.244	0.430
45–54	0.157	0.364	0.216	0.411
55+	0.108	0.310	0.176	0.381
U.S. citizen	0.949	0.221	0.968	0.176
Education
Less than high school	0.273	0.446	0.099	0.298
High school graduate	0.386	0.487	0.367	0.482
Some college	0.242	0.428	0.362	0.481
College graduate	0.099	0.299	0.172	0.377
Dependents (yes/no)	0.483	0.500	0.300	0.458
Multiple counts	0.209	0.407	0.233	0.422
Plea	0.878	0.327	0.963	0.188
Pretrial detention	0.840	0.367	0.651	0.477
Criminal history	1.962	1.482	1.416	0.957
Presumptive	131.797	111.653	158.885	103.570
Mandatory minimum	0.375	0.484	0.002	0.042
Guideline departure
No departure	0.563	0.496	0.333	0.471
Upward departure	0.066	0.249	0.021	0.143
Substantial assistance departure	0.054	0.226	0.031	0.174
Downward departure	0.316	0.465	0.614	0.487

The racial composition of individuals across the two offense types is also stark. The majority of those convicted of receipt or possession of child pornography are White (86%); there are very few Black (4%), Hispanic (10%), or Native American (<1%) individuals charged with this offense. The demography of individuals convicted of sexual abuse offenses is more mixed; 42% are White, 18% are Black, 10% are Hispanic, and 31% are Native American. The large percentage of Native Americans in this offense type likely reflects the legal jurisdiction of the federal government on Native American lands or reservations. Females also make up a larger percentage of those convicted for sexual abuse but comprise a very small proportion of both types of offenses: roughly 5% of sexual abuse cases and only 1% of child pornography cases. U.S. citizens make up the majority of those convicted in both types of offenses (95% of sexual abuse cases and 97% of child pornography cases). Regarding age and education, those convicted for sexual abuse offenses tend to be younger and have less education than those convicted for child pornography; 18% of those convicted of sexual abuse are below 25 years of age versus only 8% of those convicted of child pornography, and in 27% of sexual abuse cases, the individual had less than a high school education versus only 10% of those in child pornography cases. A higher percentage of those convicted of sexual abuse have financial dependents (48% vs. 30%), are detained awaiting sentencing (84% vs. 65%), have higher average criminal history scores (1.96 vs. 1.42), and have a mandatory minimum sentence apply (38% vs. <1%). Interestingly, those convicted of child pornography offenses have higher average presumptive minimum sentences than those convicted of sexual abuse (159 months vs. 132 months) and are more likely to receive a judicial downward departure (61% vs. 32%).⁹ Finally, although fewer in number, a slightly higher percentage of substantial assistance departures are granted in sexual abuse cases (5% vs. 3%).

Figure 1 displays the average sentence length in months for both types of cases. As can be seen in both Panels A and B, the average sentence lengths for both types of cases have increased over the years studied. For those convicted of sexual abuse offenses (Panel A), sentences declined from 2006 to 2009, and then increased precipitously thereafter. Average sentences in 2009 were roughly 95 months and increased to 145 months in 2016 and dropped to just below 140 months in 2017. Panel B reveals that the sentences for those convicted of child pornography offenses have more steadily increased from a low of about 112 months in 2010 to a high of almost 140 months in 2017. From these figures, it is apparent that the punishments for those convicted of federal sex offenses have increased substantially over the years examined here.

Figure 1.

Sentence lengths for federal sex offenses across time.

Figure 2 details the changes in the racial composition of sexual abuse and child pornography cases over time, and Table 2 reports the percentage of cases across each racial/ethnic group at the start and end of the series, as well as the percent change across these two periods. The figure reveals that there have been prominent changes in the composition of sexual abuse cases over the 12-year period assessed. Among individuals convicted of sexual abuse offenses from 2006 to 2017 (Table 2), there were declines in the percentage of individuals who were White ( $Δ =$ −7.19%) and Native American ( $Δ =$ −61.54 %), whereas the percent change increased for those who were Hispanic ( $Δ =$ 58.45%) and especially Black ( $Δ =$ 1,330.47 %).¹⁰ The numbers for those convicted of child pornography are less stark but still detail heterogeneity in the racial/ethnic composition of cases over time. Figure 2, Panel B, demonstrates that among child pornography cases, the vast majority of those convicted are White, and this has remained relatively stable over time, although as the trend shows, the share of Whites convicted has slightly dipped over time and the proportion of Hispanics convicted has slightly increased. For instance, from 2010 to 2017 (Table 2), the percentage of White individuals slightly decreased ( $Δ =$ −8.66 %), whereas Black ( $Δ =$ 56.88%), Hispanic ( $Δ =$ 89.13%), and Native Americans increased ( $Δ =$ 141.86%).¹¹

Figure 2.

Racial composition of federal sex offenses across time.

Table 2.

Percent Change in Racial/Ethnic Composition of Sexual Abuse and Child Pornography Cases Over Time.

Race/ethnicity	Sexual abuse offenders			Child pornography offenders
Race/ethnicity	2006 (%)	2017 (%)	Percent change (∆)(%)	2010 (%)	2017 (%)	Percent change(∆) (%)
White	36.43	33.81	−7.19	90.29	82.47	−8.66
Black	2.33	33.33	1,330.47	3.2	5.02	56.88
Hispanic	7.75	12.28	58.45	6.07	11.48	89.13
Native American	53.49	20.57	−61.54	0.43	1.04	141.86

Table 3 provides the results of the initial model comparing changes in sentence length over time for sexual abuse relative to child pornography cases. To test our first research question, we generated a binary variable, where 1 indicates sexual abuse cases and 0 indicates child pornography cases, and interact this indicator with the time variable (2010–2017). Accordingly, the interaction coefficient provides an estimate of whether over time, sexual abuse cases are receiving sentences that are significantly longer or shorter relative to child pornography cases. The results in model 1 indicate a nonsignificant interaction term suggesting that there are no differences in sentence length of those convicted of sexual abuse versus child pornography (b = 0.002, p > .10). Figure 3, Panel A, visualizes this result, showing that from net of covariates, from 2010 to 2017, the sentence lengths for sexual abuse cases are increasing, whereas child pornography cases are decreasing. However, this difference was not found to be statistically significant.

Table 3.

Sentence Length and Departure Outcome Models for Individuals Convicted of Sex Abuse Compared With Child Pornography in U.S. District Court From 2010 to 2017.

Variables	Model 1: OLS model of sentence length		Model 2: Logit model of downward departure
Variables	B	SE	OR	SE
Sex abuse	0.006	(0.034)	0.321***	(0.033)
Year	−0.001	(0.003)	1.087***	(0.009)
Sex abuse × Year	0.002	(0.005)	1.068***	(0.020)
Black	0.038**	(0.014)	1.603***	(0.130)
Hispanic	0.009	(0.020)	0.940	(0.064)
Native American	0.098***	(0.028)	0.885	(0.093)
Male	0.122***	(0.031)	0.446***	(0.059)
Age: 25–34 years	0.103***	(0.022)	0.564***	(0.037)
Age: 35–44 years	0.125***	(0.023)	0.417***	(0.029)
Age: 45–54 years	0.134***	(0.023)	0.369***	(0.026)
Age: 55+ years	0.100***	(0.025)	0.408***	(0.030)
U.S. citizen	0.017	(0.025)	0.917	(0.093)
High school graduate	0.007	(0.017)	0.996	(0.056)
Some college	0.018	(0.018)	0.981	(0.057)
College graduate	0.012	(0.022)	1.200**	(0.082)
Dependents (yes/no)	0.016	(0.010)	0.974	(0.037)
Multiple counts	0.112***	(0.014)	0.828***	(0.040)
Plea	0.009	(0.012)	1.351***	(0.110)
Pretrial detention	0.280***	(0.016)	0.274***	(0.012)
Criminal history	0.015***	(0.003)	0.834***	(0.014)
Upward departure	0.401***	(0.017)	—	—
Substantial assistance departure	−0.704***	(0.031)	—	—
Downward departure	−0.573***	(0.008)	—	—
Presumptive (Ln)	0.986***	(0.025)	—	—
Presumptive	—	—	1.004***	(0.000)
Mandatory Minimum	0.027	(0.023)	0.758**	(0.068)
District fixed effects	Yes		Yes
Constant	−0.415**	(0.128)	28.597***	(9.649)
Observations	17,750		18,109
R²/Pseudo R²	.610		.163

Note. Robust standard errors in parentheses. OLS = ordinary least squares; OR = odds ratio.

p < .10. *p < .05. **p < .01. ***p < .001.

Figure 3.

Panel A: Predicted probability of logged sentence length for sex abuse compared to child pornography cases from 2010 to 2017. Panel B: Predicted probability of a downward departure for sex abuse compared to child pornography cases from 2010 to 2017.

Model 2 in Table 3 provides the results predicting changes in receiving a downward departure for sexual abuse cases relative to child pornography cases across time. Results in this model demonstrate that those convicted of sexual abuse offenses are significantly more likely to receive a downward departure, relative to those convicted of child pornography (odds ratio [OR] = 1.068, p < .001). Figure 3, Panel B, illustrates these results, revealing that the likelihood of a downward departure is increasing in both child pornography and sexual abuse cases over time, however, the slope is increasingly significantly steeper for sexual abuse cases.¹²

Disaggregated Offense Type Analyses

Next, we turn to examining racial and ethnic disparities over time disaggregated by offense type. Across all models, the key parameter is the interaction term of race/ethnicity and year, which indicates whether punishments (sentence length or downward departures) are significantly increasing, decreasing, or remaining the same across racial/ethnic groups over time, net of controls of demographic characteristics and legal case processing variables. Table 4, Model 1, presents the results testing for racial/ethnic disparity in sentence length for sexual abuse cases across years 2006 to 2017.¹³ The findings demonstrate that Blacks are receiving harsher punishments over time, as indicated by the positive and statistically significant coefficient for the Black × year interaction term (b = 0.017, p < .05), net of covariates. Figure 4, Panel A, elucidates these results showing that the sentences for Whites and Native Americans have been relatively stable over the 12 years of data, but that Hispanic and Black sentences have increased, with the average sentence lengths for Blacks displaying the steepest increase over time.

Table 4.

Results of Racial Disparities in Punishment Among Sex Abuse Cases Convicted in U.S. District Court From 2006 to 2017.

Variables	Model 1: OLS model of sentence length		Model 2: Logit model of downward departure
Variables	B	SE	OR	SE
Black	−0.087	(0.060)	1.371	(0.376)
Hispanic	−0.037	(0.068)	1.518	(0.436)
Native American	0.064⁺	(0.038)	0.643*	(0.129)
Year	0.002	(0.004)	1.090***	(0.018)
White × Year	—	—	—	—
Black × Year	0.017*	(0.007)	1.047	(0.034)
Hispanic × Year	0.012	(0.008)	0.954	(0.035)
Native American × Year	−0.004	(0.005)	1.053*	(0.026)
Male	0.188***	(0.044)	0.220***	(0.036)
Age: 25–34 years	0.045*	(0.022)	0.584***	(0.059)
Age: 35–44 years	0.064*	(0.026)	0.518***	(0.059)
Age: 45–54 years	0.068*	(0.028)	0.359***	(0.045)
Age: 55+ years	0.063*	(0.031)	0.403***	(0.056)
U.S. citizen	0.019	(0.031)	0.874	(0.152)
High school graduate	0.006	(0.018)	1.022	(0.089)
Some college	0.009	(0.022)	0.847	(0.088)
College graduate	0.019	(0.031)	1.126	(0.166)
Dependents (yes/no)	0.016	(0.014)	0.908	(0.065)
Multiple counts	0.128***	(0.022)	0.724**	(0.073)
Plea	−0.075***	(0.022)	2.139***	(0.293)
Pretrial detention	0.274***	(0.037)	0.417***	(0.041)
Criminal history	0.017***	(0.004)	1.020	(0.026)
Upward departure	0.487***	(0.028)	—	—
Substantial assistance departure	−0.729***	(0.048)	—	—
Downward departure	−0.494***	(0.023)	—	—
Presumptive (Ln)	0.790***	(0.042)	—	—
Presumptive	—	—	1.007***	(0.000)
Mandatory minimum	0.205***	(0.034)	0.473***	(0.046)
District fixed effects	Yes		Yes
Constant	0.431*	(0.186)	0.705	(0.977)
Observations	5,087		5,207
R²/Pseudo R²	.788		.211

Note. Robust standard errors in parentheses. OLS = ordinary least squares; OR = odds ratio.

p < .10. *p < .05. **p < .01. ***p < .001.

Figure 4.

Sexual abuse race by year sentence length and departure marginals.

Table 4, Model 2, displays the results of the model testing for racial disparity in likelihood of receiving a departure in sexual abuse cases from 2006 to 2017. The results demonstrate that Blacks and Native Americans are more likely to receive a downward departure over time, although the interaction term is statistically significant only for Native Americans (OR = 1.053, p < .05). Figure 4, Panel B, depicts these effects, demonstrating that the likelihood of receiving a departure increases for all the race/ethnic groups studied here, but that the likelihood for Blacks and Native Americans increases more sharply over time; the increasing slope for Native Americans is statistically significant (p < .05). For instance, Blacks have a .3 probability of receiving a downward departure in 2006, and this increases to .6 probability in 2017, whereas Native Americans have a .2 probability in 2006, which increases to .45 probability in 2017.¹⁴

Table 5 displays the results of racial/ethnic differences in punishment over time among the child pornography subsample from 2010 to 2017. Table 5, Model 1, assesses changes in the sentence length outcomes, revealing that among child pornography cases, Blacks are the only racial/ethnic group who are receiving significantly shorter sentences over time (b = −0.013, p < .10), net of controls for legal and extra-legal characteristics. Figure 5, Panel A, displays the results of the predicted sentence lengths by race from 2010 to 2017, showing that sentence lengths for Native Americans are marginally increasing over time, whereas the sentence lengths of Whites, Blacks, and Hispanics are decreasing over time, with the steepest decline being for Blacks. Finally, Table 5, Model 2, displays the results of the likelihood of a downward departure. The results reveal that there are no significant differences in the effects of race/ethnicity on receiving a downward departure over the time period studied. These findings are visually displayed in Figure 5, Panel B, and demonstrate that the likelihood of receiving a downward departure among child pornography cases is increasing at similar rates for all racial/ethnic groups over time.

Table 5.

Results of Racial Disparities in Punishment Among Child Pornography Cases Convicted in U.S. District Court from 2010 to 2017.

Variables	Model 1: OLS model of sentence length		Model 2: Logit model of downward departure
Variables	B	SE	OR	SE
Black	0.096**	(0.030)	0.040	(0.202)
Hispanic	0.001	(0.051)	−0.072	(0.142)
Native American	0.076	(0.096)	−0.056	(0.623)
Year	−0.003	(0.003)	0.099***	(0.010)
White × Year	—	—	—	—
Black × Year	−0.013⁺	(0.007)	−0.062	(0.045)
Hispanic × Year	−0.004	(0.011)	−0.010	(0.031)
Native American × Year	0.006	(0.019)	0.075	(0.132)
Male	0.051	(0.036)	−0.065	(0.163)
Age: 25–34 years	0.134***	(0.029)	−0.587***	(0.085)
Age: 35–44 years	0.152***	(0.030)	−0.902***	(0.087)
Age: 45–54 years	0.173***	(0.030)	−1.011***	(0.088)
Age: 55+ years	0.134***	(0.032)	−0.954***	(0.091)
U.S. citizen	−0.009	(0.032)	−0.115	(0.123)
High school graduate	0.015	(0.022)	0.050	(0.069)
Some college	0.023	(0.023)	0.094	(0.070)
College graduate	0.026	(0.027)	0.253**	(0.080)
Dependents (yes/no)	0.014	(0.012)	−0.042	(0.043)
Multiple counts	0.078***	(0.011)	–0.135*	(0.055)
Plea	0.027+	(0.016)	0.136	(0.108)
Pretrial detention	0.242***	(0.015)	−1.295***	(0.049)
Criminal history	0.014***	(0.004)	−0.331***	(0.022)
Upward departure	0.368***	(0.019)	—	—
Substantial assistance departure	−0.700***	(0.038)	—	—
Downward departure	−0.578***	(0.009)	—	—
Presumptive (Ln)	1.089***	(0.012)	—	—
Presumptive	—	—	0.003***	(0.000)
Mandatory minimum	0.024	(0.067)	−0.704+	(0.383)
District fixed effects	Yes		Yes
Constant	−0.831***	(0.087)	3.252***	(0.385)
Observations	14,102		14,387
R²/Pseudo R²	.579		.153

Note. Robust standard errors in parentheses. OLS = ordinary least squares; OR = odds ratio.

p < .10. *p < .05. **p < .01. ***p < .001.

Figure 5.

Child pornography race by year sentence length and departure marginals.

Discussion

Negative media portrayals and federal legislation for sex offenses in general have proliferated in the past two decades, and public attitudes toward these types of offenses and the individuals convicted of them have followed suit with calls for increasingly crime-control–oriented responses (Pickett et al., 2013; Sample, 2001; Tewksbury, 2014). Despite the increased national attention on sex offenses, empirical research on the treatment of individuals charged and convicted of these offenses has been relatively sparse. The current study attempted to explore the nature of punishment for individuals convicted of federal sex offenses across roughly a decade of USSC sentencing commission data. Specifically, the purpose was to examine differences in punishment for child pornography versus sexual abuse offense convictions over time, as well as to explore the impact of changes in racial/ethnic disparities on punishment for both types of offenses during this period.

Broadly speaking, the results of this study reveal no significant differences in the punishment of those convicted of child pornography versus sexual abuse offenses over time as average sentence lengths did not significantly differ by offense type. Our results demonstrate that there are no differences in punishment practices in terms of sentence length despite criticisms that those who had physical contact with a victim are receiving sentences that are less harsh than those who are charged with child pornography offenses (Hessick, 2011). This may be evidence that sex offense convictions at the federal level are for the most serious cases, regardless of whether it is a contact offense or not. As such judges view persons convicted of sex offenses as similarly blameworthy and culpable. Results from the receipt of departure analysis, however, reveal that individuals convicted of sexual abuse offenses are significantly more likely to receive a downward departure in more recent years. Despite that the probability of departure in sexual abuse cases is lower than in child pornography cases, almost twice as low (Figure 3, Panel B), the slope of the increase in probability of departures over time is significantly steeper for sexual abuse cases.

Regarding the influence of race/ethnicity on punishment over time, several interesting results emerged. For those convicted of sexual abuse offenses, results demonstrated that Blacks are subject to receiving longer sentences over time compared with Whites. In fact, over the period studied, Black individuals convicted of sexual abuse had the largest percent increase (+1,300%) as a share of those convicted of sexual abuse in federal district courts. Aside from comprising a larger share of federal sexual abuse cases, the results showed that Blacks convicted of sexual abuse are receiving significantly longer sentence lengths over time.

However, for child pornography offenses, Blacks appear to be receiving the largest decreases in average sentence lengths over time, whereas Native Americans are experiencing increases in sentence length. One potential explanation for the divergence in the race/ethnicity disparities over time across these two offense types is differences in the racial composition of them. For instance, Blacks make up a much smaller proportion of those convicted of child pornography (4% vs. 18% of sexual abuse offenses, whereas Whites represent 86% of those convicted of child pornography and 42% of those convicted of sexual abuse). Similarly, the percentage of Blacks convicted during our time series for child pornography cases rose 56.88% compared with more than 1,300% for sexual abuse cases, and for Native Americans, their composition in child pornography cases rose 141% and decreased 64% in sexual abuse cases (see Table 1). Thus, the results suggest that as certain racial groups make up a larger share of all those convicted for a specific type of sex offense over time (Blacks for sexual abuse cases and Native Americans for child pornography cases), this tends to correspond with longer sentences for members of that racial/ethnic group.

Our disaggregated models of the departure decision reveal that for sexual abuse cases, Native Americans are increasingly more likely to receive a departure over time compared with their White counterparts. There are, however, no significant interactive race × time effects for departure in child pornography cases. Again, Native Americans make up a decreasing share of those convicted of sexual abuse over the 12-year time period studied here (61.54% decrease between 2006 and 2017). Very little research has examined the treatment of Native Americans convicted of sexual abuse in federal court, but our results reveal that at the same time that Native Americans became a smaller share of all federal sexual abuse convictions, they were increasingly likely to benefit from a downward departure in federal sexual abuse cases. Under the focal concerns perspective, as the members of a racial group make up an increasing share of all individuals in a particular offense category, judges may view them as more dangerous, blameworthy, and at high risk of reoffending, and they, therefore, receive harsher sentences than other racial ethnic groups. Conversely, as certain groups make up a decreasing share of those convicted of sexual abuse, they may become viewed as less blameworthy and a threat to the community, and therefore receive more lenient sentences than other racial groups (i.e., Native Americans receiving an increasing odds of a departure over time due to being viewed as less dangerous and at risk of reoffense).

It might also be that the reason the sexual abuse cases involving Native Americans are being prosecuted in federal court is due to the offense occurring in Indian Country where federal law has jurisdiction. In that case, judges and prosecutors might see them as deserving of a downward departure because these cases for non–Native Americans, would be prosecuted in state court not federal court. Another possibility for these findings could relate to the difficulty in garnering witness testimony and victim cooperation in these cases. According to the DOJ (2017),

although difficulties in prosecuting physical assault, sexual assault, and child molestation cases are not unique to Indian country, structural barriers in Indian country may compound the challenges . . . Victims and witnesses must often travel long distances to the Federal courthouse to testify. (p. 37)

And, “investigators and prosecutors may encounter difficulties developing the rapport and trust needed to encourage victims to cooperate.” Relatedly, it is also likely that compared with other sexual abuse cases prosecuted at the federal level (i.e., sex trafficking), the cases involving Native Americans are relatively less serious. Indeed, Native Americans were less likely than others to have a mandatory minimum to apply in their case, which might substantiate the argument that these are less serious cases, and the main reason they have been prosecuted in federal court is due to geographical jurisdiction.¹⁵

Regarding the results of direct effects across the models discussed (Tables 3 –5), certain extra-legal factors influenced the sentence length and departure decisions, net of controls for other relevant factors. Males received significantly longer sentences than females for sexual abuse cases but not for child pornography cases. Regarding age, the youngest individuals receive the shortest sentences regardless of type of offense. Those below 25 years of age receive significantly shorter sentences than all other age categories. Most of the other direct effects findings in our models were in the expected direction and similar to previous research on sentencing and departure decisions; individuals with more counts, larger criminal histories and those who were detained prior to sentencing receive significantly longer punishments and lower likelihoods of a departure (the exception is for sexual abuse cases where criminal history did not have a statistically significant effect on receipt of a departure).

The effect of presumptive sentence on both decisions was similar for both offense types; interestingly, longer presumptive sentences result in significantly longer sentences but also greater probabilities of getting a departure. Individuals subject to a mandatory minimum received significantly longer sentences in both types of offenses; mandatory minimums, however, led to a lower likelihood of a departure in sexual abuse cases and had no significant effect in child pornography cases. Regarding the effect on one’s sentence length for getting a departure, upward departures resulted in significantly longer average sentence lengths, and substantial assistance and downward departures resulted in significantly shorter average sentences in both types of cases.

The above findings notwithstanding, there are a few limitations with the current study that can be expanded upon by future research. First, although the USSC data have many advantages, including a large number of cases, and availability across multiple years, the composition of sex offenses likely differs in federal jurisdictions, as compared with state courts. Future research that reassess the questions posed in this study using data from state courts over time would be valuable. Second, the USSC data provide details on several important variables that capture both legally relevant and extra-legal factors that can influence sentencing decisions. Still, there are potentially important variables such as the type of attorney (i.e., private vs. public defender) an individual had (Hartley et al., 2010), the presence of physical evidence (Lee & Testa, 2020; Nir & Griffiths, 2018; Spears & Spohn, 1997), and details of the victim characteristics (Spohn & Spears, 1996) that could influence sentence decisions, and the commission could assist in a better understanding of the determinants of sentences by adding these variables to its current data sets. Third, the current study focused on decisions at the final sentencing stages (i.e., sentence length and guideline departure). However, future research can also assess how discretionary decisions for earlier case processing decisions have changed across time in federal sex crime cases, such as whether an individual was detained prior to trial without bail (Lee, 2019), whether an individual received a charge reduction from a prosecutor (Johnson & Larroulet, 2019; Shermer & Johnson, 2010), and the prosecutors’ decisions to change the charge from the arresting offense (Hartley & Tillyer, 2018). Finally, the aim of this study was the assess changes in sentencing for federal sex crimes over time across U.S. District courts. In doing so, we included a series of district-level fixed effects to control for interdistrict variation. However, future research should expand upon this work and assess whether the sentencing of federal sex crimes varies between court jurisdictions, and which factors may explain this variation in sentencing practices (Ulmer & Johnson, 2004).

We end with a note about the implications of this study for federal sentencing policy and future research. Our findings reveal that sentence lengths for individuals convicted of both sexual abuse and child pornography offenses increased substantially over the multiple years of data analyzed in the current study. Our results also demonstrate a shift in the racial and ethnic makeup of the individuals convicted of these offenses. Consequently, our analyses demonstrate that the shifting nature of the racial and ethnic composition of those convicted of federal sex offenses has corresponded with shifts in punishment that has increased racial and ethnic disparity in federal sentencing among sex offense cases. Previous research has found that Congressionally enacted enhancements to punishment (i.e., mandatory minimum statutes) attached to certain types of offenses (crack-cocaine) resulted in unwarranted racial disparity being reintroduced into federal sentencing (Hartley et al., 2007). As such, lawmakers seeking to enhance penalties for certain types of offenses, even where those offenses comprise a small share of the overall criminal caseload, would be wise to examine the disparate impact the increased penalties might have with regard to the extra-legal characteristics of those prosecuted for these offenses. Our findings show that there has been an increase in the proportion of Blacks convicted for sexual abuse offenses over time, and these convictions have resulted in longer sentence lengths for Blacks convicted of sexual abuse during a period where racial disparities in sentencing have been declining generally (King & Light, 2019). Congress recently passed the First Step Act in 2018 aimed at sentencing reform, specifically the reduction of mandatory minimum sentences for certain offenses, expansion of the safety valve provision, and improvement of programming within the federal Bureau of Prisons allowing for early release (through increased good time credit) of inmates who participate in these programs. In line with reforms seeking to reduce the federal prison population, Congress should further pass legislation specifically targeting individuals convicted of federal sex offenses to offer and incentivize, through increased good time credit, participation in treatment-based prison programming, which would assist in successful reentry and reduced recidivism.

Our analyses also reveal that a substantial number of those convicted of sexual abuse and child pornography offenses received judicial downward or prosecutorial-initiated (substantial assistance) departures; roughly one third and two thirds of those convicted, respectively. These findings demonstrate that individuals convicted of both types of offenses have increased likelihoods for a departure over time and, in line with previous research, in some instances, race has a significant effect in these decisions (Doerner & Demuth, 2010; Johnson et al., 2008; Ward et al., 2016). As a result, dependent on their race, individuals convicted of the same federal sex offenses are more or less likely to receive a sentence reduction via departure and arrive at different final sentences.

These results suggest that judges may be increasingly exercising their discretionary power to depart from guideline sentences based on their disagreements with sentencing policy and/or the presumptive sentences for those convicted of federal sex offenses (Kimbrough v. United States, 552 U.S. 85, 101, 2007). Indeed, some federal judges have expressed disagreement with the prescribed guideline sentences for those convicted of child pornography offenses and they depart more often in these cases (Kaiser & Spohn, 2014), but the results of this study indicate that they are perhaps increasingly dissatisfied with guideline minimum sentence lengths for those convicted of sexual abuse offenses as well. Alternatively, more serious child pornography cases might have been prosecuted in recent years and, therefore, judges have not been as likely to depart in those cases.¹⁶ A useful direction for future research would be additional qualitative work with judges that could extricate the exact mechanisms that underlie these patterns.

If judges continually depart from the guidelines citing policy disagreements with federal sex offense statutes and presumptive sentence lengths, the USSC, who reports to, and advises, Congress on federal sentencing practice and policy, should continue to monitor these departures and report back to Congress to recommend adjustments to federal sex offense statutes to promote fair and just punishment metrics. As departures greatly influence what the final sentence length will ultimately be, it is important to examine their use to ensure uniformity in treatment of those convicted of federal sex offenses across race/ethnicity. Congress should also endeavor to monitor how legislative increases in punishments for certain offenses affect law enforcement arrest, and U.S. Attorney charging, and plea bargaining practices, to ensure fairness in application of federal law.

Congress enacted the federal sentencing guidelines to realize more uniform and less disparate sentencing outcomes (Frase, 2019). The federal prison population since guideline implementation, however, has become disproportionately made up of racial and ethnic minorities. Recent USSC statistics show that just more than 70% of those serving time in federal prison are racial or ethnic minorities (USSC, 2020). Our results therefore are also in line with the limited extant research revealing that the changing racial/ethnic composition of individuals convicted of federal offenses has contributed to racial/ethnic sentencing disparities over time (Light, 2014). Even so, very little focus has been given to the temporal context of sentencing decisions, and the analytic approach taken here uncovered how racial/ethnic variation in sentence length and departure determinations changed substantially across time. Future research, therefore, should attempt to account for these contextual variations in analytic models of punishment outcomes when possible. Likewise, it would be beneficial for future work to both assess the reasons behind the dramatic changes in the racial/ethnic composition for those charged and convicted of certain crimes such as sexual abuse, and how increases in the proportion of defendants who are racial/ethnic minorities influence perceptions of prosecutors and judges.

Finally, future studies should examine these two federal sex offenses disaggregated further by the various offense types within sexual abuse (e.g., sex trafficking of a minor, child pornography production) and child pornography (e.g., possession, receipt, and distribution). As varying mandatory minima lengths apply differentially based on the statute under which the offense is being prosecuted, it would be beneficial to examine whether, for example, average sentence lengths vary significantly even within these two broad offense categories, or whether departures are disproportionately given for one type of offense (e.g., possession of child pornography). Aggregate analyses can sometimes mask variation within categories of offense type. This is especially important if there are also differences in the demographic composition of those convicted at these more refined operationalizations.

Footnotes

Authors’ Note

The authors take full responsibility for the integrity of the data, the accuracy of the analyses, and have made every effort to avoid inflating statistically significant results.

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.

ORCID iD

Richard D. Hartley

Notes

References

Adams

Flynn

(2017). Federal prosecution of commercial sexual exploitation of children cases, 2004–2013 (NCJ Publication No. 250746). Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/fpcsecc0413.pdf

Albonetti

C. A.

(1987). Prosecutorial discretion: The effects of uncertainty. Law & Society Review, 21(2), 291–313. https://doi.org/10.2307/3053523

Albonetti

C. A.

(1991). An integration of theories to explain judicial discretion. Social Problems, 38(2), 247–266. https://doi.org/10.2307/800532

Babchishin

K. M.

Hanson

R. K.

VanZuylen

(2015). Online child pornography offenders are different: A meta-analysis of the characteristics of online and offline sex offenders against children. Archives of Sexual Behavior, 44(1), 45–66.

Baumer

(2013). Reassessing and redirecting research on race and sentencing. Justice Quarterly, 30, 231–261. https://doi.org/10.1080/07418825.2012.682602

Bontrager

Barrick

Stupi

(2013). Gender and sentencing: A meta-analysis of contemporary research. Journal of Gender Race & Justice, 16, 349.

Doerner

J. K.

Demuth

S. S.

(2010). The independent and joint effects of race/ethnicity, gender, and age on sentencing outcomes in US federal courts. Justice Quarterly, 27, 1–27. https://doi.org/10.1080/07418820902926197

Droske

T. J.

(2008). Correcting Native American sentencing disparity post-Booker. Marquette Law Review, 91, 723–813.

Feldmeyer

Ulmer

J. T.

(2011). Racial/ethnic threat and federal sentencing. Journal of Research in Crime and Delinquency, 48(2), 238–270.

10.

Franklin

T. W.

(2013). Sentencing Native Americans in US federal courts: An examination of disparity. Justice Quarterly, 30(2), 310–339. https://doi.org/10.1080/07418825.2011.605072

11.

Franklin

T. W.

Henry

T. K. S.

(2020). Racial disparities in federal sentencing outcomes: Clarifying the role of criminal history. Crime & Delinquency, 66(1), 3–32.

12.

Frase

R. S.

(2019). Forty years of American sentencing guidelines: What have we learned? Crime and Justice, 48(1), 79–135. https://doi.org/10.1086/701503

13.

Hamilton

(2011). The efficacy of severe child pornography sentencing: Empirical validity or political rhetoric? Stanford Law & Policy Review, 22, 545–586.

14.

Hartley

R. D.

Madden

Spohn

(2007). Prosecutorial discretion: An examination of substantial assistance departures in federal crack-cocaine and powder-cocaine cases. Justice Quarterly, 24, 383–407. https://doi.org/10.1080/07418820701485379

15.

Hartley

R. D.

Miller

H. V.

Spohn

C. C.

(2010). Do you get what you pay for? Attorney type and its effects of criminal court outcomes. Journal of Criminal Justice, 38, 1063–1070. https://doi.org/10.1016/j.jcrimjus.2010.07.009

16.

Hartley

R. D.

Tillyer

(2012). Defending the homeland: Judicial sentencing practices for federal immigration offenses. Justice Quarterly, 29, 76–104.

17.

Hartley

R. D.

Tillyer

(2018). Examining prosecutorial discretion in federal criminal cases: Legal and extra-legal determinants of declination and charge change decisions. Justice Quarterly, 35(7), 1195–1225. https://doi.org/10.1080/07418825.2018.1530367

18.

Hessick

C. B.

(2011). Disentangling child pornography from child sexual abuse. Washington University Law Review, 88, 853–902.

19.

Hofer

P. J.

(2007). United States v. Booker as a natural experiment: Using empirical research to inform the federal sentencing policy debate. Criminology & Public Policy, 6(3), 433–460.

20.

Huebner

B. M.

Bynum

T. S.

(2006). An analysis of parole decision making using a sample of sex offenders: A focal concerns perspective. Criminology, 44(4), 961–991. https://doi.org/10.1111/j.1745-9125.2006.00069.x

21.

Johnson

B. D.

Betsinger

(2009). Punishing the “model minority”: Asian-American criminal sentencing outcomes in federal district courts. Criminology, 47(4), 1045–1090. https://doi.org/10.1111/j.1745-9125.2009.00169.x

22.

Johnson

B. D.

Larroulet

(2019). The “distance traveled”: Investigating the downstream consequences of charge reductions for disparities in incarceration. Justice Quarterly, 36(7), 1229–1257. https://doi.org/10.1080/07418825.2018.1529250

23.

Johnson

B. D.

Ulmer

J. T.

Kramer

J. H.

(2008). The social context of guideline circumvention: The case of federal district courts. Criminology, 46, 711–783. https://doi.org/10.1111/j.1745-9125.2008.00125.x

24.

Kaiser

Spohn

(2014). “Fundamentally flawed?” Exploring the use of policy disagreements in judicial downward departures for child pornography offenses. Criminology & Public Policy, 13, 241–270. https://doi.org/10.1111/1745-9133.12082

25.

King

R. D.

(2019). Cumulative impact: Why prison sentences have increased. Criminology, 57(1), 157–180. https://doi.org/10.1111/1745-9125.12197

26.

King

R. D.

Light

M. T.

(2019). Have racial and ethnic disparities in sentencing declined? Crime and Justice, 48, 365–437. https://doi.org/10.1086/701505

27.

Lee

J. G.

(2019). To detain or not to detain? Using propensity scores to examine the relationship between pretrial detention and conviction. Criminal Justice Policy Review, 30(1), 128–152. https://doi.org/10.1177/0887403416668016

28.

Lee

J. G.

Testa

(2020). Evidence, arrest circumstances, and felony cocaine case processing. Criminology, Criminal Justice, Law & Society, 21, 1.

29.

Levenson

Brannon

Fortney

Baker

(2007). Public perceptions about sex offenders and community protection policies. Analyses of Social Issues and Public Policy, 7(1), 1–25. https://doi.org/10.1111/j.1530-2415.2007.00119.x

30.

Light

M. T.

(2014). The new face of legal inequality: Noncitizens and the long-term trends in sentencing disparities across US district courts, 1992–2009. Law & Society Review, 48(2), 447–478.

31.

Light

M. T.

Massoglia

King

R. D.

(2014). Citizenship and punishment: The salience of national membership in US criminal courts. American Sociological Review, 79(5), 825–847. https://doi.org/10.1177/0003122414543659

32.

Lynch

(2019). Focally concerned about focal concerns: A conceptual and methodological critique of sentencing disparities research. Justice Quarterly, 36, 1148–1175. https://doi.org/10.1080/07418825.2019.1686163

33.

Lynch

Omori

(2018). Crack as proxy: Aggressive federal drug prosecutions and the production of Black–White racial inequality. Law & Society Review, 52(3), 773–809.

34.

Maddan

Hartley

R. D.

(2018). Towards the development of a standardized focal concerns theory of sentencing. In J. T.

Ulmer

M. S.

Bradley

(Eds.), Punishment decisions: Locations of disparity. ASC division on corrections and sentencing handbook (Vol. 2, pp. 311–335). Routledge.

35.

Mitchell

(2005). A meta-analysis of race and sentencing research: Explaining the inconsistencies. Journal of Quantitative Criminology, 21(4), 439–466.

36.

Motivans

Kyckelhahn

(2007). Federal prosecution of child sex exploitation offenders, 2006. United States Department of Justice. Bureau of Justice Statistics. https://www.bjs.gov/content/pub/pdf/fpcseo06.pdf

37.

Mustard

D. B.

(2001). Racial, ethnic, and gender disparities in sentencing: Evidence from the US federal courts. The Journal of Law and Economics, 44(1), 285–314.

38.

Nir

Griffiths

(2018). Sentencing on the evidence. Criminal Justice Policy Review, 29(4), 365–390. https://doi.org/10.1177%2F0887403416635248

39.

Pickett

J. T.

Mancini

Mears

D. P.

(2013). Vulnerable victims, monstrous offenders, and unmanageable risk: Explaining public opinion on the social control of sex crime. Criminology, 51(3), 729–759. https://doi.org/10.1111/1745-9125.12018

40.

Rachlinski

J. J.

Johnson

S. L.

Wistrich

A. J.

Guthrie

(2009). Does unconscious racial bias affect trial judges. Notre Dame Law Review, 84, 1195–1246.

41.

Sample

L. L.

(2001). The social construction of the sex offender (Unpublished dissertation). University of Missouri, St. Louis, MI.

42.

Shermer

L. O. N.

Johnson

B. D.

(2010). Criminal prosecutions: Examining prosecutorial discretion and charge reductions in US federal district courts. Justice Quarterly, 27(3), 394–430. https://doi.org/10.1080/07418820902856972

43.

Spears

J. W.

Spohn

C. C.

(1997). The effect of evidence factors and victim characteristics on prosecutors’ charging decisions in sexual assault cases. Justice Quarterly, 14(3), 501–524. https://doi.org/10.1080/07418829700093451

44.

Spohn

(2017). Race and sentencing disparity. In: E.

Luna

(Ed.), Reforming criminal justice: A report of the academy for justice on bridging the gap between scholarship and reform (Vol. 4, pp. 169–186). Arizona State University.

45.

Spohn

Holleran

(2000). The imprisonment penalty paid by young, unemployed Black and Hispanic male offenders. Criminology, 38(1), 281–306.

46.

Spohn

Spears

(1996). The effect of offender and victim characteristics on sexual assault case processing decisions. Justice Quarterly, 13(4), 649–679. https://doi.org/10.1080/07418829600093141

47.

Steffensmeier

Demuth

(2000). Ethnicity and sentencing outcomes in US federal courts: Who is punished more harshly? American Sociological Review, 65(5), 705–729.

48.

Steffensmeier

Painter-Davis

Ulmer

(2017). Intersectionality of race, ethnicity, gender, and age on criminal punishment. Sociological Perspectives, 60(4), 810–833.

49.

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(4), 763–798. https://doi.org/10.1111/j.1745-9125.1998.tb01265.x

50.

Testa

Lee

J. G.

(2020). Trends in sentencing of federal drug offenders: Findings from US district courts 2002–2017. Journal of Drug Issues. Advance online publication. https://doi.org/0022042620959071

51.

Tewksbury

(2014). Evidence of ineffectiveness: Advancing the argument against sex offender residence restrictions. Criminology & Public Policy, 13, 135.

52.

Ulmer

Painter-Davis

Tinik

(2016). Disproportional imprisonment of Black and Hispanic males: Sentencing discretion, processing outcomes, and policy structures. Justice Quarterly, 33(4), 642–681.

53.

Ulmer

J. T.

(2012). Recent developments and new directions in sentencing research. Justice Quarterly, 29, 1–40. https://doi.org/10.1080/07418825.2011.624115

54.

Ulmer

J. T.

Bradley

M. S.

(2019). Criminal justice in Indian country: A theoretical and empirical agenda. Annual Review of Criminology, 2, 337–357. https://doi.org/10.1146/annurev-criminol-011518-024805

55.

Ulmer

J. T.

Johnson

(2004). Sentencing in context: A multilevel analysis. Criminology, 42(1), 137–178. https://doi.org/10.1146/annurev-criminol-011518-024805

56.

Ulmer

J. T.

Light

M. T.

Kramer

J. H.

(2011). Racial disparity in the wake of the Booker/Fanfan decision: An alternative analysis to the USSC’s 2010 report. Criminology & Public Policy, 10, 1077–1118.

57.

United States v. Booker. (2005). Supreme Court of the United States 543 U.S. 220.

58.

United States Department of Justice. (2017). Indian country investigations and prosecutions.

59.

United States Sentencing Commission. (2009). History of the child pornography guidelines

60.

United States Sentencing Commission. (2012). Report to congress: Federal child pornography offenses.

61.

United States Sentencing Commission. (2019). Mandatory minimum penalties for sex offenses in the federal criminal justice system.

62.

United States Sentencing Commission. (2020). Quick facts federal offenders in prison June 2020.

63.

Ward

J. T.

Hartley

R. D.

Tillyer

(2016). Unpacking gender and race/ethnicity biases in the federal sentencing of drug offenders: A casual mediation approach. Journal of Criminal Justice, 46, 196–206. https://doi.org/10.1016/j.jcrimjus.2016.05.008