An Analysis of Factors Related to Prosecutor Sentencing Preferences

• Author: Gerard Rainville
• DOI: 10.1177/0887403401012004003
• Published date: 01 December 2001
• Date: 01 December 2001
• Subject Matter: Journal Article

Rainville / PROSECUTOR
CRIMINAL J
SENTENCING USTICE POLICY
PREFERENCES REVIEW / December 2001
An Analysis of Factors Related to
Prosecutor Sentencing Preferences
Gerard Rainville
American University
Three types of variables have been identified as related to prosecutor decision making
in the screening and settlement stages of criminal case-processing—legal, extralegal,
and resource variables. The current analysis examines the degree to which these
classes of variables affect prosecutor sentence preferences. Ordinary least squares
regression is used to relate factors that prosecutors regard as germane to forming sen-
tence preferences to a measure of sentence restrictiveness. Analyses reveal a dimin-
ished reliance on legal and extralegal variables in the determination of preferred sen-
tences. In their stead, the available correctional placement options within a
prosecutor’s jurisdiction as well as the personal values of prosecutors appear to
determine the level of sentence restrictiveness that prosecutors desire.
The scholarly attention devoted to the policy decisions of prosecutors is
negligible relative to the importance of the prosecutor’s role in the adminis-
tration of justice (Forst & Brosi, 1977; McDonald, 1979a). In the United
States, the prosecutor is regarded as the “chief law enforcement officer” of
their jurisdiction (American Bar Association, 1971; Jacoby, 1979)and has
become a central figure in determining the outcome of criminal complaints
(Adams & Cutshall, 1987; McDonald, 1979b).
Although the amount of research devoted to prosecutors has been inade-
quate, a small body of empirical research has examined several aspects of
the prosecutor’s decision making. Variables affecting prosecutor decision
making in regard to filing charges, in accepting a plea, and in other decision-
making areas such as in admitting evidence (Spohn & Horney, 1992)and
opting whether to prosecute juvenile offenders in adult courts (Bishop, Frazier, &
Henretta, 1989; Schiraldi & Ziedenberg, 1999)have all been examined. One
area that has not been examined extensively is the prosecutors’ decision
AUTHOR’S NOTE: A previous version of this article was presented at the American Society
for Criminology’s Annual Conference, San Francisco, 2000. The author wishes to thank
Paige Harrison and anonymous reviewers for comments on previous drafts.
Criminal Justice Policy Review, Volume 12, Number 4, December 2001
295-310
© 2001 Sage Publications
295

---

296
CRIMINAL JUSTICE POLICY REVIEW / December 2001
making in regard to sentencing.1 The purpose of this study is to present an
analysis of variables that prosecutors regard in their conception of an appro-
priate criminal sentence.
REVIEW OF PROSECUTOR
DECISION-MAKING STUDIES
To guide the current analysis, a brief consideration is made of variables
that have been shown empirically to affect prosecutor decision making at
various case-processing stages. For example, it has been noted that varia-
tions in the seriousness of the offense are correlated with both a prosecutor’s
decisions to file charges (Albonetti, 1987; Frase, 1980; Jacoby, Mellon,
Ratledge, & Turner, 1982)and to expend greater resources on a case
(Albonetti, 1986; Landes, 1971; McDonald, Rossman, & Cramer, 1979).
As such, it may be likely that such a variable informs the prosecutor’s pref-
erence for the nature of a criminal sentence.
There are three broad classes of variables that have been identified in the
prosecutor decision-making literature:
Resource-oriented variables. Such variables are not related to any partic-
ular case at hand, but rather are environmental considerations that prosecu-
tors must regard in determining their available courses of action in all cases.
For instance, the absence of pretrial detention facilities in a prosecutor’s dis-
trict may affect a prosecutor’s preference for the denial of bail. When such a
resource is scarce, it may be reserved only for those cases that most urgently
demand pretrial detention. A very basic resource-oriented variable is the
average caseload per prosecutor. The number of cases that prosecutors are
expected to handle should determine the decisions of the prosecutor as to
how to handle their respective volume of cases.
Legal variables. Legal variables are clearly case related. Such variables
generally provide justifiable grounds for opting to pursue cases in a particu-
lar way. For example, if a case involves a serious crime, a repeat offender
and/or the evidence is strong, it should not be screened out.
Extralegal variables. These are variables for which gray areas exist in
regard to the propriety of their use. For example, it is not just for prosecutors
to fail to pursue a case because they object to the lifestyle, language, or man-
ner of a victim or witness. However, when the aforementioned variables
clearly affect the reliability of the evidence in a case, such as they would if

---

Rainville / PROSECUTOR SENTENCING PREFERENCES
297
the evidence in the case consisted mainly of the testimony of this sole wit-
ness, then seemingly irrelevant “extralegal” variables may become legally
relevant (Adams & Cutshall, 1987). Prejudicial variables are an extreme
form of extralegal variables and are rarely appropriate to consider in decid-
ing how to pursue a case. Considerations of the offender’s or victim’s race
and/or gender, for example, should not affect the diligence of prosecutors in
pursuing justice in a given case.
What has not been examined is the degree to which prosecutors’personal
preferences serve as a source of variation in case-processing decisions.
Jacoby (1979, p. 77)contends that “personal considerations” of prosecutors
certainly weigh on their decisions, but empirical pursuit of the interaction
between these considerations, including “the prosecutor’s philosophy of
law and his (sic)perception of the prosecutor’s purpose” and their preferred
case outcomes have been generally lacking.
A review of the variables that have been shown, empirically, to affect
prosecutor screening and case settlement decisions are presented below in
Table 1.2 The main feature of this review is the ability to classify most
case-processing variables into three general classes—resource, legal, and
extralegal/prejudicial variables.
PROSECUTORIAL SENTENCING PREFERENCES
Although judges pronounce criminal sentences, the prosecutor often rec-
ommends sentences to the judge, especially in cases settled by pleas
(Alschuler, 1983). The privilege of recommending sentences, however,
does not represent the chief mechanism through which prosecutors affect
sentencing. In the 1980s, legislatures began to limit the discretionary power
of judges by introducing sentencing guidelines to structure the sentencing
process. Additionally, legislatures imposed three-strike laws and manda-
tory add-ons for particular charges. The net effect of sentencing reforms
was a transfer of discretion from judges to prosecutors in regard to sentenc-
ing (Morris & Tonry, 1990; Tonry, 1996). Under such conditions, the prose-
cutor has great latitude to (effectively)determine the sentence that offenders
serve by choosing which charge to file (in full awareness of the presumptive
penalty and the presence of add-ons or three-strike stipulations). Thus,
prosecutors may bargain with offenders both by sentence-bargaining (rec-
ommending leniency)and charge-bargaining (filing charges with lower
presumptive sentences). Furthermore, the power of prosecutors to affect
sentences has not been significantly altered under conditions in which plea
bargains are banned (Carns & Kruse, 1992)or constrained by legislatures

---

298
CRIMINAL JUSTICE POLICY REVIEW / December 2001
Table 1:
Variables Related to Prosecutor Decision Making at Various Stages of
Case-Processing (sources of empirical support)
Case screening
Resource variables
Screening policy due to caseload (Boland & Forst, 1985).
Legal variables
Quality of police evidence (Albonetti, 1987; Boland, Mahanna, & Sones, 1992;
Feeney, Dill, & Weir, 1983; Forst, Lucianovic, & Cox, 1977; Spears & Spohn,
1997).
Seriousness of the offense (Albonetti, 1987; Frase, 1980; Jacoby, Mellon, Ratledge,
& Turner, 1982).
Prior criminal record (Adams & Cutshall, 1987; Albonetti, 1987; Neubauer, 1974).
Extralegal or prejudicial variables
Victim relation to offender (Albonetti, 1987; Cannavale & Falcon, 1976; Forst et al.,
1977; Spears & Spohn, 1997; Stanko, 1981-1982).
Demographics of offendera (Adams & Cutshall, 1987; Spears & Spohn, 1997; Stanko,
1981-1982).
Summary (Plea)or full extent (trial-ready)prosecution
Resource variables
Defendant’s costs for trial (Emmelman, 1996; Landes, 1971; Rhodes,...