Draft
Preliminary Findings:
Estimating the Effect of Targeted Enforcement
Strategies
on Conviction and Imprisonment Rates
for New York City
Sharon E. Lansing
School of Public Administration & Policy
Nelson Rockefeller College of Public Affairs and Policy
University at Albany
State University of New York
135 Western Avenue
Milne Hall
Albany, NY 12222
Phone: (518) 457-7301
Fax: (518) 457-9700
E-Mail: lansing@ dcjs.state.ny.us
Paper submitted for presentation at the
Nineteenth International Conference of the System Dynamics Society
July 23-27, 2001 in Atlanta, Georgia, USA
Draft Paper Submitted on March 15, 2001
ABSTRACT
The growth in prison populations since the seventies has become an increasingly important area
of study in the social sciences. Explanations for this growth have largely focused on the extent to which
criminal justice policies and social forces. Interestingly, little attention has been paid to the role of system
resources and the dynamics of criminal justice system behavior in prison population growth. This paper
will argue that decision making in the criminal justice system following arrest is shaped largely by the
dynamics of system behavior, more so than law, policy or individual discretion. It contends that it is the
allocation of a system’s case processing capacity between the upper and lower courts that largely
determines the amount of growth in new prison admissions. Furthermore, the allocation of this capacity
can be affected by policing strategies that produce substantial and protracted increases in the volume of
arrests. System Dynamics is among the analytical techniques used to develop this argument.
Contents
Section 1 The Scale of Imprisonment
Section 2 Understanding the Dynamics of Criminal J ustice System Behavior
+ An Exploratory Study of Dynamic System Behavior
- Penal Philosophy and System Capacity
- Arrest Volume and Thresholds for Upper Court Prosecution
- Prosecutorial Decision Making
- Shifting Conviction Capacity
+ An Summary of Preliminary Findings and Conclusions
Section 3 A Theory and Hypotheses of Criminal J ustice System Behavior
Section 4 System Dynamics Model
Section 5 Possible Policy Implications ..
Appendix (NOT ATTACHED - see cover memorandum)
Figures 1-5
Draft Document: March 15, 2001
Estimating the Effect of Targeted Enforcement Strategies
on Conviction and Imprisonment Rates
for New York City
Section1 The Scale of Imprisonment
The growth in prison populations since the seventies has become an increasingly important area
of study in the social sciences (Tonry and Petersilia 1999). Research concermed with seeking
explanations for this growth has largely focused on understanding the extent to which criminal justice
policies and social forces have influenced this growth.
Interestingly, little attention has been paid to the role of system resources and the dynamics of
criminal justice system behavior in prison population growth. One important exception is a study
conducted by Paul McCold (1993). Using a system dynamics modeling, McCold showed that changes in
police, prosecutor and court resources were sufficient to explain the growth in new prison admissions in
New Y ork State regardless of changes in the crime rate or other exogenous factors (i.e., demographic
changes in age groups, unemployment, prison crowding, and sentencing policy) (1993:90-93).
This study will argue that decision making in the criminal justice system following arrest is shaped
largely by the dynamics of system behavior, more so than law, policy or individual discretion. In doing so,
it contends that it is the allocation of a system’s case processing capacity between the upper and lower
courts that largely determines the amount of growth in new prison admissions. Furthermore, it contends
that the allocation of this capacity can be affected by policing strategies that produce substantial and
protracted increases in the volume of arrests. System Dynamics is among the analytical techniques used
to develop this argument.
Section 2 Understanding the Dynamics of Criminal J ustice System Behavior
This study is an outgrowth of research that was concemed with understanding why conviction
rates for juveniles prosecuted as adults for serious violent offenses in the upper (felony) courts of four
New Y ork City counties - Bronx, Kings, New Y ork and Queens - had fluctuated, sometimes by as much
as 15 percent, over the past two decades.’ During the course of that research, trends in upper court
conviction rates for juveniles were compared to those for adults who were arrested for comparable
crimes. The analysis showed that trends were the same for both groups. Because the instability of
conviction rates appeared to be a systemwide phenomenon, it became clear it would not be possible to
understand, in isolation from the larger system, why juvenile conviction rates were fluctuating.
An exploratory study was subsequently undertaken to identify factors that could have contributed
to the substantial changes in upper court conviction rates for violent felony arrests. System
Dynamics(SD) methodology was used to develop a simulation model to explore criminal justice system
1 In New York State, the crimes for which juveniles can be prosecuted as adults are defined in New Y ork State Penal
Law § 10.00(18); these crimes are a subset of serious violent crimes defined as violent felony offenses in PL § 70.02.
Draft Document: March 15, 2001 -L
dynamics that might explain this behavior.’ A discussion of the analytical approach taken and findings
from this exploratory study follows.
An Exploratory Study of Dynamic System Behavior
The exploratory System Dynamics (SD) analysis examined - from a systemic perspective - how
the targeted enforcement of drug offenses in the eighties and, in the mid-1990s, a broader spectrum of
quality of life crimes affected the prosecution of offenders arrested for non-targeted offenses, particularly
violent offenses. New Y ork City’s four most populous counties - Bronx, Kings, New Y ork, and Queens
were the jurisdictions studied. Two research questions guided the exploratory analysis:
* To what extent did the growth in conviction capacity in the upper and lower courts in each of the
four counties studied keep pace over the past two decades with the growing number of arrests
for drug and quality of life offenses?
« If the growth in conviction capacity did not keep pace with the growth in arrests, to what extent
were case outcomes effected? That is, how were the resulting losses in conviction capacity
distributed across arrests offenses by offense type and seriousness with respect to the likelihood
of conviction and incarceration?
SD model parameters were calibrated with data from the New Y ork State Computerized
Criminal History (CCH) data base maintained by the State’s Division of Criminal Justice Services. The
police, prosecutors and courts are required under State law to report all fingerprintable arrests to the
State, as well as the subsequent dispositions of these cases.
Initially, widely held assumptions about criminal justice system behavior were relied upon to build
the model’s structure. When an assumption did not generate the observed behavior, it became necessary
to retum to the data base to conduct descriptive analyses that would help to explain the unexpected
system behavior. Several iterations of movement back and forth between the data and the model were
necessary to separate assumption from fact for model construction. The hypotheses guiding the current
study (and presented later in this section) evolved during the course of this iterative, model-fitting process.
A summary of the findings from which the hypotheses flow follows. Figures cited below are presented in
the Appendix which is not attached to this paper; please see cover memorandum.
Penal Philosophy and System Capacity
The number of defendants sentenced to prison in New Y ork City’s upper courts has largely been
a function of the number of cases forwarded to the upper courts for felony prosecution.’ This
relationship is illustrated in Figure 1. The knowledge that this behavior has been relatively consistent
over time, along with the fact that only defendants prosecuted in an upper court and convicted of
2 ‘The SD model is still undergoing revisions. It is not presented in this draft paper, but will be included in the final
paper.
a Although reported crime has witnessed a substantial decline during the latter half of the nineties, the overall number
of arrests for felony offenses has not declined. Thus, the decline in reported crime, alone, does not explain the substantial decline
in new prison admissions that began in the mid-nineties.
Draft Document: March 15, 2001 -2-
committing a felony offense can be sentenced to prison, is the key to unraveling the dynamics that
underlay fluctuations in the number of persons sentenced to prison in New Y ork City jurisdictions. To
understand why the number of new admissions to prison fluctuates over time, one must first understand
the dynamics of criminal justice system behavior that affect changes in thresholds for upper court
prosecution. These are dynamics that are largely shaped by penal philosophy and a system’s capacity to
prosecute cases in the upper court.
Penal philosophy. During the seventies, a number of laws were passed in New Y ork State that
provided for harsher penalties, largely mandatory prison sentences, in cases involving drug offenses (early
seventies) and serious and repeat offenders (late seventies). New Y ork State did not act alone; many
other states enacted similar statutes during the seventies.
Public support for these harsher penalties was evident in the additional federal, state, and local
monies allocated to prosecutors and courts to support the enforcement of these new laws. The additional
monies allowed systems to increase their capacity for upper court prosecutions (achieved through
reductions in thresholds for upper court prosecution) in order to promote the imposition of prison
sentences under the new laws. Figure 1 illustrates the growth in the number upper court convictions in
the four New Y ork City jurisdictions studied. As a consequence, the number of defendants sentenced to
prison began to increase in New Y ork State.
In the mid-nineties, New Y ork State demonstrated its continued support for the enforcement of
these harsher sanctions with the passage of determinate sentencing laws. These laws targeted
defendants convicted of violent felony offenses. However, despite this reaffirmation of penal policy and
the absence of any substantial decline in arrests for serious offenses, the number and percentage of
defendants sentenced to prison in New Y ork City’s criminal justice system has been declining since the
early to mid-nineties.'
Conviction capacity. The capacity of a system to enforce laws is essential if the intent of the
laws is to be realized. When the growth in a system’s capacity to prosecute cases lags behind the growth
in arrests, the capacity to convict is strained. In such an instance, the number and percentage of persons
convicted can decline irrespective of penal philosophy.
The mechanism through which conviction rates are adjusted is the threshold for upper court
prosecution. Thresholds, which vary based largely on offense type and seriousness, are determined by
prosecutors and achieved through the creation of guidelines (generally informal) that specify the legal
criteria that must be met for a case to be prosecuted in the upper court. When capacity is strained,
thresholds are raised to reduce the number of cases prosecuted. Conversely, when there is excess
capacity thresholds are lowered. The exploratory analysis tracked changes in prosecutorial thresholds by
examining changes over time in the number of convictions in a given jurisdiction relative to the number of
arrests.
* Must address the possibility of judicial back lash - would only be evident if the proportion of upper court
convictions resulting in prison sentences declined. This did not happen.
Draft Document: March 15, 2001 -3-
Arrest Volume and Thresholds for Upper Court Prosecution
Over the past two decades, the number of arrests in New Y ork City surged upward. This growth
is largely attributable to targeted enforcement strategies implemented by the New Y ork City Police
Department (NY PD) during the eighties against drug offenses and in the nineties against a broader
spectrum of quality of life offenses. It was during the initial phase of this growth, in the mid-eighties, that
conviction rates for felony offenses began to decline.
In Bronx County, for example, there was a sudden and permanent decline in the number of felony
convictions in 1990 (Figure 2). This pattem of decline (i.e., a sudden downward decline that then
stabilizes) suggests that there may have been a change in the prosecutorial screening criteria that, in
effect, raised the threshold for upper court prosecutions. Kings and New Y ork counties witnessed similar
declines in felony conviction rates during the later half of the eighties, although these declines were less
stable over time than in Bronx County. Only in Queens County was the decline in the number of felony
convictions limited to a two-year period spanning 1989 and 1990.
One possible explanation for a decline in conviction rates is a change in the mix of arrest case
characteristics (e.g., offense type and seriousness and prior record). When controls were add for court
type ( i.e., upper and lower) and arrest offense type (i.e., violent, drug, and other), little support was found
for this explanation. Graphs that track the percent change in the number of arrests (the heavy solid line);
upper court convictions (the thin line with the square marker); and lower court convictions (the thin solid
line) are displayed in Figure 4 for Bronx County.
Recall, the graph for Bronx County in Figure 2 shows that the number of convictions for felony
arrests overall declined in 1990. Controlling for upper and lower court convictions, though, Figure 3A
shows that upper court convictions ( the line with the square marker) began to decline several years
earlier (1987). The fact that this decline was offset by an increase in lower court convictions (the solid
thin line) from 1987 through 1989 explains why Figure 2 does not show an overall decline until 1990 for
Bronx County.
Figures 3B-D control for both court and offense type. The important comparison across the
graphs presented in these three figures is the percent change in the number of upper court convictions
(the line with the square marker) relative to the percent change in arrests ( the heavy solid line).
Collectively, the graphs show that the overall decline in upper court convictions (Figure 3A) was due
largely to the decline in the number of convictions for violent felony arrests. In contrast, the percent
change in upper court convictions for drug and other felony arrests kept pace with, and sometimes
exceeded, the growth in their respective arrest categories.
Though not presented in this paper, analyses found somewhat similar patterns for felony case
processing behavior in Kings and New Y ork County and, to a lesser degree, in Queens County. The fact
that the temporal occurrences of the shifts in percent changes differ across counties despite the fact that
arrest trends were similar across counties, suggests that these shifts are more likely due to changes in
prosecutorial thresholds than changes in the mix of case characteristics. Further undermining the “case
mix” hypothesis is the fact that the average seriousness of the top charge in violent felony arrest cases did
not decline during the latter half of the nineties in any of the four counties.
The role that case processing constraints play in the unraveling of this dynamic puzzle is far
more clear. The graphs presented in Figure 4 contrast annual trends spanning 1983-1999 in the number
Draft Document: March 15, 2001 -4
of arrests (in thousands) for all felony and all misdemeanor arrests with upper court conviction rates for
violent felony arrests. These graphs show the strong, inverse correlation of arrest volume and conviction
rates for violent felony offenses. In the eighties, when arrests were increasing for both felony and
misdemeanor offenses, it is highly probable that the increase in felony arrests during that period
contributed to the decline in felony convictions. It is less clear to what degree, if any, the growth
misdemeanor arrests effected upper court conviction capacity during the eighties. In the nineties, though,
when felony arrests were relatively stable, felony conviction rates declined further. Given this fact, it is
reasonable to conclude that the post-1990 decline in upper court convictions may be related to the
increases in misdemeanor “quality of life” offenses during that period.
These findings pointed the way to the next question to be answered in this puzzle of dynamic
system behavior - Why was the narrowing of prosecutorial thresholds largely constrained to cases
involving violent felony arrest offenses? This finding is counterintuitive to assumptions about the role that
offense seriousness and type play in case processing behavior.
Prosecutorial Decision Making
The purpose of targeted enforcement is to deter crime. Because deterrence can sometimes be
achieved through arrest alone, targeted enforcement is not always undertaken by the police with the
expectation that convictions will follow. However, statistics from the exploratory analysis showed that
upper and lower court conviction rates declined for non-targeted offenses rather than targeted offenses in
all four study sites. Thus, the problem became one of identifying the mechanisms within the system that
would explain the unexpected pattems of divergence in conviction rates. Attention was tumed to
understanding the dynamics of the prosecutorial screening process.
Ethically bounded prosecutorial decision making. Michael and Don Gottfredson (1988)
conducted an extensive review of research concerned with decision making in the criminal justice system.
This work provided important insights in the search for the systemic mechanism. They found that the
body of research pertaining to prosecutorial decision making had consistently identified the principal legal
correlates for the decision to prosecute as (1) a prosecutor’s belief in an offender's guilt, (2) the belief
that an offense warrants prosecution (i.e., the punishment fits the seriousness of the crime), (3) the
seriousness of the offense, and the (4) strength of evidence’ (not in order of importance). Findings
regarding the importance of an offender’s prior record in the decision to prosecute were mixed. With
respect to judicial decisions at sentencing, research has consistently found that the likelihood imprisonment
is highly correlated with the seriousness of the crime committed and the defendant's prior criminal history
(Blumstein, Cohen, Martin and Tonery 1983; Gottfredson and Gottfredson 1988).°
From among these correlates, it was reasoned that strength of evidence was the most pivotal
factor for successful prosecution. Factors that substantially contribute to the sufficiency of evidence
include the evidence gathering ability of the police and the willingness and ability of witnesses to
5 Once prosecution was undertaken, the principal reasons for prosecutors to request case dismissals were insufficient
evidence and a defendant's successful participation in a diversion program. To a lesser degree, some cases were not prosecuted
or latter dismissed at the prosecutor’ s request because of due process problems with arrest procedures and the willingness of
offenders to cooperate with law enforcement to advance the prosecution of other criminal cases.
5 ‘The extent to which prosecutorial plea-bargaining and probation pre-sentence reports help to shape judicial
decisions has also been the focus of much research. Plea bargaining is among the most controversial of prosecutorial tools.
Draft Document: March 15, 2001 -5-
participate in the prosecution of an offender (Gottfredson and Gottfredson 1991). While prosecutors may
have some influence over the quality of police evidence, they have little, if any, control over witnesses
other than police officers. Furthermore, research has shown that the victim-offender relationship was the
most often cited reason for the failure of witnesses to cooperate.
The concept of ethically bounded prosecutorial decision making is introduced at this point to
help explain the unexpected divergence in conviction rates across offense types. The concept is
grounded in the belief that the sufficiency of evidence, particularly the availability and willingness of
witnesses to provide testimony, is pivotal for successful prosecution. It contends that, in systems where
the capacity to prosecute cases is strained, prosecutors - who are ethically bound to prosecute cases
where guilt can be proven - are compelled to direct their attention toward cases for which the sufficiency
of evidence is high and away from cases, regardless of case seriousness, where the relative sufficiency
of evidence is low.
The sufficiency of evidence, particularly the availability and willingness of witnesses to provide
testimony, is generally much stronger in targeted arrest cases than in non-targeted arrest cases. This
occurs because of the nature of these arrests and the fact that the police are the primary witnesses in
such cases. Consequently, as targeted arrests increase as a proportion of all arrests, prosecutorial
capacity for prosecution and conviction, absent intervention (e.g., increased efficiencies to expand system
capacity for convictions), is slowly, and unwittingly, diverted from the prosecution of non-targeted
offenses.
Based on this concept, the unexpected divergence in conviction rates in the four New Y ork City
counties can be explained, at least in part, as a systemic response to the enormous growth in arrests for
targeted offenses. As their capacity to prosecute cases became increasingly strained, prosecutors were
ethically compelled to shift their attention toward the prosecution of targeted arrests. As a consequence,
conviction capacity was slowly, and unwittingly, diverted from the prosecution of non-targeted offenses,
particularly violent offenses, in both the upper and lower courts.’
Proportionality. Findings from the exploratory analysis also indicated there were boundaries that
prevented the percentage of violent arrests convicted in the upper court from dropping below a certain
level. That is, there appeared to be a certain percentage of cases in which offense seriousness was
perceived as so great, prosecutors were ethically bound to prosecute these cases in the upper court. Only
once these “floors” in conviction rates were reached for violent and other non-targeted felony offenses,
did upper court convictions rates for targeted offenses (i.e., felony drug arrests) begin to decline (Figure
5).
This observation lends further support the idea that there exists within legitimate systems of
criminal justice an innate sense of proportionality (discussed in the previous section). Also, it suggests that
the dynamics that structure the magnitude of a “going rate” may also be responsible for the determination
of this floor. These concepts of dynamic system behavior help to explain why upper court conviction
percentages for felony drug arrests began to decline in some counties in the latter half of the nineties, just
as those for violent offenses were stabilizing.
7 Ttis unknown at this time to what extent, if any, targeted enforcement strategies contributed to this problem by
diverting police resources from the investigation of non-targeted arrests.
Draft Document: March 15, 2001 -6-
Shifting C onviction Capacity
Finally, findings from the exploratory SD study showed that during periods when targeted
enforcement caused arrests for misdemeanor offenses to grow exponentially, conviction capacity was
diverted from the upper to the lower court. This not only caused the number of upper court convictions to
fall, but also the number of cases in which prison sentences were imposed. This occurred because the
number of defendants sentenced to prison by New Y ork City’s upper courts has largely been a function
of the number of cases forwarded to the upper courts for felony prosecution.
This finding is supported by Paul McCold’s (1993) work which showed that changes in police,
prosecutor and court resources (summary measures for system capacity), alone, were sufficient for
estimating changes in the number of observed prison admissions. The exploratory analysis, though, found
that it was not only the amount of system capacity available for prosecuting cases,’ but also the manner in
which the “conviction” capacity was allocated between the upper and lower courts that determined the
number of offenders sentenced to prison in New Y ork City courts. The exploratory research also
indicated that it was largely changes in arrest volume that determined how conviction capacity was
distributed across the upper and lower courts. Together, the literature review and the exploratory analysis
suggest that the “going” conviction rate is the mechanism through which an innate sense of proportionality
is manifested within systems of criminal justice.
Summary of Preliminary Findings and Conclusions
In summary, the key findings from the preliminary analysis include the following:?
. The number of defendants sentenced to prison in New Y ork City’s upper courts has
largely been a function of the number of cases forwarded to the upper courts for felony
prosecution. Given this knowledge, it may be possible to control the scale of
imprisonment if the dynamics that underlie decisions to prosecute cases in the upper court
can be better understood.
. There is evidence that a substantial and protracted increase in the volume of
misdemeanor arrests can cause a decline in the number and percentage of felony arrests
that result in upper court convictions.
. Changes over time in the percentage of upper court convictions for felony arrests were
not proportionally distributed across offense types; conviction rates for violent felony
arrests declined while those for drug (PL 220) felony arrests initially increased.
8 The exploratory research conducted for this study used conviction capacity (i.e., a summary measure of all factors,
including resources, that determine court capacity) rather than resources as summary measure of all efforts undertaken to effect
case processing capacity
° These findings and conclusions are based on quantitative analyses only. To date I have had no opportunities to
discuss my findings with practitioners. It is possible that such discussions could alter my interpretation of findings.
Draft Document: March 15, 2001 -7-
Section 3 A Theory and Hypotheses of Criminal J ustice System Behavior
The proposed theory of system behavior presented below is composed of six core tenets or
hypotheses developed during the course of the exploratory analysis. The hypotheses assert that it is the
dynamic systemic forces of arrest volume, conviction capacity, ethically bounded prosecutorial decision
making, and the “going” rate, working in tandem - more so than law, policy or practitioner discretion -
that largely control the magnitude of conviction and imprisonment rates. Practitioner discretion exists only
to the extent it is allowed to operate within the boundaries established by these other systemic forces. (A
causal loop diagram will be included in the final paper.)
The concept of system equilibrium is central to this theory. For the purposes of this study,
equilibrium occurs in a criminal justice system or subsystem when the percent change in arrests is the
same as the percent change in convictions. The system dynamics which control and disrupt this
equilibrium are hypothesized below.
+ Hypothesis 1. An innate sense of proportionality exists within legitimate systems of criminal
justice.
It is this innate sense of proportionality that determines the amount of variance in conviction rates across
offenses and offenders. It is operationalized through the dynamics underlaying the evolution and
maintenance of a “going” conviction rate for an offense.
* Hypothesis 2. In a system or subsystem with limited conviction capacity, the system or
subsystem will be forced to seek a new equilibrium by adjusting conviction rates downward when
a substantial growth in the number of arrests is not offset by a similar growth in the capacity to
prosecute cases.
Thus, the stability and maintenance of “going” conviction rates is bounded by a system or subsystem’s
capacity to prosecute arrest cases. At the same time, a decline in arrests does not always translate into
an increase in conviction rates because there is no intemal system mechanism controlling marginal slack.
* Hypothesis 3. Gains or loses in conviction capacity resulting from changes in arrest volume
are not proportionally distributed across offense types in a system or subsystem because of the
ethical boundaries that constrain decision making.
Specifically, gains in conviction capacity (i.e., the number of cases a system is able to prosecute/convict)
flow naturally toward the prosecution of targeted arrest offenses and losses in conviction capacity flow
naturally toward nontargeted arrest offenses. This occurs because there is a stronger likelihood of
sufficient evidence in targeted arrest cases due to the nature of the arrest and the greater availability and
willingness of witnesses (i.e., the arresting officer) to provide testimony in such cases.
+* Hypothesis 4, In a system or subsystem with limited capacity, upper court conviction rates for
targeted offenses decline only when the downward adjustment of going conviction rates for non-
targeted offenses reach their lower limit or floor.
That is, there is a certain percentage of cases in which offense seriousness is so great, prosecutors are
ethically bound to prosecute these cases. Only once these “floors” in conviction rates are reached for
Draft Document: March 15, 2001 -8-
violent and other non-targeted offenses, do upper court convictions rates for targeted offenses begin to
decline.
+ Hypothesis 5. In a system or subsystem with limited conviction capacity, conviction capacity
will be shifted from the upper to the lower court when the conviction rate floors are reached for
misdemeanor (targeted and non targeted) arrest cases.
This action is necessary to prevent misdemeanor conviction rates from declining below their baseline
rates. The hypothesis assumes the availability of upper court conviction capacity.
* Hypothesis @ A reduction in upper court capacity will cause the number of cases in which
prison sentences are imposed to decrease.
This hypothesis, which contends it is relative - rather than absolute - case severity that determines the
distribution of sanction types across convicted cases, is grounded in the work of Robert Emerson (1983).
Emerson synthesized the major research findings about social-control decision making that emerged
during the late sixties and through the early eighties in the field of criminal justice and, to a lesser degree,
the fields of medicine, sociology, and organizational behavior. He argued that:
[UJunder a variety of circumstances, the individual case is not the sole or even the most
important unity for categorizing and disposing of cases. Particular cases are in fact
processed not independently of others but in ways that take into account the implications
of other cases for the present one and vice versa. These wider holistic concems and
influences are an important organizationally-based factor that shapes decision outcomes.
(1983:425-426)
Thus, a reduction in upper court conviction capacity will cause the number of cases in which prison
sentences are imposed to decrease because the distribution of sanctions is proportional and this
distribution is relatively stable over time. '° Why the proportion of convictions in which prison sentences
were imposed remains constant over time will not be examined in this study. Literature pertinent to the
explanation of this behavior includes the work of Emerson (1983) and Heumann (1978).
This theory stops short of hypothesizing what happens when the floor in conviction rates for
targeted (i.e., drug) felony offenses is reached. Traditional assumptions about the relationship between
deterrence and punishment have been found wanting. It is a widely held assumption that crime increases
when the deterrent and incapacitative effects of any law are diminished. So far, though, violent and other
serious non-targeted crimes have continued to decline despite the decline in conviction and incarceration
rates for these offenses. The possible policy implications of the co-occurrence of declining crime,
conviction and incarceration rates are discussed in Section 5.
19 Although reported crime has witnessed a substantial decline during the latter half of the nineties, the overall number
of arrests for felony offenses has not declined. Thus, the decline in reported crime, alone, does not explain the substantial decline
in new prison admissions that began in the mid-nineties.
Draft Document: March 15, 2001 -9-
Section4 System Dynamics Model
To be added.
Section5 Possible Policy Implications
Preliminary analyses suggest that this study may provide strong evidence that, since the early
eighties, the targeted enforcement of drug and other quality of life offenses in New Y ork City has
compromised the prosecution of offenders arrested for violent and other non-targeted offenses.
Certainly, the decline in conviction and incarceration rates for these offenses was an unintended
consequence of targeted enforcement efforts. Nonetheless, such facts should evoke alarm.
It is a widely held assumption that crime increases when the deterrent and incapacitative effects
of any law are diminished. In this instance, though, one must pause to consider the fact that reported
crimes involving violent'' and other non-targeted offenses have also declined during this same period.
That these two trends are positively correlated is counterintuitive to our assumptions about crime
prevention and control.
The mid-nineties decline in violent crime in New Y ork City has been attributed to factors that are
both deterrent and incapacitative in nature. Among these factors are the following:
. the enhanced enforcement of drug and other quality of life offenses;
. the enactment of laws enhancing sanctions for violent and repeat offenders; and
. the enactment of determinate sentencing laws which increased time served through the
abolition of parole.
Targeted enforcement cannot continue to grow unrestrained indefinitely in the face of New Y ork
City’s declining crime rates. A stabilization or decline in targeted enforcement could occur as a result of
decreasing marginal returns for targeted enforcement; this occurs when cost of crime suppression begins
to outweigh it benefits. It could also occur as a result of public outcry as the balance between personal
rights and public safety tip in favor of the latter. In New Y ork City, there has been some public outcry
regarding police abuses, but it has not been sufficient to quell such police actions. And earlier this year
arrests for targeted offenses began to climb once more as the police anticipated the possibility of another
drug crises.
Although violent crime has declined - whether as a result of targeted enforcement, tougher
sanctions, the abolition of parole, or other complex social phenomena -- it is important that the concept of
offender accountability not take a back seat to the concept of deterrence. Research has shown that it is
the certainty rather than the severity of punishment that has the greater deterrent effect. Greater
offender accountability can be achieved without increasing the scale of imprisonment.
Further research will be required to substantiate the findings reported in this study. Will the
hypotheses of criminal justice system behavior presented in this study be able to explain system behavior
1 Crimes of homicide, assault and robbery involving the use of guns peaked around 1991-92 in New Y ork City and
than began a steep decent in 1992 that has continued through 1999, while non-gun homicides, assaults and robberies began a more
controlled decent two years earlier in 1990 (Zimring and Fagan).
Draft Document: March 15, 2001 -10-
in other jurisdictions across New Y ork State and the nation? Is system size a factor that alters the
applicability of the hypotheses? Whether substantiated in whole or part, it would seem that methodologies
that help us to better understand the dynamics of criminal justice system behavior should take on a greater
importance in research.
If findings are substantiated by subsequent research, the task before us become one of
determining which policy levers to adjust to achieve the proper balance between deterrence (through
targeted enforcement) and just deserts (through the scale of imprisonment). For example, what should
the policy response be to a decline in offender accountability absent any apparent effect on crime rates?
What if the deterrent effects of targeted enforcement offset the presumed negative effects of declining
offender accountability? What about the principal of just deserts? Should the going conviction rates for
violent and other serious crime be allowed to continue their downward adjustments? Or should targeted
enforcement be reduced to allow conviction rates for non-targeted crimes to recover? Or should
conviction capacity be increased to accomplish this recovery and sustain the current level of targeted
enforcement?
If serious crime is declining despite the fact that the likelihood of conviction and incapacitation for
both targeted and non-targeted offenses has declined, what does this imply about the scale of
imprisonment in New Y ork State where New Y ork City jurisdictions account for around two thirds of
new prison admissions annually? Certainly, the passage of tougher sentencing laws for serious and repeat
offenders and the abolition of parole have enhanced the public’s safety by keeping these offenders off the
streets for longer period of time. At the same time, though, it suggests that use of incapacitation may
have been unnecessarily excessive in past years for marginal offenders.
If trends in the number of offenders sentenced to prison closely parallel trends in the number of
offenders convicted in the upper court, can we control the scale of imprisonment by exerting more control
over the number and type of cases prosecuted in the upper court? The exploratory study found evidence
of floors for upper court conviction rates that were remarkably similar for violent offenses across the four
jurisdictions studied. Should we seek out these lower thresholds for all crimes? The fact that violent and
other non-targeted offenses continued to decline under lower thresholds, suggests that upper court
conviction rates could remain at these lower levels without posing a serious threat to public safety and, at
the same time, reduce the scale of imprisonment in New Y ork State. To sustain this trend, though, we
need to understand why crime rates are declining. A draft report for exploratory analysis examining the
impact of targeted enforcement on crime rates is presented in Appendix F. Preliminary findings suggest
that it has largely been the increase in non-drug quality of life crimes that is responsible for this decline.
Finally, prison and jail populations increased during the seventies despite a federally supported
nationwide moratorium on prison construction. To what extent did the increasing availability of federal
monies funneled to states and localities through the now defunct Law Enforcement Assistance
Administration (LEAA) during the seventies for system improvements contribute to the growth in upper
court conviction capacity?” Zimring and Hawkins (1991:76) give little weight to the importance of
LEAA monies and the growth in prison populations. Interestingly, though, the growth in LEAA
disbursements parallels the growth in prison populations in New Y ork State and nationwide. Is it
2 in support of the national moratorium on prison construction, LEAA monies could not be used for the construction
of new prisons during the program's early years.
Draft Document: March 15, 2001 -11-
possible that these monies contributed - indirectly - to the enormous growth in prison populations during
the seventies as a result of the increased processing efficiencies made possible through LEAA funded
initiatives? Comparable questions should be asked about the role played by federal drug enforcement
monies during the eighties.
Draft Document: March 15, 2001 -12-
Bibliography
Albonetti, C. A. and Hepbum, J. R. (1996). Prosecutorial discretion to defer criminalization: The
effects of defendant’ s ascribed and achieved status characteristics. Journal of Quantitative
Criminology 12:63-81.
Albonetti, C. A. (1998).! Direct and indirect effects of case complexity, guilty pleas, and offender
characteristics on sentencing for offenders convicted of a white-collar offense prior to sentencing
guidelines. Journal of Quantitative Criminology 14:353-378.
Ashworth, Andrew J. (1992). Sentencing reform structures. In Crime and Justice: A Review of
Research, vol.16, edited by Michael Tonry. Chicago: University of Chicago Press.
Alschuler, A. (1978).1 Sentencing reform and prosecutorial power. University of Pennsylvania Law
Review 126:550-77. From Griset 1995
Berk , Richard A., David Rauma, Sheldon L. Messinger, and Thomas F. Cooley (1981). A test of the
stability of punishment hypothesis: The case of California, 1851-1970. American Sociological
Review 46:805-829. From Zimring and Hawkins 1991.
Beha, James A. II (1977). “And nobody can get you out”: The impact of a mandatory prison sentence
for the illegal carrying of a firearm on the used of firearms and on the administration of criminal
justice in Boston. Boston University Law Review 57:96-146 (pt.1), 289-333 (pt.2). From Tonry
1992.
Blalock, Hubert M. Jr. (1979). Social Statistics. 2d ed. New Y ork: McGraw-Hill Book Company.
Blumestein, Alfred and Allen F. Beck (1999). Population growth in U.S. prisons, 1980-1996. In Crime
and Justice: A Review of Research, vol.26, edited by Michael Tonry and Joan Petersilia.
Chicago: University of Chicago Press.
Blumestein, Alfred and Jacqueline Cohen (1973).! A theory of the stability of punishment. Journal of
Criminal Law and Criminology 64:198-207.
Blumestein, Alfred, Jacqueline Cohen, and Daniel Nagin (1976).! The dynamics of a homeostatic
punishment process. Journal of Criminal Law and Criminology. 67:317-334.
Blumestein, Alfred, Jacqueline Cohen, Susan E. Martin, and Michael H. Tonry, eds. (1983). Research
on sentencing: The search for reform vol.1. Washington, D.C.: National Academy Press.
Blumestein, Alfred, Jacqueline Cohen, Soumyo Moitra, and Daniel Nagin (1981).1 On testing the stability
of punishment hypothesis: A reply. Journal of Criminal Law and Criminology. 72(4):1799-
1808.
Draft Document: March 15, 2001 -13-
Blumestein, Alfred and Soumyo Moitra (1979).! An analysis of time series of the imprisonment rate in
the states of the United States: A further test of the stability of punishment hypothesis. Journal
of Criminal Law and Criminology. 70:376-390.
Blumestein, Alfred and Soumyo Moitra (1980).! Growing or stable incarceration rates: A comment on
Calahan’s “Trends in incarceration rates in the United States since 1880.” Crime and
Delinquency. 26:91-94.
Cahalan, Margaret (1979).! Trends in incarceration in the United States since 1880. Crime and
Delinquency. 25:9-41.
Caplow, T. and Simon, J. (1999).1 Understanding prison policy and population trends. In Crime and
Justice: A Review of Research, vol.26, edited by Michael Tonry and Joan Petersilia. Chicago:
University of Chicago Press.
Cappell, Charles L. and Gresham Sykes (1991).! Prison commitments, crime and unemployment: A
theoretical and empirical specification for the U.S., 1933-85. Journal of Quantitative
Criminology. 7:155-200.
Casper, J., Brereton,D. and D Neal (1982). The implementation of the California determinate
sentencing law. Washington, DC: U.S. Department of Justice. From Gottfredson et al (1988
From Griset 1995
Chaiken, J., T. Crabill, L. Holliday, M. Jaquette, M. Lawless, and E. Quade. (1976). Criminal justice
models: An overview. National Institute of Law Enforcement and Criminal Justice, Law
Enforcement Assistance Administration, U.S. Department of Justice.
Clear, T., J., Hewitt, and R. Regoli (1978).! Discretion and the determinate sentence: its distribution,
control, and effect on time served. Crime and Delinquency. 24:428-45. From Griset 1995
Crutchfield, Robert D., George S. Bridges, and Susan R. Pitchford (1994). Analytical and aggregation
biases in analyses of imprisonment: Reconciling discrepancies in studies of racial disparity.
Journal of Research in Crime and Delinquency 31:166-182.
Davis, K. (1969).! Discretionary justice: a preliminary inquiry. Urbana: University of Illinois Press.
From Griset 1995
Emerson, Robert M. 1983. Holistic effects in social control decision-making. Law and Society Review.
17:425-455.
Friedman, L. (1993).! Crime and punishment in American history. New York: Basic Books. From
Griset 1995
Gomme, I. M. and Hall, M. P. (1995).! Prosecutors at work: Role overload and strain. Journal of
Criminal Justice 23:191-200.
Gottfredson, Michael R. and Don M. Gottfredson. (1988). Decision making in criminal justice:
Toward a rational exercise of discretion New York: Plenum.
Draft Document: March 15, 2001 -14-
Grabosky, Peter N. (1984). The variability of punishment. In Toward a General Theory of Social
Control, vol. 1, edited by Donald Black. Orlando, FL: Academic Press.
Green, E. (1961).! Judicial attitudes in sentencing. London: MacMillian.
Greenberg, David (1977). The dynamics of oscillatory punishment processes. Journal of Criminal Law
and Criminology 68:643-651. From Zimring and Hawkins 1991.
Griset, P. L. (1995). The politics and economics of increased discretion over time served: a New Y ork
case study. Justice Quarterly 12:307-323.
Griset, P. L. (1996).! Discretion, disparity, and discrimination in sentencing: Where have all the critics
gone? Judges’ Journal 35:2-9
Hill, G. D., S.S. Harris and J. L. Miller (1985). The etiology of bias: Social huristics and rational decision
making in deviance process. J Res Crime 22:135-162. From Miller 1994
Hudzik, John K. (1984). Federal aide to criminal justice: Rhetoric, results, lessons. Washington,
DC: The National Criminal Justice Association.
Huemann, Milton (1978). Plea-bargaining: The experiences of prosecutors, judges, and defense
attorneys. Chicago: University of Chicago Press.
Joint Committee on New Y ork Drug Law Evaluation (1978). The nation’s toughest drug law:
Evaluating the New York experience. A project of the Association of the Bar of the City of
New Y ork and the Drug Abuse Counsil, Inc. Washington, D.C.: U.S. Government Printing
Office. From Tonry 1992.
Lagoy, Stephen P., Frederick A. Hussey and John H. Kramer (1978).1 A comparative assessment of
determinate sentencing in the four pioneer states. Crime and Delinquency 24:385-400.
Lansing, Sharon (1999). Youth discounts and sentencing equity in New York State. Paper presented
at the American Society of Criminology Annual Meeting, Toronto, Canada, November 18, 1999.
Loftin, Colin, Milton Heumann, and David McDowell (1983). Mandatory sentencing and firearms
violence: Evaluating an altemative to gun control. Law and Society Review 17:287-318. From
Tonry 1992.
Loftin, Colin, and David McDowell (1981). “One with a gun gets you two”: Mandatory sentencing and
firearms violence in Detroit. Annals of the American Academy of Political and Social Science
455:150, From Tonry 1992.
Luhman, Niklas (1985).1 Ecological communication. Translated by John Bednarz, Jr. Chicago:
University of Chicago Press. (Discusses feedback loops in institutional behavior.)
Newman, D. (1966).! Sociologists and the administration of justice. In Sociologists at Work, edited by
A. Shostak. Homewood, Illinois: Dorsey. 177-87 From Griset 1995
Draft Document: March 15, 2001 -15-
McAllister, William, James A tchinson and Nancy Jacobs (1991).1 A simulation model of pretrial felony
case processing: a queuing system analysis. Journal of Quantitative Criminology 7:291-314.
McCold, P.E. (1993). The role of fiscal policy in producing prison population dynamics: a trend
analysis and dynamic simulation of felony offender processing in New York State, 1975-
1988. Ph.D. dissertation. State University of New Y ork at Albany, Rockefeller College of
Public Affairs and Policy, School of Criminal Justice.
Miethe, Terrence D. and Charles A. Moore (1985). Socioeconomic disparities under determinate
sentencing system: a comparison of preguideline and postguideline practices in Minnesota.
Criminology 23:337-363.
Mieth, Terrence D. and Charles A.. Moore (1989).! Sentencing guidelines: Their effect in
Minnesota. Washington, D.C.: National Institute of Justice.
Miller, J. Langley and John J. Sloan, III (1994). A study of criminal justice discretion. Journal of
Criminal Justice 22:107-123.
Morgenthau, Robert (1984).! LEAA and prosecution. In Federal Aide to Criminal Justice: Rhetoric,
Results, Lessons, by John K. Hudzik. Washington, DC: The National Criminal Justice
Association. 281-292
Myers, Laura and Sue Titus Reid (1995). The importance of county context in the measurement of
sentence disparity: the search for routinization. Journal of Criminal Justice 23:223-241.
From Griset 1995
Rauma, David (1981a). A concluding note on the stability of punishment: Reply to Blumestein, Cohen,
Moitra and Nagin. Journal of Criminal Law and Criminology 72(4):1809-1812. From
Zimring and Hawkins 1991?.
Rauma, David. (1981b). Crime and punishment reconsidered: Some comments on Blumestein’s stability
of punishment hypothesis. Journal of Criminal Law and Criminology. 72(4):1772-1798.
From Zimring and Hawkins 1991?
Richardson, George P. 1991. Feedback thought in social science and systems theory. Philadelphia:
University of Pennsylvania Press.
Rossman, David, Paul Froyd, Glen L. Pierce, John MacDevitt, and William J. Bowers (1979).! The
impact of the mandatory gun law in Massachusetts Report to the National Institute of Law
Enforcement and Criminal Justice, Law Enforcement Assistance A dministration, U.S.
Department of Justice, Washington, D.C. From Tonry 1992.
Rausche, Georg and Kirchheimer, Otto (1939). Punishment and social structure. New Y ork:
Columbia University Press.
Sorokin, Pitirim (1937).! Social and cultural dynamics, vol.2. New York: American Book Company.
Spader, D. J. (1984). Rule of law vs. rule of man: The search for the golden zigzag between conflicting
Draft Document: March 15, 2001 -16-
fundamental values. Journal of Criminal Justice 12:379-394.
Stolzenberg, Lisa and Steward J. D’ Alessio (1994).! Sentencing and unwarranted disparity: An empirical
assessment of the long-term impact of sentencing guidelines in Minnesota. Criminology 32:301-
310.
Tonry, Michael (1987).! Sentencing guidelines and their effects. In The Sentencing Commission and
its Guidelines, edited by Andrew von Hirsh, Kay A. Knapp and Michael Tonry. Boston:
Northeaster University Press.
Tonry, Michael (1987b).! Sentencing reformimpacts. Washington: National Institute of Justice.
Tonry, M. (1988).1 Structuring sentencing. In Crime and Justice: A Review of Research, vol. _,
edited by Michael Tonry and Norval Morris. Chicago: University of Chicago Press..
Tonry, Michael (1992). Mandatory penalties. In Crime and Justice: A Review of Research, vol.16,
edited by Michael Tonry. Chicago: University of Chicago Press.
Tonry, Michael (1993). Sentencing commissions and their guidelines. In Crime and Justice: A Review
of Research, vol.17, edited by Michael Tonry. Chicago: University of Chicago Press.
Tonry, Michael and Joan Petersilia (1999). American prisons at the beginning of the twenty-first century.
In Crime and Justice: A Review of Research, vol.26, edited by Michael Tonry and Joan
Petersilia. Chicago: University of Chicago Press.
Triplett, R. and Miller, S.L. (1994).! Case processing in the Harris County, Texas criminal justice system:
a comparison across crime types. Journal of Criminal Justice 22:13-26.
U.S. Sentencing Commission (1991). Special report to the Congress: Mandatory minimum penalties
in the federal criminal justice system. Washington, D.C.: U.S. Sentencing Commission..
From Tonry 1992.
U.S. Sentencing Commission (199_). 1996 Sourcebook of Federal Sentencing Statistics. Washington,
D.C.: U.S. Sentencing Commission.
U.S. Sentencing Commission (199_). 1999 Sourcebook of Federal Sentencing Statistics. Washington,
D.C.: U.S. Sentencing Commission.
Utz. P. Determinate Sentencing in Two California Courts. Cited by J. Cohen and M. Tonry, Sentencing
reforms and their impacts, in A. Blumestein, ed., Research on Sentencing: The Search for
Reform ch.7. Washington: D.C.: National Academy Press, 1983. From Gottfredson et al
1988)
Vera Institute of Justice (1977).! Felony arrests: Their prosecution and disposition in New York
City’s courts. A Vera Institute of Justice Monograph. New Y ork, New Y ork.
Walker, S. (1992). Origins of the contemporary criminal justices paradigm: The American Bar
Association survey, 1953-1969. Justice Quarterly. 9:201-30. From Griset 1995
Draft Document: March 15, 2001 -17-
Walker, S. (1993).1 Taming the system The control of discretion in criminal justice 1950-1990.
New Y ork: Oxford University Press.
Waller, Irwin and Chan, Janet (1974). Prison use: a Canadian and international comparison. Criminal
Law Quarterly 17:47-71. From Zimring and Hawkins 1991.
Weibush, Richard G., Christopher Barid, Barry Krisberg, and David Onek (1995). Risk assessment and
classification for serious, violent, and chronic juvenile offenders. In A Sourcebook: Serious,
Violent & Chronic Juvenile Offenders, edited by James C. Howell, Barry Krisbery, J. David
Hawkins, and John J. Wilson. Thousand Oaks: Sage Publications.
Wooldredge, J. D. (1998). Analytical rigor in studies of disparities in criminal case processing. Journal
of Quantitative Criminology. 14:155-179.
Worden, A.P. (1995). The judge’s role in plea bargaining: an analysis of judges’ agreement with
prosecutors’ sentencing recommendations. Justice Quarterly 12:257-278.
Zatz, Majorie S. and Alan J. Lizotte (1985). The timing of court processing: towards linking theory and
method. Criminology 23:313-335.
Zimring, Franklin E. and Gordon Hawkins (1991). The scale of imprisonment. Chicago: University of
Chicago Press.
Zimring, Franklin E. and Jeffry Fagan (1999).
Draft Document: March 15, 2001 -18-
