Considerations in the Estimation of Impacts

As noted earlier, the impact of a new sentencing law on prison population depends on the extent to which sentencing under the new law would differ from current practices. The previous chapter took the first step necessary in the estimation of impacts by providing an empirical identification of how offenders are currently being sentenced. identification focuses on the impacts of a new law at a disaggregate level, i.e., for each particular group of offenders. The real impact of concern,

Such an

on the other hand, is the aggregate one, i.e., how many total offenders would be committed to state prison, and what total state prison population would result.

Clearly, the total size of the impact on state prison populations will depend on four factors:

1. The classes of offenders which a new sentencing law affects;
The number of offenders convicted in each of those classes;

2°

3.

4.

The proportion of those offenders who are currently receiving
sentences which are inconsistent with the new law;

The magnitude of the sentencing changes needed to achieve
conformity with the law.

Thus, for example, a law may specify sentences for a wide variety of offender
classifications (depending on their offenses, their prior records, etc.),
but if the total number of convicted offenders in those classifications

is small, then the total impact of the law will be small, regardless of how different the new sentences are from current practice. On the other hand, even if a new law specifies sentences which are very close to current sentences,

39.

if there is a large number of convicted offenders in that classification, then the number of new inmates in prison could be large.

Consequently,

then, the total impact of a new law is a summation of the impacts on individual subgroups of convicted offenders; each individual impact depends both on the size of the subgroup and the change in sentencing it would

undergo.

This, however, is simply an estimate of what would happen if the law were implemented precisely as it was stated. The extent to which this would actually take place, however, depends upon how the principal actors in the sentencing process judges, prosecutors, defense attorneys, and

respond to the law. A number of factors can be identified

which affect this response:

•The decision by prosecutors to bring offender characteristics
(such as firearm use and previous convictions) which are
specified in the legislation to the attention of the sentencing
judge;

The ability of prosecutors to prove the existence of those
characteristics when challenged by the offender;

•The willingness of judges to sentence in accordance with the law; and

•The willingness and ability of prosecutors to appeal cases where a judge has, in fact, not sentenced in accordance with the law.

These factors, in turn may be influenced by the size of the impacts and by capacity constraints within the criminal justice system. For example, if the expected impact of sentencing in accordance with the law would be an increase in prison populations beyond the capacity of the prison system,

As pointed out in Chapter II and in Appendix B, even the statement of some proposed legislation is not always very precise.

40.

then judges might be less willing to sentence in that fashion; if prosecutors were already overworked, they might be unwilling to appeal very many of the cases which are not sentenced in accordance with the law.

There are other factors that are not directly related to the sentencing process itself which might also affect the impact of a sentencing law. The criminal justice system may adapt in a variety of ways to reduce the impact of a law even though sentencing per se is completely in accordance with the law. For example, prosecutors might arrange for pre-trial diversion of offenders to some form of probation, and thus render moot the question of how they are sentenced. served by releasing inmates from their sentences earlier than in the past in order to reduce prison populations. These responses need not occur only with respect to those individuals who are affected by the legislation, but could be general adaptations in response to the pressures created by the new law.

Or the parole board might reduce the actual time

An example of the effects of diversion occurred in Pennsylvania between 1970 and 1975. After the introduction of the Accelerated Rehabilitative Disposition (ARD) program (which allowed defendants to be placed in a probationary status without being convicted) in 1972, the probability of a court case resulting in conviction decreased from about 60% to 43%, while the probability of a conviction resulting in a state prison sentence increased from 6% to 10%. Since, the probability of any court case resulting in a state prison sentence remained at about 4%, it appeared that those offenders who were now being diverted were those who would previously have been convicted but not sentenced to prison. The same process might be presumed to take place for those offenders for whom prison sentences would be required under a new sentencing law, but where the prosecutor considers a prison sentence inappropriate.

41.

The possibility

(or even the likelihood) of these various responses does not invalidate the usefulness of an estimation of impact which assumes both full compliance with the law and the constancy of other criminal justice system processing. The extent of resistance in the sentencing process and of adaptation in other parts of the criminal justice system is probably directly related to the size of the impact, and thus proposals which are projected to have large impacts if implemented are probably more likely to result in significant adaptation to reduce the impacts. The adaptation which does take place may have undesirable side effects, and in point of fact, might very well make the "cure" (1.e., the enactment of the new sentencing law) worse than the "illness". Since a sentencing law may serve a variety of goals reducing disparity in sentencing, insuring a minimum of punishment for the most serious offenders, etc. -and still have a relatively small impact on the prisons and on other parts of the criminal justice system, a law with a small potential impact in the latter sense is preferable to a similar law with large potential impacts. Thus, an impact estimate that assumes complete compliance and no adaptation can be of critical importance in assessing the feasibility of a particular sentencing proposal.

e.g.,

Methodology for State Prison Population Impact Estimation

Presuming that the majority of sentencing proposals which might be considered by the Legislature would be structured similarly to those discussed in Chapter II,