FORUM on Corrections Research

Intensive Supervision of Offenders on Prerelease Furlough: An Evaluation of the Vermont Experience

During the 1988 presidential campaign, a furor erupted over inmate furloughs when the Republican candidate, George Bush, accused his Democratic opponent of being "soft on crime." To prove his point, he cited the case of an inmate from the Democrat's home state who had committed a sexual assault and stabbing while on a furlough.

A national survey indicated that, in light of this publicity, several states restricted furloughs and three states discontinued them altogether. Nevertheless, the survey also reported high success rates for furlough programs.⁽³⁾Ironically, it was during this period that the Vermont Department of Corrections greatly expanded its prerelease furlough program.

As the term suggests, prerelease furloughs are authorized absences from correctional facilities for inmates who are nearing release. They allow offenders to prepare for life outside prison walls and adjust to more freedom. The program is similar to Canada's temporary absence (TA) program.

In Vermont, on any given day during the early and mid-1980s, inmates on furlough in the community constituted about 2% of the sentenced population, about 12 people. In 1990, this percentage climbed to about 10% of the sentenced population in Vermont (or 80 offenders).

The prerelease furlough program has become an important mechanism for relieving overcrowding in Vermont corrections facilities. In the spring of 1988, however, supervision of many inmates on prerelease furlough was greatly intensified. The aim of this study was to compare the effects of this intensified supervision with the effects of previous supervision practices on community safety, offender control and postrelease criminal activity.

Origin of the Program

Intensive supervision of inmates on prerelease furlough evolved from a house arrest initiative that began in 1988 in Chittenden county (in the northwestern part of Vermont). As a measure to relieve overcrowding, the house arrest program was designed as an alternative for misdemeanant offenders and non-violent felons with short sentences (maximum sentence of no more than 120 days). Corrections department officials select who is admitted to the program, and correctional officers monitor home confinement. Supervision includes several face-to-face contacts per week as well as random daily telephone checks. Alcohol and drug testing are also standard features of the program.

Before March 1988, supervision of inmates on prerelease furlough was almost non-existent compared with the house arrest program. The normal procedure for monitoring offenders on prerelease furlough - offenders who had served long sentences - included, at most, twice weekly face-to-face contacts or telephone checks.

The superintendent of the Chittenden correctional facility felt that house arrest resources should be used to monitor furloughed inmates. He noted that people under house arrest consisted of a few misdemeanants who were being intensively monitored while furloughed inmates, who had much more serious criminal histories, were being virtually ignored. He received permission from his superiors to have resources from the house arrest program diverted to supervise furloughed inmates as intensively as inmates under house arrest.

Thus, the intensive supervision of inmates on prerelease furlough began. The policy of supervising furloughed inmates spread to the southwestern part of the state in 1989. Now, almost all inmates on prerelease furlough in Vermont are intensively supervised. Department officials felt comfortable with such a policy in part because they worried that increased reliance on furloughs to control overcrowding would lead to the release of more dangerous people to the street.

Analysis

Our analysis examines two questions. First, are inmates under intensive supervision different from those monitored under previous furlough supervision practices? Second, are the outcomes different?

We examined two groups of inmates on prerelease furlough. One group consisted of 36 offenders who were furloughed from the Chittenden facility between January 1986 and April 1988. The furlough episodes for this group numbered 40 (an inmate can be furloughed more than once). Members of this group received minimal supervision. We labelled these offenders Group 1.

The other group, Group 2, was furloughed from the Chittenden facility between March 1988 and November 1991. They numbered 69, with the number of furlough episodes being 105. This group received intensive supervision.

There were no statistically significant differences in the average minimum sentence and the average maximum sentence of the two groups.

There was, however, an important difference between the two groups in their current offence type. Contrary to the prediction of some, those released under intensive supervision were not more dangerous.

Indeed, a significantly higher proportion of offenders in this group were non-violent offenders. Our conclusion is that the pressure of crowding encouraged officials to furlough more offenders who were eligible according to department policy.

Our analysis also showed that the population furloughed under intensive supervision was at no greater risk to reoffend than past populations, if felony record is an indication. There was no statistically significant difference in the proportion of offenders with a felony history. Again, we conclude that more offenders who were eligible were furloughed.

Finally, there was no significant difference in the average length of each furlough episode.

Figure 1 shows a comparison between the outcomes of the two groups' furloughs. While one in five furlough episodes ended in revocation under previous supervision practices, about one in two were revoked under intensive supervision. In the case of the former, all revocations were for violations of furlough conditions. Regarding the latter, 2 revocations were for arrests, and 3 furloughs were revoked for criminal actions for which there was a subsequent conviction; the other 48 revocations were for violations of conditions.

Figures 2 and 3 compare the therapeutic or rehabilitative effects of intensive supervision with those of previous practices. Offenders in each group were tracked for one year after release from prison. Figure 2 shows that 15% of offenders in Group 1 and 19.1% of those in Group 2 were subsequently convicted of another crime. This difference was not statistically significant.

Moreover, Figure 3 indicates that, of those who were subsequently reconvicted, individuals in neither group were more likely to be violent than those in the other. Of offenders in Group 1, one in six (16.7%) subsequently committed a violent crime, while 4 in 17 (23.5%) of offenders in Group 2 were convicted of a violent act. Again, the difference was not statistically significant.

---

Figure 1

---

Figure 2

---

Figure 3

Conclusion

The results of this study cast some doubt on the usefulness of intensive supervision for offenders on prerelease furlough. Though offender control is improved, it does not seem to affect public safety.

Whether the benefits gained in offender control are worth the costs in resources and further prison crowding (due to returning offenders) is a much discussed subject among practitioners and academics.⁽⁴⁾It costs about US$8,000 per year to supervise intensively each offender on furlough. Lacking budget estimates for previous practices, we can assume that the costs were minimal. Hence, a return to such practices could probably save money.

Nevertheless, there are reasons for maintaining intensive supervision. In an era of hardening public attitudes toward crime, corrections professionals have stressed surveillance and de-emphasized service brokerage for offenders in the community.⁽⁵⁾With an increasing number of offenders on furlough, corrections officials may want to maintain a public surveillance presence in the interest of good public relations.

In addition, Vermont is one of the few jurisdictions in the United States never to have been under court order to release inmates because of overcrowding. It is quite clear that the present furlough program has helped in this regard. Moreover, the delicate policy decisions regarding whom to release are still the jurisdiction of elected and appointed officials rather than the judiciary. This situation provides for a high degree of public accountability.

It is likely that events will eventually cause changes in practices, however. The offender population is projected to grow well beyond prison capacity. Eventually, street supervision resources may be insufficient to continue intensive surveillance. At that point, supervision practices may necessarily become less intensive yet, as this research suggests, more appropriate for monitoring offenders on the street.

---

⁽¹⁾William Bagdon, Research Associate, Agency of Human Services, Vermont Department of Corrections, 103 South Main Street, Waterbury, Vermont 05671-1001, United States.
⁽²⁾James E. Ryan, Senior Research and Statistics Analyst, Agency of Human Services, Vermont Department of Corrections, 103 South Main Street, Waterbury, Vermont 05671-1001, United States.
⁽³⁾Su Perk Davis, "Number of Furloughs Increasing - Success Rates High,"Corrections Compendium,16(1991): 10-21.
⁽⁴⁾James M. Byrne, Arthur Luguno and Christopher Baird, "The Effectiveness of the New Intensive Supervision Programs,"Research in Corrections,2 (1989): 1-48.
⁽⁵⁾James M. Byrne, "Reintegrating the Concept of Community into Community-Based Corrections,"Crime and Delinquency,35(1989): 471-499.