Weighing the Value of the Bargain: Prosecutorial Discretion After Sentencing Guidelines

Abstract

There is little empirical research to indicate whether the introduction of sentencing guidelines displaces discretion from judges to prosecutors. In the handful of studies that examine the hydraulic displacement of discretion, discretion is usually measured by the rate of charge bargaining. The current study uses an alternative methodology—calculating the value of the bargain—to examine the effect of sentencing guidelines on prosecutorial discretion in the District of Columbia Superior Court. It measures the impact of charge bargaining on sentence length in a sample of 266 pre-guidelines sentences and 263 post-guidelines sentences, and finds that the rate of charge bargaining did not change after the introduction of guidelines, but that the impact of bargaining on sentence length increased slightly. Although the amount of displacement of discretion in the D.C. Guidelines was modest, the study demonstrates that alternative measures (value of the bargain) might reveal displacement of discretion when traditional measures (rate of the bargain) do not.

Keywords

prosecutorial discretion sentencing guidelines charge bargaining

Introduction

Historically, U.S. judges possessed extraordinary discretion in sentencing (Stith & Cabranes, 1998). They were generally free to impose any sentence authorized by law, and their sentencing decisions were virtually unreviewable. In 1972, the U.S. Supreme Court held that “a sentence imposed by a federal district judge, if within statutory limits, is generally not subject to review” (United States v. Tucker, 1972, p. 447). In addition, because each judge functionally operated as a law unto himself or herself, identical defendants—appearing before different judges—could receive wildly different punishments. Thus, a defendant in one courtroom might be sentenced to a lengthy prison term; whereas in the courtroom next door, an identical defendant, convicted of the same crime, might receive only probation (King & Klein, 2004).

Judge Marvin E. Frankel (1972) famously criticized this kind of sentencing disparity as “judicial lawlessness” (p. 1), and he called for sentencing guidelines that could restrict and channel judicial authority (Frankel, 1973). Sentencing guidelines seek to increase uniformity and reduce disparity by basing sentences upon the characteristics of the offense and the offender (Luna, 2005; Ostrom, Ostrom, Hanson, & Kleiman, 2008; Spohn, 2002). Guidelines sentencing became enormously popular in the United States during the late 1970s (Pillsbury, 1989) as many states adopted guidelines to produce greater uniformity in sentencing.

Many researchers have concluded that sentencing guidelines do reduce disparity (Frase, 2000; Gorton & Boies, 1999; Hartley, Maddan, & Walker, 2006; Hunt & Connelly, 2005; Kramer & Lubitz, 1985; Kramer & Ulmer, 2009; Miethe & Moore, 1985; Minnesota Sentencing Guidelines Commission, 1984; Pennsylvania Commission on Sentencing, 1984; Stolzenburg & D’Alessio, 1994; Washington State Sentencing Guidelines Commission, 1992). However, some critics have argued that criminal justice reforms have merely displaced disparities away from the judge at the sentencing phase, to the prosecutor at the stages of charging and plea bargaining (Alschuler, 1978; Coffee & Tonry, 1983; Engen, 2008; Lagoy, Hussey, & Kramer, 1979; Nagel & Schulhofer, 1992; Savelsberg, 1992; Tonry & Coffee, 1987). This is commonly referred to as the “hydraulic displacement of discretion” thesis:

An often-invoked simile likens the discretion-ridden criminal justice system to a set of hydraulic brakes. If you push down on one point, the displaced volume of fluid will exert pressure and “bulge out,” reappearing elsewhere in the mechanism. Similarly, discretion in the criminal justice system can never be extinguished; it is simply dislodged and shifted to other system parts. (McCoy, 1984, p. 256)

Prior Research on Prosecutorial Charging and Charge Bargaining Decisions

When compared with other decisions within the criminal justice system, such as the police decision to arrest, the judicial decision to grant pretrial release, or the judicial sentencing decision, the prosecutor’s decision to charge a defendant with a crime is underresearched (Albonetti, 1997). Although the Bureau of Justice Statistics previously published data from the Prosecutor’s Management Information System (PROMIS)—which aggregated information from Manhattan; Los Angeles; Washington, D.C.; and other U.S. jurisdictions—this was discontinued in 1992. According to Spears and Spohn (1997), previous research suggests that initial charging decisions are based primarily on legal factors, such as the seriousness of the offense, the strength of evidence in the case, and the defendant’s culpability. More recently, Spohn, Beichner, and Davis-Frenzel (2001) examined the prosecutor’s decision of whether or not to charge in sexual assault cases. They find that charging decisions principally reflect the prosecutor’s assessment of the likelihood of conviction. This assessment, in turn, appears to be based upon typologies of rape and rape victims, the victim’s willingness to participate in preliminary interviews, the victim’s willingness to cooperate with prosecution efforts, and any admission that claims were fabricated.

As with the initial decision of whether to charge, McDonald (1979) noted that the prosecutor considers numerous factors when determining whether to downgrade or dismiss charges in exchange for a guilty plea, including the seriousness of the offense, the culpability of the offender, and the strength of the case. Unsurprisingly, the literature suggests that prosecutors are most likely to pursue cases involving serious crimes, with stronger evidence, and more culpable defendants (Shermer & Johnson, 2010).

McDonald (1979) also noted that attributes of the defendant and victim can play a role in the decision of whether to charge bargain, including age, sex, race, marital status, social class, political or family connections, demeanor, history of employment, drug use, alcohol use, psychiatric problems, physical health problems, military service, and length of local residence. Some studies indicate that extralegal offender and victim characteristics, such as race or gender, influence prosecutorial decision making, whereas other studies fail to find such an effect (Shermer & Johnson, 2010). Studying the relationship between offender characteristics and the prosecutor’s decision to reduce the number of charges, Ball (2006) did not find a statistically significant relationship between offender characteristics, such as race and ethnicity, sex, age, and employment status, and the likelihood of receiving a count reduction.

Practical considerations might also influence whether a prosecutor will engage in charge bargaining. For instance, the willingness of victims and other witnesses to testify at trial and pretrial conferences can affect the evaluation of a case (McDonald, 1979). Another key consideration is the prosecutor’s caseload. Most prosecutors extend plea offers in the majority of their cases because they lack the time and resources to go to trial in all of them (Davis, 2007). The prosecutor’s relationship with the defense attorney and the attorney’s reputation for honesty, willingness to go to trial, and competency at trial are also contributing factors (McDonald, 1979). Although the existing research on prosecutorial decision making provides some important insights, “much of it is dated, has been constrained to small sample sizes, [is] limited to particular offenses (e.g., burglary, robbery, or sexual assault), or [is] conducted in specific locales” (Shermer & Johnson, 2010, p. 400).

Prior Research on the Hydraulic Displacement of Discretion Thesis

One theoretical perspective that may shed some light on the displacement of discretion thesis is the court communities perspective. As Eisenstein and his colleagues have noted, sentencing reforms do not operate in a vacuum; they “take place in the real world of criminal court communities” (Eisenstein, Flemming, & Nardulli, 1988, p. 296). Thus, what might initially appear to be a blunt transfer of discretion from judges to prosecutors might actually conceal a collective effort to maintain individualized sentencing in the face of judicial reforms (Tonry & Coffee, 1987). In 1958, Ohlin and Remington noted that when structural changes are imposed, courtroom actors respond to maintain stability in the workplace by engaging in accommodative responses.

Thus, prosecutors may seek to preserve individualized sentencing post-guidelines through their control of charging and charge bargaining decisions (Wooldredge & Griffin, 2005). Reitz (1998) described an example in which the prosecutor charges a lesser offense because the probable legal penalty is too great for a particular case, or in which the parties plea bargain to avoid the full force of the penalty. Prosecutors’ decisions over plea bargaining confer enormous power. McCoy (1984) noted that the process of plea negotiations “produces nearly binding recommendations for the sentencing judge; in effect, [plea bargaining] may virtually become sentencing” (p. 258). Yet, the reallocation of discretion does not necessarily produce conflicts within the courtroom community. Judges, too, “may attempt to maneuver around guidelines by being more tolerant of certain plea agreements” (Wooldredge & Griffin, 2005, p. 302). And, Frase (1999) explained that in a “properly balanced guidelines system” with “reasonable sentence severity levels,” in which courts retain substantial sentencing discretion due to broad guidelines ranges and/or flexible departure powers, it is “rare that prosecutorial decisions will produce sentences which judges strongly disapprove, yet are powerless to prevent” (p. 69; see also Miethe, 1987). McCoy (1984) found that a determinate sentencing reform left intact those courtroom norms under which guilty pleas were approved and, in fact, “probably made the plea negotiation process even smoother . . . because defendants knew exactly what sentence they would receive after pleading guilty” (p. 265; see also Tonry & Coffee, 1987). McCoy’s findings suggest “that many defendants were able to negotiate so that they received basically the same sentences under the new law as they would have under the old” (p. 265). Engen’s (2008) review of research on the hydraulic displacement thesis explained that prosecutors and other court actors circumvent sentencing reforms through mechanisms such as charge reductions because the laws conflict with local norms or “going rates” for punishment, because court actors perceive the laws to be unjust in individual cases, or because reducing the charges gives prosecutors a powerful means of enticing (or coercing) defendants to plead guilty.

Another perspective used to examine prosecutorial decision making is focal concerns theory, which was first proposed by Steffensmeier, Ulmer, and Kramer (1988) to frame hypotheses regarding the effects of race, gender, and age on judicial sentencing decisions. They posit that three focal concerns influence judges and other criminal justice actors in reaching sentencing decisions. The three focal concerns are the offender’s blameworthiness and the degree of harm caused to the victim, protection of the community, and practical implications of sentencing decisions.

According to Steffensmeier et al. (1988), the first focal concern, blameworthiness, is ordinarily associated with the just deserts philosophy of punishment, where the severity of the sentence increases with the culpability of the defendant and the harm caused by the offense. The second factor, protection of the community, typically focuses on incapacitation and deterrence. Under this perspective, predictions about offender dangerousness (i.e., the risk of recidivism) are based on attributions predicated on factors, such as the nature of the offense (e.g., violent or property); the offender’s criminal history; the facts of the crime, such as use of a weapon; and the characteristics of the offender, such as drug dependency, education, employment, or family history.

The third focal concern, practical constraints and consequences, consists of both organizational and individual facets. Organizational concerns include maintaining working relationships among courtroom actors, ensuring the stable flow of cases, and being sensitive to local and state correctional crowding and resources. Practical consequences at the individual level include concerns about the offender’s ability to do time, health condition, special needs, and the disruption of ties to children and other family members. Because the three focal concerns and their interplay are highly complex, and because judges rarely have sufficient information about the case or the defendant, they develop a “perceptual shorthand” to make determinations such as who is dangerous and who is not. This shorthand is linked to race, gender, and age attributions (Steffensmeier et al., 1988, p. 767).

Some scholars have applied the focal concerns perspective to prosecutorial decision making. Kramer and Ulmer (2009), for instance, discussed the focal concerns of prosecutors within the context of their application of mandatory minimums:

[P]rosecutors frequently de-mandatorize cases in exchange for plea agreements, in that negotiated plea agreements tend to protect offenders from mandatory minimums. In terms of focal concerns, plea agreements and trial convictions may signal a variety of different things about defendant blameworthiness and dangerousness or practical constraints to prosecutors. On the one hand, working with the prosecution to reach a plea agreement . . . might indicate remorse . . ., which might lower perceived moral blameworthiness. In addition, . . . pleading guilty and showing remorse are often seen as a first step toward rehabilitation . . . In addition, plea agreements and trials entail a variety of practical constraints and consequences . . . Rewarding eligible defendants for pleading guilty by foregoing the imposition of mandatory sentences encourages guilty pleas and moves cases through the system . . . plea agreements allow prosecutors to avoid the uncertainties of trials to obtain convictions they might not otherwise obtain. (pp. 185-186)

In addition, Spohn and Holleran (2001) used the focal concerns perspective to analyze prosecutor charging decisions. They note that the focal concerns that guide charging decisions are similar, but not identical to those guiding judicial sentencing decisions. Although prosecutors consider the seriousness of the offense, the degree of harm to the victim, and the culpability of the suspect when filing charges, they take into account a different set of practical constraints and consequences than judges. Although prosecutors are also concerned about maintaining relationships with other members of the courtroom work group, they focus on the likelihood of conviction rather than the social costs of punishment. They must predict how the victim, the suspect, and the incident will be viewed and evaluated by the judge and jurors at later phases of the criminal justice process (Spohn & Holleran, 2001). Because these predictions are uncertain, prosecutors also develop a perceptual shorthand that incorporates stereotypes of real crimes and credible victims. Consequently, they consider not only the legally relevant factors but also the background, character, and behavior of the victim; the relationship between the suspect and the victim; and the willingness of the victim to cooperate as the case moves forward (Spohn & Holleran, 2001).

Finally, Hartley, Maddan, and Spohn (2007) attempted to identify the factors that affect prosecutorial decision making regarding substantial assistance departures for crack cocaine and powder cocaine defendants prosecuted in federal courts.¹ Among other things, they found that prosecutors were less likely to file substantial assistance departure motions for offenders charged with offenses involving crack cocaine (rather than powder cocaine), and they were more likely to file motions for offenders facing mandatory minimum sentences. Offenders charged with more serious crimes and with more substantial criminal histories had a greater likelihood of receiving departures. Female offenders, White offenders, and more educated offenders also had a greater likelihood of receiving these types of departures.

Although the authors could not state with certainty what motivated prosecutors to file motions for substantial assistance, they write that it appears that their decisions are guided by two focal concerns: a desire to secure additional prosecutions and convictions of drug offenders and a belief that the sentences imposed on some types of drug offenders are unreasonably harsh. They conclude, “Our findings suggest that prosecutors use these motions, which are highly discretionary and generally unreviewable, to reduce the sentences of offenders perceived to be sympathetic and non-dangerous” (Hartley et al., 2007, p. 391).

Although the court community and focal concerns perspectives shed important light on the hydraulic displacement thesis, there are unanswered empirical questions regarding whether sentencing reforms actually shift discretion from judges to prosecutors. The paucity of empirical research is due largely to the fact that prosecutors do not typically report information about charging decisions (Forst, 2002). Engen’s (2008) review of research provided a helpful description of the limited empirical record and unanswered questions related to the hydraulic displacement of discretion thesis. As Engen explained, although prosecutors may adapt their charging practices in response to new laws or presumptive sentencing guideline systems that restrict judicial discretion, it is not clear how this type of circumvention happens and in what kinds of cases; whether these reforms have actually shifted control over sentencing to prosecutors, and to what degree; and whether prosecutorial adaptation has undermined the goals of these reforms such as increasing uniformity and reducing unwarranted disparities.

Even if prosecutors often bargain around sentencing guidelines, this is not enough, Engen explains, to demonstrate that they usurp sentencing authority from judges. If, for example, prosecutors do not plea bargain in certain types of cases, judges may dismiss certain charges, reduce sentences for underlying offenses to offset enhancements in others, or find defendants guilty of misdemeanors rather than felonies. Although prosecutors controlled the charging decision in these cases, they did not control sentencing. The critical test for the displacement hypothesis, according to Engen, is whether the relative control of prosecutors over sentencing increases under new sentencing laws relative to that of judges.

It is not clear to what extent mandatory or presumptive sentencing laws have shifted control over sentencing to prosecutors, nor is it clear whether punishment is more or less uniform under these laws than it would be otherwise. The evidence simply is incomplete. In the past 20 years, very few studies have examined charging decisions regarding mandatory minimums with individual-level data, and even fewer have examined charging decisions in states with presumptive sentencing guidelines. Most importantly, very few studies have estimated the effect of charging decisions in these cases on either the likelihood of imprisonment or the length of sentences ordered, net of other factors. (Engen, 2008, p. 85)

Although the empirical record on questions related to the hydraulic displacement of discretion is limited, there are a handful of studies examining charging and plea bargaining practices before and after sentencing reforms that limit judicial discretion. In 1987, Miethe analyzed differences in prosecutorial decision making before and after the 1980 implementation of sentencing guidelines in Minnesota. He predicted that rates of charge bargaining would increase after guidelines implementation (because prosecutors would have more power to encourage defendants to plead guilty). He analyzed the severity of initial charges and the rates of charge dismissals, charge reductions, sentence negotiations, and plea bargains, as well as examining extralegal characteristics such as sex, race, and employment status. Miethe, however, found few differences in the rates of charge bargaining and sentencing negotiation. He also found little evidence that extralegal factors influenced plea bargaining. Rather, any differences in prosecutorial decision making appeared to result from an increased emphasis on legal and offense-specific factors.

Empirical study of the hydraulic displacement thesis stalled for almost 20 years, until Wooldredge and Griffin (2005) conducted a similar analysis in Ohio. Ohio’s guidelines, implemented in 1996, were less restrictive than Minnesota’s were. Given the presumption that prosecutorial discretion would be greater in a more restrictive sentencing system, Wooldredge and Griffin predicted that they would also find little evidence of hydraulic displacement. Yet, they reported a modest increase in the rate of charge reductions among defendants who pled guilty, suggesting that the reform might have influenced prosecutorial discretion, at least in terms of charge bargaining. However, like Miethe (1987), Wooldredge and Griffin concluded that increased charge reductions did not result in increased extralegal disparity.

In 2014, Vance and Oleson examined the hydraulic displacement thesis using data from the District of Columbia Superior Court, which implemented voluntary sentencing guidelines in 2004. They examined the effects of guidelines on charge bargaining outcomes (measured by reductions in charges, charges dropped, and reductions in guidelines severity measures [between original charge and the charge of conviction]). They did not identify any statistically significant effects of the guidelines on prosecutorial discretion. Therefore, to better examine hydraulic displacement, it might be useful to examine the factors that influence the exercise of prosecutorial discretion, particularly in the decisions of whether to charge and whether to engage in charge bargaining.

The Value of the Bargain

Although the studies listed above have measured prosecutorial discretion by asking whether the rate of charge bargaining increased post-guidelines, another approach, examining the value of the bargain rather than the rate of the bargain, might provide a different estimate of displaced discretion, and thereby prove to be valuable to policy makers. Rather than measuring changes in the rate of charge bargaining, Piehl and Bushway (2007) recommended measuring the difference (in months) that a sentence is reduced through charge bargaining.

[I]n a world in which 90% of the convictions end as a result of a plea bargain, we believe the substantive value of interest is the difference between what the person would have received if they had not pled and what they received as the result of the plea bargain. (p. 107)

To weigh the value of a charge reduction, Piehl and Bushway (2007) asked what would have happened to the person if he or she did not plea bargain. Although this is unobservable directly,

we do have data on what other people who are convicted of these offenses receive at sentencing, and to the extent to which these people are similar to the person who plea bargain, we can use this information to create an estimate of our desired counterfactual. (p. 108)

They first estimate a model for the actual sentence length at plea using all available information about the case and criminal history, including the charge at conviction. Then, using those same estimates, they create a predicted sentence for these same individuals, in which the estimate is made using the charge at arraignment instead of the charge at conviction. If there is no charge bargaining, predicted and actual values will be the same. If, however, charge bargaining has occurred, predicted sentences (using charges at arraignment) should be higher than the actual values (using charges at conviction). The difference for each individual

will be an estimate of the value of the charge bargain in terms of the sentence avoided by pleading guilty. Or alternatively, the difference will be an estimate of the size of the discretion exercised by the prosecutor in assigning the charge. (p. 108)

Using a weight of the bargain approach, Piehl and Bushway compared two guidelines systems by using State Court Processing Statistics data: a voluntary system (in Maryland) and a presumptive system (in Washington). Piehl and Bushway found that although the rate of charge bargaining (specifically, the rate at which people pled down from a felony charge to a misdemeanor conviction) was higher in Maryland, its impact on sentences was actually greater in Washington. Unlike previous studies using the traditional “existence of the bargain” approach, the findings are consistent with the hypothesis that stricter guidelines lead to more displacement of discretion to the prosecutor (Reitz, 1998). Piehl and Bushway argue that their methodology reveals a superior view of the impact of sentencing guidelines upon prosecutorial discretion. In light of their claim, we applied the Piehl and Bushway (2007) methodology to data collected from the District of Columbia.

Method

Research Setting

In June 2004, the District of Columbia implemented voluntary sentencing guidelines to increase fairness and consistency in sentencing for felony cases in the D.C. Superior Court. Prior to that time, there was no mechanism to structure judicial sentencing discretion within wide statutory ranges. The D.C. Sentencing Commission recognized that there was “substantial unexplained variability in sentencing” and that “[b]asic fairness requires that similarly situated offenders should receive similar sentences for similar crimes” (District of Columbia Sentencing Commission, 2003, p. v). The D.C. guidelines employ a matrix where the intersection between the offense of conviction and the criminal history indicates the proposed sentence. The guidelines use one matrix for all nondrug convictions (in which nine offense severity groups comprise the “Master Grid,” which is displayed in the appendix) and one matrix for drug cases (in which three offense severity groups comprise the “Drug Grid”).

In December 2006, the Commission conducted a preliminary evaluation of the guidelines, and concluded that they had reduced the degree of unexplained variation in sentences: 87.9% of all sentences imposed fell within guideline ranges (District of Columbia Sentencing Commission, 2006). The Commission also recognized the possibility that judicial discretion could be displaced to prosecutors, noting that “sentencing guidelines by design limit the discretion and power of judges, and many believe that in doing so, guidelines transfer some of that discretion and power to prosecutors—giving them too much power” (District of Columbia Sentencing Commission, 2006, p. 76). To assess this possibility, the Commission granted the principal author research access to all relevant charging and sentencing data collected by the D.C. Superior Court with the exception of the names of the defendants and the sentencing judges. The D.C. Superior Court was selected as the jurisdiction to study because the data that were collected on prosecutorial charge bargaining within the D.C. Superior Court are rarely available to researchers.

Description of Data

The data were drawn from two sources: the Court Information System (CIS), a database that contains conviction and sentencing-related information, and the “CourtView” system, which tracks individual cases from charging to sentencing. In CourtView, authorized personnel can review individual cases to examine docket sheets, charging documents, pleadings, and other court documents. Although access to the CourtView system made it possible for the principal author to capture data about prosecutorial decision making, this involved painstaking manual data entry. Accordingly, just a sample of pre-guidelines and post-guidelines felony cases were examined.

It was not possible to ascertain from the data which of the 2004 cases were pre- or post-guidelines; therefore, all 2004 cases were excluded, and a random selection of cases sentenced in 2003 (pre-guidelines) and 2005 (post-guidelines) were collected for analysis. Because there are only three wide offense severity groups in the Drug Grid, there was virtually no movement in these charge categories from the initial charge to conviction and, because some distance traveled between the categories is necessary to test the Piehl and Bushway (2007) technique, drug cases were excluded. Also excluded, because of the complexity of manual coding, were the 13.6% of cases involving multiple convictions. Ultimately, then, 266 single-count, nondrug convictions were selected from 2003 (pre-guidelines), and 263 single-count, nondrug convictions were selected from 2005 (post-guidelines). These 529 felony convictions represented 23.5% of the 2,254 nondrug cases from calendar years 2003 and 2005.

The dependent variable is the reduction in sentence length (in months). The independent variables include demographic characteristics, criminal history characteristics, and case processing characteristics. The measured demographic variables are age, sex, and race. The measured criminal history variables are prior misdemeanor and prior felony arrests within the District of Columbia. Information on traffic-related matters was not collected. No prior criminal record information was available from jurisdictions other than the District of Columbia, although there is no reason to believe that the ability to measure out-of-state convictions changed after the guidelines. Therefore, a relative comparison between the pre- and post-guidelines samples should not be affected. The case processing variables included type of attorney (whether the defendant had a publicly funded private attorney appointed by the Court under the Criminal Justice Act [CJA], an attorney from the District of Columbia Public Defender Service [PDS], or a privately retained attorney) and whether the offender was charged by grand jury indictment. The government’s interest in saving time and avoiding trial can result in favorable bargains before indictment: Many D.C. Superior Court judges grant greater sentencing concessions for pleas entered before indictment.

Methodology

To build our empirical model, we begin with the traditional model for explaining sentence length using case characteristics and criminal history:

{sentence length}_{i} = α + β {criminal history}_{i} + γ {crime severity}_{i} + ε_{i} .

Following Piehl and Bushway (2007), our estimating equation includes several modifications to this general approach in Equation 1. First, we log the sentence length to take into account nonlinearities across the sentencing range. Second, to examine the effect of charge bargaining on sentence length, this study uses a sentencing variable, which is a continuous variable measuring months of imprisonment, with sentences to probation treated as zero months of imprisonment.

With these modifications, our estimating equation becomes the tobit² equation:

\begin{array}{l} Ln (sentence length) = α + β_{1} {prior felony charges}_{i} + \\ β_{2} {prior misdemeanor charges}_{i} + β_{3} {indicted}_{i} + \\ β_{4} {attorney type}_{i} + β_{5} age + β_{6} gender + β_{7} race + \\ β_{8} {offense severity group}_{i} + ε_{i} . \end{array}

We enter the Sentencing Guidelines’ offense severity groups in the Master Grid associated with the charge at conviction. Using the estimated coefficients from Equation 2, we then calculate a predicted sentence length. The most important variables collected for this study are the specific names of every charge initially filed by the prosecutor. This information is then used to create a new variable for the most severe offense severity group initially charged by the prosecutor and the offense severity group for the charge at conviction. Because we know the offense severity group for the charge at arraignment, we use this to create a second predicted value for each individual. In other words, we create an estimate of the expected sentence at arraignment, again using the estimated coefficients from Equation 2. For both the pre-guidelines period and the post-guidelines period, we then calculate the difference between each person’s expected latent sentence at arraignment and the predicted latent sentence. The actual sentence is censored at zero, whereas the latent sentence is the “real” values of the sentencing variable if it was not censored. In this case, the latent sentence is unknown, but it is estimated using the tobit regression model. Actual sentences imposed, reported in Table 1, are well above the predicted latent sentences, which makes sense, because the latent values will in every case be less than or equal the actual value of zero for the censored cases.³

Table 1.

Descriptive Statistics: Pre-Guidelines and Post-Guidelines Samples (N = 529).

	Pre-guidelines (n = 266)	Post-guidelines (n = 263)
Demographic variables
Age at sentencing	32.09 (10.03)	32.48 (10.78)
Male (0, 1)	88.72	84.41
African American (0, 1)	93.98	93.52
Criminal history variables
Prior D.C. felony arrests	3.07 (3.69)	2.71 (3.50)
Prior D.C. misdemeanor arrests	3.51 (5.10)	3.20 (4.21)
Case processing variables
Conviction by plea (0, 1)	96.24	92.78
Attorney type
CJA (1)	82.95*	69.44*
PDS (2)	12.40*	23.02*
Retained (3)	4.65*	7.54*
Indicted (0, 1)	94.66*	87.64*
Type of conviction offense
Arson	0	0.38
Assault	15.04	14.45
Burglary	2.26	4.56
Carjacking	0.38	0.38
Cruelty to children	0.38	0
Destruction of property/unlawful entry	0	0.76
Escape/Bail Reform Act violation	34.96	31.56
Homicide	3.01	4.94
Inchoate	6.40	6.84
Kidnapping	0.75	0.38
Robbery	5.26	3.42
Sex offense	2.63	2.67
Theft, fraud, forgery	1.13	1.52
Unlawful use of automobile	13.53	10.65
Weapon	13.16	16.35
Final sentence	26.91 (45.56)	23.16 (40.42)

Note. CJA = Criminal Justice Act; PDS = Public Defender Service.

p < .05.

Results

Table 1 provides descriptive statistics for the measures in this study. Statistical tests did not reveal significant differences in the demographic profile or prior criminal record of the offenders between the pre- and post-guidelines samples. The tests did reveal an unexplained difference in attorney type, with CJA panel attorneys becoming less common and PDS defenders becoming more common after guidelines implementation. Conversations with defense attorneys revealed that PDS attorneys handle a higher proportion of more serious offenses, but crime types did not change after guidelines implementation. There was also a significant decrease in the rate of grand jury indictments, possibly indicating that prosecutors and defense attorneys were able to negotiate plea agreements more quickly (agreeing to waive the indictment requirement) after guidelines implementation. According to interviews with criminal justice practitioners in the District, the guidelines made the plea negotiation process easier for both the prosecutor and defense counsel by increasing the predictability of the sentencing exposure for each charge. Mean sentence lengths for Master Grid offenses decreased from 26.9 months before the guidelines to 23.2 months post-guidelines. This decrease, however, was not statistically significant, t(514) = 0.993, p = .321.

Table 2 presents the results of the tobit regression of sentence length. For pre-guidelines sentences, prior felony charges from the District of Columbia and guidelines offense severity Groups 6 to 9 were significant variables. Post-guidelines, prior felony charges, sex, a privately retained attorney, and guidelines offense severity Groups 5 to 9 were significant variables. The overall variance explained by these models is modest: pseudo-R² scores of 6% and 12%, respectively.

Table 2.

Tobit Regressions of Ln(Sentence Length): Pre-Guidelines and Post-Guidelines.

	Pre-guidelines (n = 243)		Post-guidelines (n = 235)
	b ₁	(SE₁)	b ₂	(SE₂)
Constant	5.99*	1.84	5.59	1.50
Prior D.C. felony charges	0.08*	0.04	0.09*	0.03
Prior D.C. misdemeanor charges	0.01	0.03	0.03	0.03
Age	−0.00	0.01	0.00	0.01
Male	0.22	0.35	0.60*	0.29
African American	−0.66	0.48	−0.43	0.38
Indicted	0.24	0.51	0.12	0.30
PDS attorney	0.37	0.34	−0.12	0.25
Retained attorney	−1.00	0.55	−0.94*	0.39
Convicted Group 9	−4.04*	1.65	−4.80*	1.42
Convicted Group 8	−4.15*	1.65	−4.32*	1.43
Convicted Group 7	−4.19*	1.85	−3.22*	1.52
Convicted Group 6	−3.50*	1.67	−3.14*	1.45
Convicted Group 5	−2.16	1.72	−2.81*	1.51
Convicted Group 4	−1.82	1.90	−1.51	1.89
Convicted Group 3	−0.78	1.80	−0.86	1.50
Convicted Group 2	Omitted	Omitted	Omitted	Omitted
Pseudo R²	.06		.12

Note. PDS = Public Defender Service.

p < .05.

Table 3 contains the predicted uncensored latent sentences estimated by the tobit model based on the charge at conviction and also by applying the coefficients of the model to the arraignment characteristics of each individual. Because the sentence lengths were logged, we have taken the antilog so that the results are in months. As expected, the predicted sentence length is higher when applying the estimated coefficients to the arraignment characteristics. Before the guidelines, if defendants had been sentenced based on their arraigned offense (using sentence magnitudes associated with conviction offenses), their sentences would be 30.3% shorter. For defendants sentenced after the guidelines, the reduction from charge bargaining is 35.9%. Using logged tobit values, a two-tailed t test was conducted to compare the differences between the predicted values of the bargain before and after guidelines. There was slightly more “value” to a charge bargain after the guidelines (M = 0.45, SD = 0.86) than there was before the guidelines (M = 0.36, SD = 0.77), although the difference is not significant, t(476) = −1.14, p = .25.

Table 3.

The Role of Charge Bargaining on Sentence Length: Tobits Before and After Sentencing Guidelines.

	Master Grid
Predictions^a based on coefficient estimates pre-guidelines
Predicted sentence using conviction charge (in months)	8.84
Predicted sentence using prosecutor charge (in months)	12.68
Change as a result of charge bargain	−3.84 (−30.3%)
Predictions based on coefficient estimates from post-guidelines
Predicted sentence using conviction charge	8.21
Predicted sentence using prosecutor charge	12.81
Change as a result of charge bargain	−4.60 (−35.9%)

These predictions are the uncensored latent sentence estimated by the tobit model. Because the sentence lengths were logged, we have taken the antilog so that the results are in months.

These results are consistent with Reitz’s (1998) prediction that structured sentencing will lead to increased discretion by prosecutors. The finding that there was a small increase in post-guidelines discretion persists whether we use ordinary least squares regression, tobit regression, or linear probability models on the incarceration decision.

Discussion and Conclusion

Because limited data are available on prosecutorial charge bargaining, the data obtained for this study present a unique opportunity to further test the methodology developed by Piehl and Bushway (2007) and determine whether it leads to a different result than the traditional method for measuring rates of charge bargaining. Although when they measured rates of charge bargaining, Vance and Oleson (2014) did not find evidence of hydraulic displacement of discretion after the 2004 introduction of guidelines in the District of Columbia, the current study found evidence of a modest shift of discretion to the prosecutor after the guidelines. This means that, even when examining a single data set (e.g., statistics from the D.C. Superior Court), different methodological approaches (e.g., the existence of the bargain vs. the value of the bargain) might reveal different associations between sentencing guidelines and their effects on prosecutorial discretion. Thus, additional research with expanded measurement methods should be conducted to further explore the impact of sentencing guidelines on the allocation of discretion within courtroom communities.

The current study provides much-needed empirical information about the displacement of discretion under sentencing guidelines, but the validity of the study is limited in several ways. First, the data were restricted to one jurisdiction, from just one pre-guidelines year and one post-guidelines year, and—because prosecutor data had to be manually coded by the primary author—were limited to a random sample of 23.5% of the processed cases. For the same reason, the data included only cases involving a single-count felony conviction. Second, neither bargaining over the number of charges nor sentencing bargaining (given a certain charge) were examined. Prosecutorial discretion has other consequences, such as limiting which cases enter the criminal justice system and how a defendant’s criminal history is measured. These decision points were not included in the study. Third, changes in initial charging practices post-guidelines cannot be examined by these data, either. This is problematic because, as demonstrated by Piehl and Bushway (2007), there is often a large movement from felony to misdemeanor charges as a result of introduced guidelines, and this movement was not captured by these data. Fourth, drug cases were not examined because the three offense severity groups in the Drug Grid did not provide sufficient movement from initial charge to conviction for application of the value of the bargain technique. Fifth, relatively little overall variance is explained by the pre-guideline model (pseudo r² of 6%) and the post-conviction model (pseudo r² of 12%). Nevertheless, despite these considerable limitations in the data, the current study is sufficient to test the utility of the Piehl and Bushway (2007) approach for measuring prosecutorial discretion.

As Engen (2008) wrote, answering the empirical question of how much the charging decision affects the severity of the sentence is a critical question upon which any conclusion regarding the displacement of discretion rests. Without first establishing this effect, “conclusions about the effects of prosecutorial discretion on disparity in sentencing and conclusions about whether these laws shift control over sentencing to prosecutors are premature” (p. 85).

The current findings raise serious policy issues as well. They confirm the findings of Piehl and Bushway (2007), who noted that measuring the value of the charge bargain produced a different estimate of the transfer of discretion than was produced by measuring the rate of bargaining. This should be of great interest to policy makers in addition to researchers, because the measurement of rates might suggest that no significant hydraulic displacement has occurred (e.g., Vance & Oleson, 2014), whereas the measurement of values might suggest that discretion has been transferred from the judge to the prosecutor (e.g., the current study). One measure reassures policy makers that the allocation of discretion has not been affected by the introduction of guidelines; the other measure indicates that discretion has in fact been displaced (as predicted by many researchers who theorized about hydraulic displacement). In the same way, methodological approaches that measure only direct race effects in sentencing disparity research fail to capture the influence of indirect and interactive effects (Spohn, 2015). Therefore, for both hydraulic displacement and racial disparity, the analytical model that is adopted, itself a function of the data that are available for analysis, can determine the ultimate findings. Accordingly, policy makers might wish to ensure that sufficient data are collected within sentencing systems to facilitate the accurate measurement of discretion (among prosecutors as well as judges).

Policy makers might also want to consider alternatives or adjustments to overly presumptive and rigid guideline systems. After all, sentencing guidelines were developed to standardize sentencing (Frankel, 1973; Hofer, Loeffler, Blackwell, & Valentino, 2004). But if disparity is not reduced but merely displaced, redirected from the judge—an impartial finder of fact who provides a transparent statement of reasons—to the prosecutor—an adversarial party in the case, without any written account of the exercise of discretion—then guidelines might camouflage and obfuscate sentencing disparities, making them harder to correct. “[R]ules to control discretion may shift the discretion so that it is invisible to the researcher” (Piehl & Bushway, 2007, p. 121). Accordingly, the “cure” of guidelines might be worse than the “disease” they were created to alleviate (Uelmen, 1991).

Restrictive, presumptive guidelines systems raise constitutional obstacles for policy makers as well as increasing the likelihood that discretion will be displaced upstream. In particular, recent Sixth Amendment jurisprudence from the U.S. Supreme Court forbids judges from imposing sentences in presumptive guideline systems that are not supported by facts admitted by the defendant or found beyond a reasonable doubt by a jury (e.g., Blakely v. Washington, 2004; Cunningham v. California, 2007). This constitutional prohibition does not apply to voluntary/informational guideline systems (United States v. Booker, 2005).

If guidelines are purely informational and/or voluntary (as in the District of Columbia, and many other U.S. jurisdictions; Kauder & Ostrom, 2008), they typically rely on a rudimentary set of metrics—typically crime seriousness on one axis and criminal history on another (Luna, 2005)—to generate a grid of available sentences. But if judges are afforded sufficient discretion and provided with sound information related to the defendant’s risk of future recidivism and the culpability and harm associated with the defendant’s past offenses, they—likely more than any other actor in the criminal justice system—are well positioned to exercise discretion in a deliberate and transparent manner.

Footnotes

Appendix

Appendix.

Master Grid.

	Criminal history score
Ranking group Most common offenses	0-½ A	¾-1¾ B	2-3¾ C	4-5¾ D	6+ E
Group 1 First degree murder w/armed First degree murder	360-720	360-720	360-720	360-720	360+
Group 2 Second degree murder w/armed Second degree murder First degree sex abuse First degree sex abuse w/armed	144-288	156-300	168-312	180-324	192+
Group 3 Voluntary manslaughter w/armed First degree child sex abuse Carjacking while armed Assault with intent to kill w/armed Armed Burglary I	90-180	102-192	114-204	126-216	138+
Group 4 Aggravated assault w/armed Voluntary manslaughter	48-120	60-132	72-144	84-156	96+
Group 5 Possession of firearm – crime of violence Armed robbery Burglary I Obstruction of justice Assault with intent to kill	36-84	48-96	60-108	72-120	84+
Group 6 Assault - Deadly Weapon Robbery Aggravated assault Second degree child sex abuse Assault with intent to rob	18-60	24-66	30-72	36-78	42+
Group 7 Burglary II Third degree sex abuse Negligent homicide Assault w/I to commit mayhem Attempt second degree sex abuse	12-36	18-42	24-48	30 -54	36+
Group 8 Carrying pistol w/o license Unauthorized use of vehicle Attempt robbery Attempt burglary First degree theft	6-24	10-28	14-32	18-36	22+
Group 9 Escape/prison breach Bail Reform Act Receiving stolen property Uttering Forgery	1-12	3-16	5-20	7-24	9+

Note. White/unshaded boxes—prison only. Dark shaded boxes—prison or short split permissible. Light shaded boxes—prisons, short split, or probation permissible.

Authors’ Note

The views expressed in this article are those of the authors alone and do not reflect the official position of the Administrative Office of the U.S. Courts.

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

Author Biographies

Stephen E. Vance, JD, PhD, is a senior attorney at the Administrative Office of the U.S. Courts. He received his JD from the Sandra Day O’Connor College of Law, Arizona State University, and his PhD from the University of Maryland, Department of Criminology and Criminal Justice. He was a Supreme Court Fellow assigned to the U.S. Sentencing Commission in 2009-2010.

Kerry M. Richmond, PhD, is an associate professor of criminal justice and chair of the Department of Criminal Justice and Criminology at Lycoming College, Williamsport, Pennsylvania. Her research interests include correctional program evaluation, gender and crime, and prisoner reentry. She received her PhD from the University of Maryland, Department of Criminology and Criminal Justice, College Park, Maryland.

James C. Oleson is an associate professor of criminology at the University of Auckland and a visiting fellow in the Department of Sociology at the Rijksuniversiteit Groningen. He received his MPhil and PhD from the University of Cambridge and his JD from the University of California, Berkeley (Boalt Hall). He served as a 2004-2005 Supreme Court Fellow at the Administrative Office of the U.S. Courts, and then served as Chief Counsel of the Criminal Law Policy Staff from 2005 to 2010.

Shawn D. Bushway is a professor of public administration and policy at the Rockefeller College of Public Affairs and Policy at the University at Albany, SUNY.

References

Albonetti

C. A.

(1997). Sentencing under the Federal Sentencing Guidelines: An analysis of the effects of defendant characteristics, guilty pleas, and departures, 1991–1992. Law and Society Review, 31, 601-634.

Alschuler

A. W.

(1978). Sentencing reform and prosecutorial power: A critique of recent proposals for “fixed” and “presumptive” sentencing. University of Pennsylvania Law Review, 126, 550-577.

Ball

J. D.

(2006). Is it a prosecutor’s world? Determinants of count bargaining decisions. Journal of Contemporary Criminal Justice, 22, 241-260.

Blakely v Washington, 542 U.S. 296 (2004).

Coffee

Tonry

(1983). Hard choices: Critical tradeoffs in the implementation of sentencing reform through guidelines. In Tonry

Zimring

(Eds.), Reform and punishment: Essays in criminal sentencing (pp. 155-203). Chicago, IL: University of Chicago Press.

Cunningham v California, 549 U.S. 270 (2007).

Davis

(2007). Arbitrary justice: The power of the American prosecutor. New York, NY: Oxford University Press.

District of Columbia Sentencing Commission. (2003). Annual report to the Council of the District of Columbia. Washington, DC: Author.

District of Columbia Sentencing Commission. (2006). Annual report to the Council of the District of Columbia. Washington, DC: Author.

10.

Eisenstein

Flemming

R. B.

Nardulli

P. F.

(1988). The contours of justice: Communities and their courts. Boston, MA: Little, Brown.

11.

Engen

R. L.

(2008). Have sentencing reforms displaced discretion over sentencing from judges to prosecutors? In Worrall

J. L.

Nugent-Borakove

M. E.

(Eds.), The changing role of the American prosecutor (pp. 73-90). Albany: State University of New York Press.

12.

Forst

(2002). Prosecution. In Wilson

J. Q.

Petersilia

(Eds.), Crime: Public policies for crime control (pp. 509-536). San Francisco, CA: ICS Press.

13.

Frankel

M. E.

(1972). Lawlessness in sentencing. University of Cincinnati Law Review, 41, 1-54.

14.

Frankel

M. E.

(1973). Criminal sentences: Law without order. New York, NY: Hill & Wang.

15.

Frase

(1999). Sentencing guidelines in Minnesota, other states, and the federal courts: A twenty-year retrospective. Federal Sentencing Reporter, 12(2), 69-82.

16.

Frase

(2000). Is guided discretion sufficient? Overview of state sentencing guidelines. St. Louis University Law Journal, 44, 425-446.

17.

Gorton

Boies

J. L.

(1999). Sentencing guidelines and racial disparity across time: Pennsylvania prison sentences in 1977, 1983, 1992, and 1993. Social Science Quarterly, 80(1), 37-54.

18.

Hartley

R. D.

Maddan

Spohn

C. S.

(2007). Prosecutorial discretion: An examination of substantial assistance departures in federal crack-cocaine and powder-cocaine cases. Justice Quarterly, 24, 382-407.

19.

Hartley

R. D.

Maddan

Walker

(2006). Sentencing practices under the Arkansas Sentencing Guideline Structure. Journal of Criminal Justice, 34, 493-506.

20.

Hofer

P. J.

Loeffler

Blackwell

Valentino

(2004). Fifteen years of guidelines sentencing: An assessment of how well the federal criminal justice system is achieving the goals of sentencing reform. Washington, DC: U.S. Sentencing Commission.

21.

Hunt

Connelly

(2005). Advisory guidelines in the post-Blakely era. Federal Sentencing Reporter, 17, 233-242.

22.

Kauder

N. B.

Ostrom

B. J.

(2008). State sentencing guidelines: Profiles and continuum. Williamsburg, VA: National Center for State Courts.

23.

King

N. J.

Klein

S. R.

(2004). Beyond Blakely. Federal Sentencing Reporter, 16, 316-332.

24.

Koons-Witt

B. A.

Sevigny

E. L.

Burrow

J. D.

Hester

(2014). Gender and sentencing outcomes in South Carolina: Examining the interactions with race, age, and offense type. Criminal Justice Policy Review, 25, 299-324.

25.

Kramer

J. H.

Lubitz

R. L.

(1985). Pennsylvania’s sentencing reform: The impact of commission established guidelines. Crime and Delinquency, 31, 481-500.

26.

Kramer

J. H.

Ulmer

J. T.

(2009). Sentencing guidelines: Lessons from Pennsylvania. Boulder, CO: Lynne Rienner.

27.

Lagoy

Hussey

Kramer

(1979). The prosecutorial function and its relation to determinate sentencing structures. In McDonald

(Ed.), The prosecutor (pp. 209-237). Beverly Hills, CA: Sage.

28.

Luna

(2005). Gridland: An allegorical critique of federal sentencing. The Journal of Criminal Law and Criminology, 96, 25-106.

29.

McCoy

(1984). Determinate sentencing, Plea bargaining bans and hydraulic discretion in California. Justice System Journal, 9, 256-275.

30.

McDonald

(1979). The prosecutor. Beverly Hills, CA: Sage.

31.

Miethe

T. D.

(1987). Charging and plea bargaining practices under determinate sentencing: An investigation of the hydraulic displacement of discretion. Journal of Criminal Law & Criminology, 78, 155-176.

32.

Miethe

T. D.

Moore

C. J.

(1985). Sentencing guidelines: Their effect in Minnesota. Washington, DC: National Institute of Justice.

33.

Minnesota Sentencing Guidelines Commission. (1984). The impact of Minnesota’s sentencing guidelines: Three year evaluation. St. Paul: Minnesota Sentencing Guidelines Commission.

34.

Nagel

I. H.

Schulhofer

S. J.

(1992). A tale of three cities: An empirical study of charging and bargaining practices under the Federal Sentencing Guidelines. Southern California Law Review, 66, 501-566.

35.

Ohlin

L. E.

Remington

F. J.

(1958). Sentencing structure; its effect upon systems for the administration of criminal justice. Law and Contemporary Problems, 23, 495-507.

36.

Osgood

D. W.

Finken

L. L.

McMorris

B. J.

(2002). Analyzing multiple-item measures of crime and deviance II: Tobit regression analysis of transformed scores. Journal of Quantitative Criminology, 18, 319-347.

37.

Ostrom

Hanson

Kleiman

(2008). Assessing consistency and fairness in sentencing—A comparative study in three states. Washington, DC: National Institute of Justice.

38.

Pennsylvania Commission on Sentencing. (1984). 1984 report: Sentencing in Pennsylvania. State College: Pennsylvania Commission on Sentencing.

39.

Piehl

A. M.

Bushway

S. D.

(2007). Measuring and explaining charge bargaining. Journal of Quantitative Criminology, 23, 105-125.

40.

Pillsbury

S. H.

(1989). Understanding penal reform: The dynamic of change. Journal of Criminal Law & Criminology, 80, 726-780.

41.

Reitz

(1998). Modeling discretion in American sentencing systems. Law & Policy, 20, 389-428.

42.

Savelsberg

(1992). Law that does not fit society: Sentencing guidelines as a neoclassical reaction to the dilemmas of substantivized law. American Journal of Sociology, 97, 1346-1381.

43.

Shermer

L. O. N.

Johnson

B. D.

(2010). Criminal prosecutions: Examining prosecutorial discretion and charge reductions in US federal district courts. Justice Quarterly, 27, 394-430.

44.

Smith

D. A.

Brame

(2003). Tobit models in social science research some limitations and a more general alternative. Sociological Methods & Research, 31, 364-388.

45.

Spears

J. W.

Spohn

(1997). The effect of evidence factors and victim characteristics on prosecutors’ charging decisions in sexual assault cases. Justice Quarterly, 14, 501-524.

46.

Spohn

(2002). How do judges decide: The search for fairness and justice in punishment. Thousand Oaks, CA: Sage.

47.

Spohn

(2015). Evolution of sentencing research. Criminology & Public Policy, 14, 225-232.

48.

Spohn

Beichner

Davis-Frenzel

(2001). Prosecutorial justifications for sexual assault case rejection: Guarding the “gateway to justice.” Social Problems, 48, 206-235.

49.

Spohn

Holleran

(2001). Prosecuting sexual assault: A comparison of charging decisions in sexual assault cases involving strangers, acquaintances, and intimate partners. Justice Quarterly, 18, 651-688.

50.

Steffensmeier

Ulmer

Kramer

(1988). The interaction of race, gender and age in criminal sentencing: The punishment cost of being young, black and male. Criminology, 36, 763-797.

51.

Stith

Cabranes

J. A.

(1998). Fear of judging: Sentencing guidelines in the federal courts. Chicago, IL: University of Chicago Press.

52.

Stolzenburg

D’Alessio

S. J.

(1994). Sentencing and unwarranted disparity: An empirical assessment of the long-term impact of state sentencing guidelines in Minnesota. Criminology, 32, 301-310.

53.

Tonry

Coffee

(1987). Enforcing sentencing guidelines: Plea bargaining and review mechanisms. In von Hirsch

Knapp

Tonry

M. H.

(Eds.), The sentencing commission and its guidelines (pp. 142-176). Boston, MA: Northeastern University Press.

54.

Uelmen

G. F.

(1991). Federal sentencing guidelines: A cure worse than the disease. American Criminal Law Review, 29, 899-905.

55.

United States v. Booker, 543 U.S. 220 (2005).

56.

United States v. Tucker, 403 U.S. 443 (1972).

57.

Vance

S. E.

Oleson

J. C.

(2014). Displaced discretion: The effects of sentencing guidelines on prosecutors’ charge bargaining in the District of Columbia Superior Court. Criminal Justice Policy Review, 25, 347-377.

58.

Washington State Sentencing Guidelines Commission. (1992). A decade of sentencing reform: Washington and its guidelines, 1981-1991. Olympia: Washington State Sentencing Guidelines Commission.

59.

Wooldredge

Griffin

(2005). Displaced discretion under Ohio Sentencing Guidelines. Journal of Criminal Justice, 33, 301-316.