Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Legislative Budget and Finance Committee

VIEWS: 52 PAGES: 119

									                     Legislative Budget and Finance Committee
                           A JOINT COMMITTEE OF THE PENNSYLVANIA GENERAL ASSEMBLY
                                      Offices: Room 400 Finance Building, 613 North Street, Harrisburg
                                       Mailing Address: P.O. Box 8737, Harrisburg, PA 17105-8737
                                Tel: (717) 783-1600 • Fax: (717) 787-5487 • Web: http://lbfc.legis.state.pa.us




SENATORS
JOHN R. PIPPY
  Chairman
JAY COSTA, JR.
WAYNE D. FONTANA
ROBERT B. MENSCH
DOMINIC PILEGGI
JOHN N. WOZNIAK

                           Statewide Complement Levels
REPRESENTATIVES
                          of Probation and Parole Officers
ROBERT W. GODSHALL
  Secretary
DAVID K. LEVDANSKY
  Treasurer
STEPHEN E. BARRAR
JIM CHRISTIANA
H. SCOTT CONKLIN
ANTHONY M. DELUCA




                     Conducted Pursuant to Senate Resolution 2008-367
EXECUTIVE DIRECTOR
PHILIP R. DURGIN




                                                                                                                 October 2009
                                             Table of Contents
                                                                                                                         Page

       Summary and Recommendations .................................................... S-1
I.     Introduction ........................................................................................              1

II.    Findings ..............................................................................................            4
       A. The PBPP Has Shifted From a Law Enforcement to an Evidence-
          based Re-entry Approach as a Way to Help Offenders Transition to
          the Community ........................................................................................          4
       B. Field Agent Complement Is Based on the Offender Population and Its
          Levels of Supervision; However, Actual Staffing Varies Considerably
          Depending on District Office Turnover and Vacancies ............................                               10
       C. Turnover and Vacancies Are Key Factors Generating Caseload/Work-
          load Inequities Between and Among Supervision Units ..........................                                 24
       D. The Standards for Offender Supervision and Frequency of Contacts
          Are Uniform Among District Offices but Staffing Levels Affect Actual
          Practice ...................................................................................................   46
       E. The PBPP Is Acting to Address Issues With Hiring and Retention of
          Field Agents Affecting the Philadelphia District Office.............................                           57
       F. The Board Has Identified Technological Needs for the Parole
          Supervision Function ...............................................................................           72
III.   Background ........................................................................................               82

IV.    Appendices .........................................................................................              93
       A. House Resolution 2008-367 ....................................................................                 94
       B. Pending Legislation Related to PBPP .....................................................                      96
       C. District Office Organization Charts ..........................................................                 97
       D. Caseload and Workload Distribution September 3, 2009 ........................ 103
       E. Response to This Report......................................................................... 105




                                                             i
Summary and Recommendations
       Senate Resolution 2008-367 directs the Legislative Budget and Finance
Committee (LB&FC) to study the statewide complement of probation and parole of-
ficers. The resolution specifically directs us to identify possible inequities in fund-
ing distribution, caseload assignment, hiring, training and retention, required levels
of offender supervision, and frequency of contacts.

      The Pennsylvania Board of Probation and Parole is an independent agency of
the Commonwealth created to oversee the Commonwealth’s probation and parole
laws. The Board has exclusive power regarding all parole matters as to any person
sentenced by any court to a maximum term of two years or more in a state or county
prison. The Board also supervises any person placed on probation when the court
by special order directs such supervision. Additionally, the Board supervises of-
fenders sentenced by other states pursuant to the Interstate Compact for Adult Of-
fender Supervision.

       As of June 30, 2009, the PBPP was responsible for supervising 31,179 offend-
ers with a complement of 505 parole agents. Supervision of offenders in the com-
munity is the responsibility of the field agents working in district offices and sub-
offices. The Philadelphia and Chester District Offices comprise the Eastern region;
the Allentown, Harrisburg, Scranton, and Williamsport District Offices comprise
the Central region; and the Erie, Pittsburgh, Mercer, and Altoona District Offices
comprise the Western region. There are also 15 sub-offices throughout the state.
See Exhibit 1 on page 5 for a map indicating the locations of these offices.

      We found:

The PBPP Assigns Field Agent Positions Equitably, but Due to Turnover
and Other Factors, Actual Workloads Vary Considerably Between Districts
and Between Units Within the Districts

        PBPP assigns parole agent positions to district offices based on the expected
state offender population and levels of supervision for those offenders for the up-
coming fiscal year. Specifically, PBPP uses a workload model that assigns a time
value to the requisite supervision activities for the various levels of supervision as-
signed to the offenders. Under this model, parole agents are considered to have a
full workload when they reach 131 hours of work per month. The Board notes that
due to temporary work assignments, vacancies and leave (military, long-term sick,
etc.), a difference of up to 10 percent (145 hours) is a temporarily manageable work-
load capacity. Thus, the PBPP’s workload formula results in an average caseload
target of 66 offenders per field agent, acceptable up to 73.



                                          S-1
       While the workload model provides an objective and uniform method for allo-
cating agent positions to the various district offices, it does not necessarily result in
uniform workloads in practice. As shown below, in four of the 10 PBPP districts
(Chester, Harrisburg, Philadelphia, and Mercer), over 40 percent of the active field
agents have workloads above the 131 hour threshold established by the PBPP.
When calculated using 145 hours, over 19 percent of the active field agents in these
four districts (up to 58 percent in Mercer) exceeded the PBPP workload capacity.


                                     Caseload and Workload Distribution
                                                 (July 2, 2009)

                                     # of     Field Agents          % of       Field Agents With         % of
      District Office               Field    With Caseloads         Field          Workloads             Field
                                   Agentsa   100 or Greater        Agents      Greater Than 131b        Agents
  Allentown ................         52              0              0.0%                9                17.3%
  Altoona ...................        21              0              0.0                 5                23.8
  Chester ...................        33              0              0.0                15                45.5
  Erie .........................     23              0              0.0                 4                17.4
  Harrisburg ...............         56              1              1.8                25                44.6
  Mercerc ...................        36            10              27.8                29                80.6
  Philadelphiad...........          114            14              12.3                53                46.5
  Pittsburgh ...............         59              1              1.7                19                32.2
  Scranton .................         29              0              0.0                 4                13.8
  Williamsport ............          13              0              0.0                 3                23.1
    Total.....................      436            26               6.0%              166                38.1%
_______________
aFilled field agent positions carrying a caseload. In general, this number does not include agents on extended medi-
cal or military leave and includes agents currently at BTA.
bThese workloads include only hours used for offender supervision. Actual workload hours would likely be higher,
resulting in more agents with workloads over 131, and include other duties such as Board of Pardons investigations.
cMercer and Venango Counties do not have county adult probation departments, therefore, the district office provides
all adult probation services. Many of the county cases are of relatively short duration.
dThe agent count does not include the FAST Unit. That unit has four agents and each has a caseload of 100 or
greater and a workload greater than 131.

Source: Pennsylvania Board of Probation and Parole.



       Mercer has the highest percentage of parole agents with workloads greater
than 131, but Mercer is unique in that it provides probation services for both state
and county offenders in Mercer and Venango Counties. Although high workloads
exist in the Chester and Harrisburg Districts, particular concern has been directed
at Philadelphia because it is by far the largest district and in recent years has had
high percentages of offenders at the enhanced and maximum supervision levels.

       Moreover, as the table below shows, the difference between the complement
level and active agents can be substantial. This has been a particular problem in


                                                       S-2
                                                           Disposition of PBPP Field Agents*

                                                                                                                      Statewide Total
                                                                                                               2009    2008    2007     2006
                           Complement .........................................................                 505     485     415      411
                           ASCRAs.................................................................               16      16       0        0
                           Non-ASCRA Complement ...................................                             489     469     415      411
                           Vacancies ..............................................................              13      30      22        9
                                                                  a
                           Not Available/On Leave ........................................                       18      17       7        5
                                                                             b
                           Re-Assignment or Suspension .............................                              0       3       3        3
                                             b, c
                           Paid Leave               ........................................................      3       3       2        2
                           Field Office Adjusted Active Agent Total...........                                  455     416     381      392
                           Filled Wage Positionsd...........................................                     13     N/A     N/A     N/A
_______________
*As of June 30.
a
 This category includes staff who are out on leave without pay for military or medical reasons or staff out on Worker’s
Compensation Act or Enforcement Officer Disability Benefits Law (Heart and Lung Act) leave.
b For June 30, 2006, 2007, and 2008, this number was estimated by taking the average of the FY 2008-09 number
for each district office and applying that percentage to calculate a number for each district each year.
cPaid leave includes paid time off such as vacation and sick leave that is usually scheduled or of brief duration.
dThis category reflects the use of annuitants.

Source: PA Board of Probation and Parole.


the Philadelphia District Office (discussed below). Since January 2009, the PBPP
has hired 36 field agents to fill statewide complement vacancies with 20 of those
agents being assigned to Philadelphia.

The PBPP Has Moved From a Law Enforcement Approach to Supervising
Offenders to an Evidence-based Re-entry Approach, Which May Require
Additional Resources

       We also found that the current workload model used to determine staffing
needs was developed in 2002, prior to the PBPP’s implementation of the evidence-
based re-entry approach to supervision. 1 The evidence-based re-entry approach,
which the PBPP adopted in 2006, changed the focus of supervision from a law en-
forcement model to one that seeks to change offender behavior to enable the offend-
er to function in society. A comparison of the law enforcement and evidence-based
re-entry parole models is shown below.


                                                            
1Evidence-based   practices are changes to the culture and core supervision practices and procedures of the agen-
cy and are integrated into all re-entry services. Re-entry refers to the services the Board provides to help an
offender transition from prison to the community to successfully complete parole. Throughout the report we use
the term “evidence-based re-entry” when referring to these practices.

                                                                                           S-3
Comparison of Law Enforcement and Evidence-based Re-entry Parole Models

            Law Enforcement Approach                              Evidence-based Re-entry Approach
Monitors compliance with the conditions of parole.         Monitors compliance with the conditions of parole.

                                                           Engages the offender in efforts to change behavior
                                                           and reduce risk through:
Brings violations of those conditions to the attention     -staff/offender interactions designed to enhance
of the authorities.                                          motivation for change; and
                                                           -providing interventions/programs that address the
                                                            specific risks and needs of the offender.

                                                           Responds to violations by holding the offender re-
                                                           sponsible but uses a proportional and problem-
Responds to violations with sanctions including rein-
                                                           solving sanction/response. This would include rein-
carceration.
                                                           carceration in the case of very serious violations, or
                                                           where public safety was in danger.



       One of the expected effects of the evidence-based re-entry approach to super-
vision is a decrease in the number of technical parole violations (TPV) for which of-
fenders are reincarcerated since evidence-based re-entry emphasizes alternatives to
reincarceration. Formal Board action to re-incarcerate for a TPV has decreased in
overall number and as a percentage of the state sentenced population in the last
four fiscal years. The number of criminal parole violations (CPV) that resulted in
formal Board action increased in the overall number and percentage of the state
sentenced population. In her FY 2009-10 Budget Presentation before the House
Appropriations Committee, the PBPP Chairman noted that Pennsylvania’s one-
year recidivism rate has decreased from 28 percent in 2003 to 21 percent by 2007,
and the three-year rate of 49 percent is below the national average of 50 percent.

      The additional requirements associated with the evidence-based re-entry fo-
cus have reportedly increased the time necessary for agents to adequately supervise
offenders. Although additional contacts with the offenders may not be required, the
time needed for those contacts has increased due to the more complex issues field
agents must address, e.g., assisting the offender with employment needs.

       Although the PBPP has made efforts to offset the additional time needed for
re-entry supervision with other time efficiencies, no study has been done to verify
that the overall time requirements have remained the same for the various levels of
supervision.




                                                         S-4
The PBPP Uses Standard Screening Tools to Establish an Offender’s Level
of Supervision, but High Caseloads and Workloads Affect the Ability of
Field Agents to Comply With the PBPP’s Supervision Standards

       The PBPP uses standard screening tools to determine the likelihood an of-
fender will re-offend and the level of supervision applicable to the offender. Similar-
ly, the standards that detail parole agent activities, such as the minimum number
of offender contacts, are also uniform throughout the state. For example, a field
agent supervising an offender at the maximum level of supervision is to have six
face-to-face contacts per quarter with that offender and two collateral contacts per
month.

       The actual supervision an offender receives, however, may be less than called
for by the standards. PBPP internal audits we reviewed identified instances where
various supervision standards were not met by the field agents. These include both
administrative, e.g., failure to record start and stop times for contacts, and opera-
tional, e.g., insufficient offender contacts, deficiencies. In many of these instances,
the inspection reports note that staffing constraints, typically due to vacancies and
leave situations, have created high caseload and workload levels, making it difficult
for agents to meet the supervision standards. As noted above, the PBPP seeks to
maintain an average workload of 131 hours per month per field agent and a result-
ing caseload average of 66 offenders per field agent. As shown in the table below,
the PBPP audits found that these caseload and workload goals have often been ex-
ceeded.

                                     Examples of Audit Findings

                                      Active   Average    Caseload    Average     Workload
   District           Unit            Agents   Caseload    Range      Workload      Range
Chester        Delaware Unit            8         75       55 - 93      109       82.3 - 139.8
Philadelphia   Northeast Unit 6         5         87       46 - 127     122       61.3 - 188.1
Philadelphia   West Unit 2              4        124       92 - 138     169      124.9 - 189.1
Harrisburg     Dauphin Unit 4           6         90       21 - 144     144       23.2 - 193.5
Williamsport   Williamsport Unit 1      8         61       17 - 76      104       27.9 - 138.5
Erie           Erie East                7         69       42 - 86      119       91.6 - 141.0
Pittsburgh     Mon Valley South         4         95       94 - 98      152      140.8 - 160.0



Field Agent Turnover Has Been Particularly High in the Philadelphia District
Office

       The statewide average adjusted turnover of parole agents in the district
offices was 11.5 percent in FY 2007-08 and 11.7 percent in FY 2008-09 compared to
about 26 percent and 16 percent, respectively, in the Philadelphia District Office.
These high turnover rates have resulted in higher caseloads for the agents who

                                                  S-5
remain in the Philadelphia District Office during the time the positions remain va-
cant and while new agents are being trained.

       High turnover has also resulted in the parole agents in the Philadelphia Dis-
trict Office having lower median years of experience than the other district offices.
At the end of 2008, the Philadelphia District Office had three supervision units
where at least half of the field agents had one year or less experience in the posi-
tion. Only one other unit statewide was similarly staffed.

       Medical and military leave status also affects the number of active field
agents and thus the caseload and workload levels. At the end of FY 2008-09, 18
agents statewide were on extended medical or military leave. Half of those were
from the Philadelphia District Office. While an employee is on this type of leave,
the PBPP does not have a vacancy it can fill, since the employee will be returning to
the position, so the caseload and resulting workload of that employee must be dis-
tributed to other field agents, or, in some cases, covered by the unit supervisor.

The PBPP Is Making Efforts to Address Turnover in the Philadelphia Dis-
trict Office

       The PBPP initiated several actions to address the high turnover rate in the
Philadelphia District Office. Several years ago, the Board implemented an em-
ployment requirement that generally restricts transfers from the seniority unit dur-
ing the first two years of employment as a field agent. With few exceptions, an
agent is required to remain in his seniority unit (Philadelphia and Chester District
Offices are one seniority unit, and every other District Office is an individual senior-
ity unit) for two years.

       In September 2008, the Board also implemented a retention incentive pro-
gram for agents in the Philadelphia District Office that provides an additional two
step increment in pay after two years of continuous service in the district office.
Another two step increment is available after four years of continuous service in the
district. These increases are in addition to all other longevity increases and general
pay increases and are applicable to incumbent agents in the district office. If an
agent transfers out of the Philadelphia District Office, any increments awarded un-
der the incentive program will be taken away. An experienced agent voluntarily
transferring into the Philadelphia District Office will immediately get the applica-
ble increments depending on his or her experience.

      The Board is also using annuitants in several districts and is seeking to hire
wage employees who will be able to step into full-time salaried positions as field
agents as soon as a position becomes vacant.




                                          S-6
The PBPP Has Considered Technology Improvements for Offender Super-
vision but Has Encountered Implementation Difficulties

       The Board has been pursuing automation and technological enhancements as
ways to enhance the Board’s capacity to track and supervise offenders. These ef-
forts include:

•   Developing, in cooperation with the Pennsylvania Department of Corrections, an
    Integrated Offender Case Management System (IOCMS). The IOCMS will large-
    ly replace existing offender management systems. When fully implemented, the
    IOCMS will ensure that all offender-based information from both agencies is
    available for management purposes in one integrated system. The IOCMS
    project has had significant delays, but final implementation is anticipated by
    2013.
•   Exploring the use of electronic devices to automate parole agent data collection
    and reporting. Currently, PBPP’s parole agents record all contacts with offend-
    ers on paper records, and the results of this manual recordkeeping process are
    not linked with offender history in a central database. Since spring 2009, PBPP
    has been investigating the Blackberry®, an electronic mobile communication de-
    vice, for possible use by agents in recording and reporting contact information.
    However, the two pilot projects testing the application of this technology did not
    produce satisfactory results. Due to the unsuitability of the Blackberry, the
    PBPP reports it has terminated this research and is now planning a pilot study
    using 30 of the new “Netbook” class of portable computers, to determine their ef-
    fectiveness to improve agent productivity and provide real time data.
•   Expanding the use of electronic monitoring. Since 2005, the PBPP has been con-
    sidering the use of Global Positioning Systems (GPS) to track offenders. GPS
    uses orbiting satellites to pinpoint the physical location of objects or persons.
    The Board currently uses electronic monitoring with radio frequency equipment
    as appropriate, and at any given time, approximately 300 electronic monitoring
    devices are in use around the state.

    The PBPP has conducted two pilot studies on GPS and has concluded that GPS
    hardware and software, despite some improvements, still had limited functional-
    ity and reliability. The PBPP intends to regularly review GPS technology’s
    progress to determine whether, at a future time, GPS systems can be successful-
    ly deployed.




                                          S-7
                               Recommendations:

1. The PBPP should recalculate the hours needed for supervision to reflect the
   re-focusing of probation and parole from enforcement to evidence-based re-
   entry practices.

   The length of time needed by a field agent to supervise an offender was calcu-
   lated several years ago, prior to the change to an evidence-based re-entry focus.
   As a result, it may not provide adequate time for agent activities, including, for
   example, implementing community alternatives to reincarceration, particularly
   in regions of the state where social services are scarce.

2. PBPP audit plans should include steps to check that the corrective steps iden-
   tified in response to annual audit reports are actually implemented.

   If an annual audit identifies deficiencies, the district director is to submit a cor-
   rective action plan. However, the implementation of that plan was not reviewed
   as part of the following year’s audit. We recommend that the PBPP continue
   and complete this process to ensure that each prior year’s corrective action plan
   was actually implemented. In the past six months the PBPP has acknowledged
   this deficit and is reportedly implementing an audit process that closes this loop.

3. The PBPP central office should develop trend information to better manage
   caseload and workloads.

   The PBPP maintains daily information on active field agent caseloads and work-
   loads. But it can be difficult to identify developing trends using daily reports,
   especially at the central office level. We recommend the daily reports of individ-
   ual agents be plotted over time or otherwise aggregated to better highlight de-
   veloping trends both at the individual agent and unit level.

4. The PBPP has also taken several steps that we recommend be continued, in-
   cluding:

   − Proceed with its plan to hire wage employees, as the budget allows, in the
     Philadelphia District Office and consider the use of wage employees in other
     district offices should turnover and the subsequent difficulty in hiring to fill
     the vacancies in a timely manner, or extended leave situations, become an is-
     sue.
   − Continue to monitor the field agent turnover rate in the Philadelphia District
     Office, especially transfers to other district offices, to determine whether the re-
     tention incentive is reducing the field agent turnover as intended.
   − Continue with its plan to periodically review the status of GPS technology to
     determine its applicability and potential effectiveness for offender supervision.

                                          S-8
   The PBPP should contact the Pennsylvania counties and other states currently
   using GPS to learn how they may have addressed the weaknesses noted in the
   Board’s pilot projects. The PBPP should report on the results of its studies
   and make recommendations regarding any statutory amendments necessary to
   implement GPS for this function, as well as the potential cost, to the General
   Assembly.
− Continue to study the use of other technology, e.g., netbooks, to assist the field
  agents with timely access to information. Any technology identified for use
  should be implemented with future IOCMS functions in mind.




                                       S-9
S-10
I. Introduction
 

       Senate Resolution 2008-367 directs the Legislative Budget and Finance
Committee (LB&FC) to study the statewide complement of probation and parole of-
ficers and identify relevant inequities in funding distribution, caseload assignment,
hiring, training and retention, required levels of offender supervision, and frequen-
cy of contacts. Please see Appendix A for a copy of Senate Resolution 2008-367.
 
                                    Study Objectives
 
    The study seeks to identify possible relevant inequities at the regional and dis-
trict level in:

       •   funding distribution;
       •   caseload assignment;
       •   hiring;
       •   training and retention;
       •   required levels of offender supervision; and
       •   frequency of contacts.

                             Scope and Methodology

       We met with staff of the Pennsylvania Board of Probation and Parole (PBPP),
regional directors, district office directors, and current and former field agents. We
reviewed PBPP policies and procedures related to the determination of level of su-
pervision and the requirements for each level of supervision. To evaluate workload
and caseload distribution among the district offices, we reviewed the Board’s
Monthly Program Reports from July 2001 to August 2009.

       To identify relevant inequities in funding distribution, we met with key
PBPP staff in the PBPP’s Office of Administrative Services and with regional and
district office directors in selected areas of the state. We also reviewed agency
budget requests and appropriations hearings presentations and other financial and
statistical information.

        To evaluate the consistency of implementation of Board policies and proce-
dures related to level of supervision, we reviewed the most recent Bureau of Offend-
er Re-entry Coordination audits of field supervision units for the Philadelphia Dis-
trict Office and a sampling of the audits for the remaining district offices. In addi-
tion, we reviewed case files and field books for a sample of offenders in several dis-
tricts that we visited. Due to Criminal History Record Information Act (CHRIA)
restrictions, all case files and field books were redacted by the Board prior to our
                                           1
review. The redaction was restricted to the names and certain identifying informa-
tion of the offenders and certain staff; however, due to redaction needs, we did not
independently select the files for review.

       To determine the reasonableness of caseload and workload standards estab-
lished by the Board, we reviewed reports by the American Probation and Parole As-
sociation (APPA), the American Correctional Association (ACA), and the Association
of Paroling Authorities International (APAI). We also sent an electronic question-
naire to the contiguous states, Michigan, and Georgia to provide recent comparative
information and spoke with nationally recognized experts in the field of probation
and parole about matters under review and for a general understanding of the sub-
ject matter and national trends. We also reviewed recent reports on the PBPP re-
leased by the Pennsylvania Department of Auditor General.

       On September 29, 2008, after a police officer in Philadelphia was killed by a
paroled felon, the Governor suspended the parole release of all offenders pending an
independent review of the Department of Corrections and PBPP’s procedures re-
garding the parole preparation and release. Information contained in this report
reflects, in part, the impact of this suspension, and we identify that impact as it re-
lates to specific discussions in the report. Parole for non-violent offenders was
reinstated on October 20, 2008, and for violent offenders on December 1, 2008.

       Although the parole of violent offenders, and the re-offending by those of-
fenders, was under review at the time of our study, the actions of the PBPP in pa-
roling these individuals was not within the scope of this study. As such, we did not
review or comment on that process. We did, however, review the reports released
from that review and have noted certain of the reports’ conclusions and recommen-
dations as appropriate to our study. Additionally, since the focus of the resolution
was on the supervision of offenders function, we did not review the activities of the
PBPP staff located in corrections facilities.

                                Acknowledgements
 
       We thank Catherine McVey; Chairman of the Board of Probation and Parole,
and her staff, in particular, our liaison Richard Dash, Director of the Office of Ad-
ministrative Services, for their assistance with this study. In addition we thank the
regional directors, district directors, and field agents we met with for their time and
insights, which added to our understanding of the supervision function. We also
thank Peggy Burke, Center for Effective Public Policy; William D. Burrell, Correc-
tions Management Consultant; and Dr. John S. Goldkamp, Ph.D., Chair of the
Criminal Justice Department at Temple University, for their assistance with this
study.




                                           2
                                  Important Note

        This report was developed by Legislative Budget and Finance Committee staff.
The release of this report should not be construed as an indication that the Commit-
tee or its individual members necessarily concur with the report’s findings and rec-
ommendations.

       Any questions or comments regarding the contents of this report should be di-
rected to Philip R. Durgin, Executive Director, Legislative Budget and Finance
Committee, P.O. Box 8737, Harrisburg, Pennsylvania 17105-8737.




                                          3
II. Findings

A. The PBPP Has Shifted From a Law Enforcement to an Evidence-
based Re-entry Approach as a Way to Help Offenders Transition to
the Community

       The Pennsylvania Board of Probation and Parole (PBPP) is an independent
agency of the Commonwealth created to oversee the Commonwealth’s probation and
parole laws. 1 The Board has exclusive power regarding all parole matters as to any
person sentenced by any court to a maximum term of two years or more in a state or
county prison. Additionally, the Board is to supervise any person placed on proba-
tion when the court, by special order, directs such supervision. The Board provides
all adult probation and parole services in Mercer and Venango Counties since those
counties do not have adult probation departments. The Board also supervises of-
fenders sentenced by other states where a request is accepted pursuant to the Inter-
state Compact for Adult Offender Supervision. Please see Appendix B for pending
legislation related to the PBPP.

       Parole is the release of certain inmates from incarceration to continue serving
their sentences in the community under varying degrees of supervision. Inmates
sentenced to incarceration for more than two years become eligible for parole con-
sideration after serving the minimum sentence imposed by the sentencing court.
Probation is a court ordered sanction that allows a person to remain in the commu-
nity under the supervision of a probation officer. Failure of the offender to comply
with parole or probation requirements may result in the individual being incarce-
rated.

       The Board of Probation and Parole maintains a central office in Harrisburg
and 52 offices statewide (10 district offices, 15 sub-offices, and 27 institutional offic-
es). The offices are divided into three geographical regions: Eastern, Central, and
Western. The Philadelphia and Chester areas comprise the Eastern region; Allen-
town, Harrisburg, Scranton, and Williamsport areas comprise the Central region;
and Erie, Pittsburgh, Mercer, and Altoona comprise the Western region. The 15
sub-offices are located throughout the state and the 27 institutional offices are lo-
cated in each state prison facility. See Exhibit 1 for a map indicating the locations
of these offices.

       Supervision of parolees in the community is the responsibility of the field
agents who work out of the district offices and sub-offices. The district offices are
also responsible for conducting parole investigations (e.g., pardon investigations,

1The Board is one of fourteen adult probation and parole field services agencies with accreditation status (au-

dited and found to be in compliance with ACA standards) by the American Correctional Association (ACA). This
includes 10 state agencies and 4 county agencies.

                                                      4
                                                                                                                                                        Exhibit 1
                                                                                                                State Probation Regions and Offices

              Erie DO

                       Erie

                          SCI Albion
                                                                                                                                                                               Tioga                         Bradford
                                                                                                                                                Potter
                                                               Warren                          Mckean                                                                                                                                                  Susquehanna
                SCI Cambridge Springs                                                                                                                                                                                                                                                    SCI Waymart


                                                                                                                                                                               Lycoming
       Crawford                                                                                                                                                                                                                                                                      Wayne
                                                                                                                                                                                                                                                                    Lacka-
                                                                                                                                                                                                             Sullivan                           Wyoming             wanna
                                                                                                                                                                      Williamsport DO       SCI Muncy
                                                                                                                                                                                                                                                                            Scranton DO
                                                       Forest
                                Venango                              SCI Forest
                                                                                                                                                            Clinton
         Mercer                                                                                                Cameron                                                                                                                                                                            Pike
                                                                                                                                                                                                                                                       SCI Dallas
         Mercer DO
                                       Franklin SO                                            Elk
                                                                                                                                                                                                                                                         SCI Retreat
                                                                                                                                                         SCI Rockview                                 Montour
                                                                                                                                                                                        Union
                                                                                                                                                                                                                        Columbia                                                  Monroe
               SRCF Mercer
                                                      Clarion                                                                             Centre                                                                                                Luzerne
                                                                                                                                                                                                    Northum-
                                                                                                                                                                                                     berland
     Lawrence                 Butler                                                                                                                                                Snyder                                                                      Carbon
                                                                                                                                                                                                    SCI Coal Twp.

                                                                                         Jefferson             Quehanna Boot Camp
                                  Butler SO
5




                                                   Armstrong                                                                                                                                                                       SCI Mahanoy
                                                                                                                                                                                                                                                                            Northamp-
                                                                                                                   Clearfield                                                                                                       Schuylkill                                 ton
                                                                                                                                                                                                                                                                Lehigh
                                                                                                             SCI Houtzdale
                                                                                                                                                                                                                             SCI Frackville                                    Allentown DO
       Beaver                                                                             Indiana                                                                                                                                                   Berks
                                                                                                                                                                 Mifflin
     Beaver Falls SO                                                                                                                   SCI Smithfield
                         Allegheny                                                                        SCI Cresson
                                                                                                                                                                                                                                                  ReadingSO                                   Bucks
                                                                                  SCI Pine Grove
         North Shore SO
                                Pittsburgh DO                                                                                                                                  Juniata
                                                                                                       Cambria                         SCI Huntingdon
                                                                                                                                                                                                       Dauphin
        SCI Pittsburgh                                                                                                    Altoona DO

                                              Westmoreland                                                                   Blair
             Mon-Valley SO                                    Greensburg SO                                                             Huntingdon                                                                     Lebanon
                                                                                                                                                                                    Perry
                                          SCI Greensburg                                                                                                                                                                                                                     Montgomery
                                                                                                                                                                                                             Harrisburg DO
        Washington                                                                                                                                                                  SCI Camp Hill                                                               SCI Graterford
                                                                                                   SCI Laurel Highlands                                                                                                           Lancaster
                                                                                                                                                                                                                                                                                                               PHL CW
                                        SCI Fayette                                                                                                                         Cumberland                                                                                 Eagleville SAVE                         PHL NE
                                                                                                                                                                                                                                 Lancaster SO                                                                  PHL NW
                                                                                                                                                                                                                                                                                                              PHL West
                                                                                                                          Bedford
          Greene                                                                        SCI Somerset                                   Fulton                                                             York SO
                                          Fayette                                                                                                        Chambersburg SO                                                     Central
                                                                                                                                                                                                                                                       Chester DO
                                                                                                                                                                                                                             Admin.                                                                        Philadelphia DO
                                                                                        Somerset                                                                                                                                                                                                         Philadelphia
                       SCI Greene                                                                                                                                                       Adams                                                                       SCI Chester
                                                                                                                                                            Franklin                                                                                                                     Delaware
                                                                                                                                                                                                            York                                        Chester



                                       PBPP District Office                                                                                                PBPP Sub-Office                                                     State Correctional Institution



    Source: PA Board of Probation and Parole.
pre-sentence investigations, and home plan investigations) and tracking abscond-
ers. The regional office has operational control and responsibility for all field staff
within their region. Additionally, the regional director is responsible for supervis-
ing all parole staff assigned to the state correctional institutions and local county
prisons within the region. 2

Organization of the District Offices
       A district office consists of a director; secretarial staff; one or more Assess-
ment, Sanctioning and Community Resource Agents (ASCRAs); and field agents
with both general and, in some cases, specialized caseloads. District office field
agents are organized into units and, in larger districts, units within sub-offices. A
sub-office has a deputy district director. See Appendix C for organizational charts
of the district offices.

       A unit typically consists of a supervisor and several agents, ranging from four
(one unit) to ten (two units) with most units (26 units) having seven agents and the
vast majority (59 units) having from six to eight agents. 3 Most units are responsi-
ble for supervising a general caseload, which would include all types of offenders,
although in certain areas of the state, the offender population supports designated
units for specific types of offenders, e.g., sex offenders and offenders with mental
health issues. As shown in Appendix C, districts range in size from Williamsport
with 2 units (although it is the largest district geographically covering 11 counties)
to Philadelphia with 21 units. Only Philadelphia has a designated mental health
unit and a special offenders unit. Pittsburgh has a designated special needs unit.
In other districts, individual field agents may have special caseloads.

       Although the majority of parole agents are field agents who supervise a
general caseload, as part of the evidence-based re-entry focus 4 discussed below,
some are designated as transitional coordinators (TC) who supervise newly re-
leased offenders for the first 90 days. 5 These cases are then assigned to the other
field agents. ASCRAs, discussed in more detail below, are another type of specia-
lized agent and do not carry individual caseloads. In addition to these types of


2An  institutional office is located at a state correctional institution. The staff is responsible for processing of-
fenders who are scheduled for parole interviews. These agents prepare summarization reports (a comprehen-
sive document concerning each offender) and compile all materials necessary for Board members and hearing
examiners to conduct parole interviews. The staff does not supervise offenders in the community and are there-
fore not part of the focus of this study.
3As reflected on organizational charts for the districts as of April 1, 2009.
4Evidence-based practices are changes to the culture and core supervision practices and procedures of the agen-

cy and are integrated into all re-entry services. Re-entry refers to the services the Board provides to help an
offender transition from prison to the community to successfully complete parole. Throughout the report we use
the term “evidence-based re-entry” when referring to these practices.
5In response to recommendations made by Dr. John Goldkamp in his December 1, 2008, report, Restoring Parole

and Related Processing for Categories of Violent State Prisoners: Findings and Recommendations II, the PBPP
requires those offenders in the “most violent” category to be released to a community corrections center and su-
pervised at maximum supervision with a curfew for 90 days and all state-sentenced offenders to be supervised
at maximum supervision with a curfew for 90 days.

                                                              6
specialized units and agents, Philadelphia has a dedicated transport unit. 6 Trans-
portation of offenders in other districts is handled by general field agents.

       According to a Board official, the PBPP generally considers a unit to consist
of one supervisor, six field agents, and one clerical staff. Individual districts, how-
ever, may assign fewer or more field agents depending on the population needs of
the geographic area for which the unit is responsible. The district offices use census
tracks to establish the geographical territory and assign staff using zip codes. As
the census numbers change, the territory assigned to a particular unit, or the staff-
ing of a unit, may change to respond to the new numbers. The Philadelphia District
Director noted that the 10-field-agent unit reflected on the April 2009 organization
chart in his district included filled positions where the field agent was on military or
medical leave. In that case, it served no purpose to move the positions to another
unit until the field agents were active.

Fast Units

       The Board created a Fugitive Apprehension Search Team (FAST) in the Phil-
adelphia District Office in 2002 to improve its efforts in recovering absconders.
Since that time, FAST units have been added to Pittsburgh and Harrisburg, and
Erie, Scranton, and the Reading Unit of the Allentown District Office have general
agents who are assigned as needed for locating absconders.7 FAST agents are re-
sponsible for locating and apprehending absconders and do not have general offend-
er population supervision responsibilities. The FAST complement is included in the
district office complement of field agents.

        Board officials credit the FAST units and agents with reducing the absconder
rate from 6.6 percent of the supervised offender population in 2003 to 4.3 percent of
the supervised offender population in 2008. The number of days an offender is
missing has also decreased from an average of 211 days in 2003 to an average of 142
days in 2007. The national absconder rate was reported by the Bureau of Justice
Statistics as 9 percent for adults on probation and 7 percent for adults on parole in
2007. The PBPP has calculated a comparable national absconder rate of 10 per-
cent. 8




6Due  to county prison overcrowding, parolees who are arrested for technical parole violations cannot remain in
the county prison overnight, necessitating a transfer to an SCI. This occurs daily. In the case of female offend-
ers, transport to the SCIs that house women occurs on a weekly basis.
7The PBPP reports that it generally takes 100 to 150 absconders to justify a complete unit of agents strictly as-

signed to FAST duties.
8To make the rate comparable, the PBPP combined the probation and parole cases, removed out-of-state cases,

and removed non-state supervised cases from the national data.

                                                            7
Assessment Sanctioning and Community Resource Agents

        In 2006, the PBPP implemented an evidence-based re-entry focus for parole,
shifting away from a more traditional enforcement focus. Re-entry is the recogni-
tion that, once offenders have been held to account and served their time, their abil-
ity to transition to the community as law abiding citizens is in the interest of public
safety. 9 Therefore, parole must address the offender’s particular needs, e.g., addic-
tion and employment. Part of the evidence-based approach used includes an inten-
sive supervision focus during the first 90-day re-entry period. See Exhibit 2 for a
brief comparison of the law enforcement and evidence-based re-entry models of su-
pervision. The implementation of an evidence-based re-entry focus included the
creation of the ASCRA position. ASCRAs provide technical assistance to parole
agents when an offender is having difficulty adjusting to life outside of prison, has
not responded positively to sanctions for parole violations, and is therefore in jeo-
pardy of recommitment to prison. ASCRAs were first funded in FY 2007-08 and are
included in the overall complement of field agents. The statewide complement in-
cludes 16 ASCRAs. Each district office has at least one ASCRA, with the Philadel-
phia District Office having four and the Allentown, Harrisburg, and Pittsburgh dis-
trict offices each having two.

                                                     Exhibit 2
     Comparison of Law Enforcement and Evidence-based Re-entry Parole Models

              Law Enforcement Approach                            Evidence-based Re-entry Approach
 Monitors compliance with the conditions of parole.        Monitors compliance with the conditions of parole.

                                                           Engages the offender in efforts to change beha-
                                                           vior and reduce risk through:
                                                           -staff/offender interactions designed to
 Brings violations of those conditions to the atten-
                                                            enhance motivation for change; and
 tion of the authorities.
                                                           -providing interventions/programs that
                                                            address the specific risks and needs of the
                                                            offender.
                                                           Responds to violations by holding the offender
                                                           responsible but uses a proportional and problem-
 Responds to violations with sanctions including
                                                           solving sanction/response. This would include
 reincarceration.
                                                           reincarceration in the case of very serious viola-
                                                           tion, or where public safety was in danger.

Source: Developed by LB&FC staff in consultation with the Center for Effective Public Policy.


       These agents maintain information and develop resources concerning
individual and group treatment available to offenders in the community, serve as
local experts in matters pertaining to evidence-based practice, facilitate cognitive
behavioral groups of offenders addressing identified criminogenic needs, and are

9A   Call to Action for Parole, Peggy Burke and Michael Tonry (2006 Center for Effective Public Policy).

                                                              8
available to advise parole supervision staff in matters related to proper administra-
tion of assessment instruments. ASCRAs also serve as the district resource in mat-
ters relating to offender employment and vocational training opportunities. AS-
CRAs do not supervise a general caseload of offenders.

Field Agent Complement and Distribution
       The statewide complement for field agents for FY 2008-09, including AS-
CRAs, is 505. Table 10 on page 31 shows the complement level of field agents in
each district for the last four fiscal years. Field agent positions are assigned to dis-
tricts based on the anticipated number of offenders requiring supervision, their le-
vels of supervision, and the resulting workload requirements. See Finding C for
additional discussion of the caseload and workload standards. Table 8 on page 27
shows the distribution of offenders by district over the same period.

       As shown on Table 1 below, each district’s complement of field agents is gen-
erally in proportion to the district’s number of offenders. For example, the Chester
District Office has had about 8 percent of the total complement of field agents and
about the same percentage of the total number of offenders. The other districts si-
milarly have a proportionate share of the field agent complement. See Finding B
for more discussion of field agent distribution.

                                                               Table 1
                                     Percent Comparison of Offender Population,
                                       Parole Agents, and Agent Expenditures
                                                          (June 2009)
          District Office               Offender Populationa     % of Total    # of Parole Agentsb   % of Total
     Allentown ...............                 3,439                11.0%             56               11.1%
     Altoona ..................                1,277                 4.1              23                4.6
     Chester ..................                2,501                 8.0              39                7.7
     Erie ........................             1,435                 4.6              24                4.8
     Harrisburg .............                  3,983                12.8              61               12.1
     Mercer ...................                3,034                 9.7              46                9.1
     Philadelphia ...........                  8,942                28.7             140               27.7
     Pittsburgh ..............                 3,944                12.6              69               13.7
     Scranton ................                 1,761                 5.6              30                5.9
     Williamsport ...........                   863                  2.8              17                3.4
                                                                           c
       Totals .................              31,179                100.0%            505              100.0% c

   _______________
   aTotal offender population (including absconders, detainees, and unclassified offenders).
   bAuthorized complement for Parole Agents 1s and 2s, including ASCRA staff.
   c May not add due to rounding.

   Source: Developed by LB&FC staff using information obtained from the PBPP.

                                                                     9
B. Field Agent Complement Is Based on the Offender Population and
Its Levels of Supervision; However, Actual Staffing Varies Considera-
bly Depending on District Turnover and Vacancies


      The complement of parole agent positions assigned to a PBPP district office is
based on the expected state offender population and levels of supervision for that
population. The number of field agent positions that are required is annually de-
termined by the application of a workload model that is based on a methodology de-
veloped by the National Institute of Corrections (NIC).

       By identifying staffing needs, and by focusing particularly on the field agent
positions and the time and tasks required to properly supervise offenders, the work-
load model is designed to ensure the fair and equitable distribution of resources to
each of the Board’s ten district offices. However, the number of active agents in at
least one district, Philadelphia, lags behind authorized complement due to issues
with maintaining staff.

                                                               Workload Model

       The workload model used by the PBPP is designed to forecast the number of
staff needed to adequately supervise the estimated offender population. According
to PBPP officials, the model used by the Board is one developed by the NIC. As de-
scribed in a 1983 report 1 prepared for the NIC, a workload approach to parole and
probation management rests on the following assumptions:

              •       There is a quantifiable amount of time one can expect from an employee.
              •       The amount of work assigned to an employee must be completed during a
                      prescribed period of time.
              •       The time it “normally” takes to supervise a particular type of case or com-
                      plete an investigation can be measured.
              •       The aggregation of time factors constitutes a method of resource allocation
                      superior to that of caseload equalization.

       The NIC report, which presents a workload model similar to the one used by
the PBPP, states that the purposes of the workload systems are to: (1) provide data
for budget justification and support; (2) enable an agency to appropriately allocate
its resources; and (3) enhance agency accountability. Budget requests based on
workload measures are, in the opinion of many correctional administrators, more
easily defended than those based on caseload ratios, and generally result in a more
                                                            
1Workload Measures for Probation and Parole, authored by Brian Bemus, Gary Arling, and Peter Quigley (pre-
pared under a grant from the National Institute of Corrections, U.S. Department of Justice), May 1983.

                                                                     10
equitable distribution of agency resources. See Finding C for a detailed discussion
of workload determination and Exhibit 4 (on page 35), which provides an example of
how the number of hours per month necessary to supervise a particular offender is
calculated.

Parole Agent Staffing and the Budgetary Process

       While a valuable tool, the workload model used in the preparation of the
PBPP’s annual Agency Budget Request is only one aspect of the normal budgetary
process. As the following description provided by PBPP officials points out, many
factors can result in the original request being modified throughout the budget
process:

      For FY 2005-06 budget preparations, the Governor’s mid-year budget
      briefing in December 2004 demonstrated that the economy was slowly
      coming out of the 2001-02 recession and the peak in growth in 2004
      would be sustained. Although the Board originally requested agent
      positions, it was able to modify its budget request following discussions
      with the Governor’s office. The Board was able to meet its needs
      through further efficiencies in operations and through authorization
      from the budget office to reallocate positions within existing comple-
      ment levels to support critical activities.

      In FY 2006-07, the Board did not request an increase in complement
      due to general budget instructions to maintain an austerity approach
      to the budget. The new chairman began an effort to critically review
      all programs and operations to achieve more efficient decision making,
      develop a re-entry bureau, streamline operations, and develop a multi-
      year plan to reduce recidivism. This analysis resulted in a significant
      budget request in the subsequent two years to implement a multi-year
      plan of action to reduce recidivism.

      In PBPP’s October requests for FY 2007-08 and 2008-09, the Board re-
      quested Field Agent complement increases of 78 and 23 agents, respec-
      tively. During the normal updates between October and January,
      these requests evolved into a final request of 71 and 20 agents in the
      Governor’s Executive Budget.

        PBPP officials informed us that personnel/parole agent positions are assigned
to offices based on the supervision needs of the state offender population, with sup-
port staff assignments following parole agent assignments. Therefore, budget ex-
penses are incurred based on filled complement positions.




                                         11
                           Analysis of District Office Personnel Expenditures
       The focus of our analysis is the cost of carrying out the offender supervision
function which, for purposes of analysis, we defined as the annual expenditures for
the Parole Agent 1 and Parole Agent 2s within their respective district locations.
We examined expenditure data over four consecutive fiscal years (FY 2005-06
through FY 2008-09). In general, we looked for trends or expenditures that ap-
peared outside of the norm (using percentages to establish the statewide or district
averages).

Proportion of Parole Agent Expenditures to DO Personnel Expenditures
       We found that parole agent expenditures 2 represented between 70 and 73
percent of all district office personnel costs during the period. In addition to parole
agent expenditures, the total personnel costs of a district office would also include
expenditures supporting a district director, clerical staff, and parole supervisors.
Table 2 reflects how the percentages of parole agent expenditures to each district’s
total personnel expenditures compare to the statewide average. The data shows
that the districts consistently spend about two-thirds or more of their personnel ex-
penditures for offender supervision. Further, the table shows that the statewide
average percentage of the parole agent expenditures in relation to the total district
personnel expenditures has increased each year (from 69.7 percent in FY 2005-06 to
72.6 percent for FY 2008-09). Over the four-year period, total parole agent expendi-
tures increased by $8,114,377 (29.7 percent), and, during the same period, the pa-
role agent complement increased by 94 (22.9 percent), from 411 to 505. During the
same period, the supervised offender population increased by 2,807 (10 percent),
from 28,372 to 31,179.

Relationship of Offender Population to Agents and Agent Expenditures
      We compared each district’s number of offenders and authorized parole agent
complement in each of the last four years. We used the authorized complement of
Parole Agent 1s and 2s which includes 16 ASCRA agents in FY 2007-08 and FY
2008-09. The offender population represents the total district and/or statewide of-
fender population, including absconders, detainees, and unclassified offenders. 3 As
shown on Table 3, we calculated the number of offenders supervised by each agent,
                                                            
2Parole  agent expenditure information obtained from the PBPP includes all Parole Agent 1s and Parole Agent
2s working from a district office location. These expenditures include all costs (for example, salaries, benefits,
travel, and training expenses) for parole agents, most of whom are assigned caseloads. However, for FY 2007-08
and FY 2008-09, the parole agent expenditures also include funding for 16 ASCRA agents who are not assigned
caseloads. The accounting system would not easily permit the separation of ASCRAs from other Parole Agent
1s and 2s for purposes of this analysis. Therefore, the parole agent expenditures for FY 2007-08 and FY 2008-
09 represent approximate dollar amounts that are slightly higher than expenditures for parole agents supervis-
ing caseloads.
3An absconder is an offender that fails to report for supervision, changes an approved residence without permis-

sion, and whose whereabouts is unknown. A detainee is an imprisoned offender accused of a new violation or
crime pending adjudication. An unclassified offender is an offender available for supervision, but whose super-
vision level is still under review.

                                                               12
                                                     Table 2
                      Comparison of Parole Agent Personnel Costs With
                        Total Personnel Costs in the District Offices
                                     (FY 2005-06 Through FY 2008-09)
                                     FY 2005-06                                       FY 2006-07
                                                        Percent                                           Percent
                                                         Parole                                            Parole
       District    DO Personnel      Parole Agent       Agent to     DO Personnel      Parole Agent       Agent to
       Office      Expenditures      Expendituresa        DOb        Expenditures      Expendituresa        DOb

Allentown            $3,887,738        $2,859,674          73.6%       $4,182,595        $3,044,034          72.8%
Altoona               1,643,391         1,070,896          65.2         1,785,395         1,168,843          65.5
Chester               3,546,898         2,508,771          70.7         3,999,077         2,784,999          69.6
Erie                  1,735,921         1,236,398          71.2         1,841,633         1,331,029          72.3
Harrisburg            4,565,020         3,441,576          75.4         5,070,960         3,836,491          75.7
Mercer                3,392,766         2,241,161          66.1         3,615,225         2,408,939          66.6
Philadelphia         10,750,499         7,092,159          66.0        11,352,559         8,257,513          72.7
Pittsburgh            6,179,690         4,278,822          69.2         6,560,658         4,490,528          68.4
Scranton              2,110,211         1,565,445          74.2         2,286,803         1,680,198          73.5
Williamsport          1,365,802           999,934          73.2         1,405,449         1,014,339          72.2

 Statewide          $39,177,936      $27,294,836           69.7%      $42,100,354       $30,016,913          71.3%


                                     FY 2007-08                                       FY 2008-09
                                                        Percent                                           Percent
                                                         Parole                                            Parole
       District    DO Personnel      Parole Agent       Agent to     DO Personnel      Parole Agent       Agent to
       Office      Expenditures      Expendituresa        DOb        Expenditures      Expendituresa        DOb

Allentown            $ 4,610,233       $ 3,493,087       75.8%        $ 5,446,435      $ 4,143,102         76.1%
Altoona                1,974,244         1,372,845       69.5           2,193,361         1,626,661        74.2
Chester                3,650,635         2,596,255       71.1           3,755,982         2,824,856        75.2
Erie                   2,086,485         1,485,598       71.2           2,336,803         1,736,656        74.3
Harrisburg             5,401,840         4,096,727       75.8           5,641,922         4,270,148        75.7
Mercer                 3,905,008         2,657,218       68.0           4,164,673         2,875,489        69.0
Philadelphia          11,755,450         8,365,025       71.2          13,355,217         9,506,545        71.2
Pittsburgh             6,726,747         4,600,977       68.4           7,268,941         4,965,644        68.3
Scranton               2,391,735         1,807,544       75.6           2,812,544         2,154,539        76.6
Williamsport           1,627,519         1,201,166       73.8           1,796,112         1,305,573        72.7

 Statewide           $44,129,896       $31,676,442       71.8%        $48,771,990      $35,409,213         72.6%
_______________
aThis column represents district office expenditures associated with the Parole Agent 1 and Parole Agent 2 positions
only.
bThe percentages in this column represent the proportion of parole agent expenditures to the total personnel expend-
itures of each district office.
Source: Developed by LB&FC staff using information obtained from the PA Board of Probation and Parole.
                                                        13
                                                         Table 3
                                      Comparison of Offender Population,
                                 Parole Agents, and Expenditures Per Offender*


June 2006
                             Offender      # of Parole   Offenders per   Parole Agent    Expenditures
                            Populationa     Agentsb      Parole Agent     Expensesc      per Offender
Allentown .........           3,254            40               81       $2,859,674        $ 879
Altoona ............          1,058            16               66        1,070,896         1,012
Chester ............          2,291            33               69        2,508,771           1,095
Erie ..................       1,217            18               68        1,236,398           1,016
Harrisburg ........           3,752            53               71        3,441,576             917
Mercer .............          2,481            38               65        2,241,161             903
Philadelphia .....            8,850           116               76        7,092,159             801
Pittsburgh ........           3,856            61               63        4,278,822           1,110
Scranton ..........           1,556            22               71        1,565,445           1,006
Williamsport .....              828             14              59          999,934           1,208
  Statewide ......           29,143           411              71        $27,294,836      $ 937
                                          10% Range          64 - 78     10% Range      $843 - $1,030



June 2007
                             Offender      # of Parole   Offenders per   Parole Agent    Expenditures
                            Populationa     Agentsb      Parole Agent     Expensesc      per Offender

Allentown ...........         3,399           41                83       $3,044,034       $     896
Altoona ..............        1,124           16                70        1,168,843           1,040
Chester ..............        2,307           33                70        2,784,999           1,207
Erie ....................     1,294           18                72        1,331,029           1,029
Harrisburg ..........         3,792           53                72        3,836,491           1,012
Mercer ...............        2,611           38                69        2,408,939             923
Philadelphia .......          8,776          119                74        8,257,513             941
Pittsburgh ..........         3,908           61                64        4,490,528           1,149
Scranton ............         1,520           23                66        1,680,198           1,105
Williamsport .......            837            13               64        1,014,339           1,212
  Statewide ........         29,568          415                71       $30,016,913       $1,015
                                          10% Range         64 - 78      10% Range      $914 - $1,117




                                                           14
Table 3 (Continued)


 June 2008
                            Offender      # of Parole    Offenders per   Parole Agent        Expenditures
                            Populationa   Agentsb        Parole Agent    Expensesc           per Offender
 Allentown .......             3,578           52              69        $3,493,087             $ 976
 Altoona ..........            1,300           20              65         1,372,845              1,056
 Chester ..........            2,504           38              66         2,596,255              1,037
 Erie ................         1,443           23              63         1,485,598              1,030
 Harrisburg ......             4,162           60              69         4,096,727                984
 Mercer ...........            2,886           44              66         2,657,218                921
 Philadelphia ...              9,410          136              69         8,365,025                889
 Pittsburgh ......             4,226           68              62         4,600,977               1,089
 Scranton ........             1,658           28              59         1,807,544               1,090
 Williamsport ...               930            16              58          1,201,166              1,292
   Statewide ....            32,097           485              66        $31,676,442            $ 987
                                          10% Range         60 - 73      10% Range            $888 - $1,086




 June 2009
                             Offender      # of Parole   Offenders per   Parole Agent        Expenditures
                            Populationa     Agentsb      Parole Agent     Expensesc          per Offender
 Allentown ..........          3,439           56              61        $ 4,143,102            $1,205
 Altoona .............         1,277           23              56          1,626,661              1,274
 Chester .............         2,501           39              64          2,824,856              1,129
 Erie ...................      1,435           24              60          1,736,656              1,210
 Harrisburg .........          3,983           61              65          4,270,148              1,072
 Mercer ..............         3,034           46              66          2,875,489                948
 Philadelphia ......           8,942          140              64          9,506,545              1,063
 Pittsburgh .........          3,944           69              57          4,965,644              1,259
 Scranton ...........          1,761           30              59          2,154,539              1,223
 Williamsport ......            863            17              51          1,305,573              1,513
    Statewide ...            31,179           505              62        $35,409,213            $1,136
                                          10% Range        56 - 68       10% Range        $1,022 - $1,249

_______________
*Bolded numbers are outside the 10 percent variance.
aTotal offender population (including absconders, detainees, and unclassified offenders).
bAuthorized complement for Parole Agent 1s and 2s, including ASCRA staff.
cActual expenditures (including wages, salaries, and benefits) of Parole Agents 1s and 2s, including ASCRA staff.

Source: Developed by LB&FC staff using information obtained from the PBPP.

                                                          15
i.e., caseload. We also compared the district’s parole agent expenditure for each of-
fender. We expected the districts to be generally within 10 percent of the statewide
average.

       For purposes of our review, we used a 10 percent variance from the statewide
average as an indication of inequity among the districts, recognizing that vacancies
and seniority of agents would directly affect the expenditure calculation. As Table 3
shows, in the last four years, the caseloads were outside the 10 percent variance
seven times, with two of those instances, the Allentown District Office in June 2006
and June 2007, being over 10 percent higher than the statewide average. The other
five instances were over 10 percent lower than the statewide average. Although the
Philadelphia Office was within the 10 percent variance from the statewide average,
it exceeded the statewide average in each year. As noted in Finding C, the state-
wide caseload average has been declining.

       In the case of the field agent expenditures by offender, variations were more
frequent, and tended to be over 10 percent higher than the average. In all four
years, the Pittsburgh and Williamsport District Offices have exceeded the statewide
average by over 10 percent. As shown on the table, the Williamsport District Office
had the highest expenditures per offender in each year. This may be directly attri-
buted to the small size of the district and the stability of the work force. Districts
such as Williamsport with a more experienced, stable complement have higher ex-
penditures per offender than districts such as Philadelphia with less experienced,
highly fluctuating agent complements.

        Proportional Relationship of District Supervision Levels

       The preceding analyses have treated each district’s offender population as es-
sentially requiring the same level of supervision as the population in every other
district. If this were the case, the percentage breakout of each district’s offender
population by level of supervision (i.e., enhanced, maximum, medium, minimum,
and administrative) would be approximately the same. However, it is clear from
PBPP statistical information that this is not the case.

       Table 4 indicates the variation in numbers and percentages of the statewide
offender population under active supervision for the month of June in each of the
last four years (2006 through 2009). This information reflects the adjustments to
the level of supervision made in 2006, which resulted in decreasing the level of su-
pervision applicable to certain offenders. It also shows the recent increases in levels
of supervision due to the response to the violence in Philadelphia and the increase
due to the recommendations in the Goldkamp reports (see Finding D).




                                          16
        Table 4 indicates that over 40 percent of the actively supervised offender
population was categorized at the lowest supervision level (i.e., the minimum level
but also including the administrative level introduced in January 2009). This level
increased, both numerically and by percentage, between 2006 and 2008, reaching a
high of 13,479 offenders classified as minimum (54.4 percent of the offender popula-
tion of 24,770). This trend may reflect a combination of the shift toward the evi-
dence-based re-entry focus (see Finding C) and the application of the offender
screening instrument that reviews each offender’s risk for re-offense and needs (see
Finding D) using evidence-based practices. A guiding belief undergirding this ap-
proach is that by reducing an offender’s level of supervision, when possible, the abil-
ity to supervise higher risk offenders is thereby increased.

                                                            Table 4
                      Statewide Supervised Population, by Level of Supervision* 
                                 (Months of June 2006 Through 2009)

               Fiscal Year Ending                     June 2006       June 2007    June 2008    June 2009 
      Enhanced ...................................       182             241          186           390 
       % of State Total .......................            0.8%            1.1%         0.8%           1.6%

      Maximum ...................................      6,661           5,694        5,280         6,521 
       % of State Total .......................           30.9%           25.8%        21.3%         26.7% 
      Medium ......................................    5,785           5,890        5,825         7,098 
       % of State Total .......................           26.8%           26.7%        23.5%         29.0% 
      Minimum ....................................     8,936          10,246       13,479         8,984 
       % of State Total .......................           41.4%           46.4%        54.4%         36.8% 
      Administrativea ...........................                                                 1,445 
       % of State Total .......................                                                       5.9% 
      Statewide Total ..........................      21,564          22,071       24,770        24,438 
_______________ 
*These numbers do not include offenders who are not classified or are non-reporting, i.e., absconders and detainees.
aThe PBPP Law was recently amended by Act 2008-83 to require an eligible offender to be placed on admin-
istrative parole one year after release on parole and until the maximum sentence date under certain condi-
tions.

Source: PBPP Monthly Program Reports, June 2006, June 2007, June 2008, and June 2009.


       The levels of supervision are an important consideration in this analysis be-
cause the workload model recognizes that a field agent must dedicate more time in
the proper supervision of offenders in the enhanced and maximum supervision le-
vels than in the medium, minimum, and administrative supervision levels, and ac-
cordingly uses different values for each level in the application of the model.

                                                                17
        According to PBPP officials, the workload model takes into account each dis-
trict’s level of supervision in determining staffing levels. District parole agent as-
signments are based on the workload model (total workload hours divided by 131
equals the number of agents needed) and each district’s total workload is the result
of the individual offender supervision levels (total hours necessary to supervise the
entire district specific population) plus the hours needed to complete the investiga-
tions and other duties in their district.

        Table 5 presents information for FY 2005-06 through FY 2008-09 on each dis-
trict’s number and percentage of offenders under active supervision (i.e., after sub-
tracting offenders who are not classified and the absconders and detainees). An ex-
amination of individual districts demonstrates that, for any given point in time,
each district’s offender population may reflect a different set of characteristics. For
example, the “enhanced” offender group in Philadelphia represented 0.5 percent (37
offenders) of the district’s population in June 2008. However, by June 2009, the en-
hanced group was 3.6 percent of the district’s population (230 offenders). Similarly,
while Philadelphia’s “maximum” population represented nearly 44 percent of its
population in June 2006, that offender group represented only 28 percent of its pop-
ulation in June 2009. As noted earlier, the increase to the enhanced level is in reac-
tion to recent incidents involving parolees in Philadelphia. The decline in the max-
imum population is due to the adjustments made to the level of supervision as a re-
sult of the assessment of the application of the Level of Service Inventory-Revised
(LSI-R) that tended to reduce the level of supervision applicable to certain offend-
ers.

    Further Analysis of Parole Agent Staffing at the District Offices

      Just as the characteristics of the offender population can affect how many
agents are necessary for proper supervision, so too does the extent to which various
forms of leave (paid and unpaid) and personnel actions (e.g., hiring decisions, re-
tirements, resignations, and terminations) occur substantially change the
caseload/workload assignments for the remaining on-the-job agents who may be re-
quired to supervise increased caseloads.

      The complement of parole agents shown on Table 6 constitutes the maximum
agent staffing authorized for the district offices. As a matter of course, and for vari-
ous reasons, individuals are not always available to perform their duties, thereby
reducing the number of agents supervising parolees. Table 6 shows the disparity
for June 30 in each of four consecutive years between the authorized number of
agents (i.e., complement) and the actual number of agents available to supervise of-
fenders (see Table 9 in Finding C for a more detailed breakdown by district).




                                           18
                                                                               Table 5

                                                    Supervised Population by Level of Supervision and District*
                                                                 (Months of June 2006 Through 2009)
     June 2006
                                          % of                     % of                   % of                 % of                 % of
                                         District                 District               District             District             District   District
           District           Enhanced    Total        Maximum     Total     Medium       Total     Minimum    Total     Admin.a    Total      Total

     Allentown........          13        0.6%           673       29.1%      656        28.4%         971     42.0%                           2,313
     Altoona ...........         9        1.1            172       20.3       270        31.8          398     46.9                              849
     Chester...........         13        0.8            601       35.1       503        29.4          595     34.8                            1,712
     Erie .................     11        1.2            199       21.2       269        28.6          461     49.0                              940
     Harrisburg ......          24        0.9            821       29.6       594        21.4        1,339     48.2                            2,778
     Mercer ............         5        0.2            263       12.6       626        30.1        1,188     57.1                            2,082
     Philadelphia ...           45        0.7           2,684      43.9      1,551       25.4        1,827     29.9                            6,107
     Pittsburgh .......         48        1.7            934       33.0       827        29.2        1,024     36.1                            2,833
     Scranton .........         12        1.0            234       18.7       333        26.6          672     53.7                            1,251
19




     Williamsport....            2        0.3             80       11.4       156        22.3          461     66.0                              699

     June 2007

     Allentown........          15        0.6%           566       23.5%      736        30.6%       1,092      45.3%                          2,409
     Altoona ...........         7        0.8            175       19.1       314        34.4          418      45.7                             914
     Chester...........         20        1.1            493       28.1       585        33.3          657      37.4                           1,755
     Erie .................      2        0.2            181       17.8       318        31.2          518      50.8                           1,019
     Harrisburg ......          32        1.1            794       27.3       679        23.3        1,406      48.3                           2,911
     Mercer ............         6        0.3            214        9.9       611        28.2        1,333      61.6                           2,164
     Philadelphia ...           95        1.6           2,126      35.5      1,267       21.2        2,498      41.7                           5,986
     Pittsburgh .......         54        1.9            872       30.0       861        29.6        1,118      38.5                           2,905
     Scranton .........          6        0.5            180       14.1       351        27.5          740      57.9                           1,277
     Williamsport....            4        0.5             93       12.7       168        23.0          466      63.7                             731
 Table 5 (Continued)
 June 2008
                                          % of                       % of                        % of                        % of                        % of
                                         District                   District                    District                    District                    District         District
           District           Enhanced    Total     Maximum          Total        Medium         Total       Minimum         Total        Admin.a        Total            Total

     Allentown........          21        0.8%          618           22.6%         627          23.0%         1,463          53.6%                                       2,729
     Altoona ...........         2        0.2           228           20.8          382          34.9            484          44.2                                        1,096
     Chester...........         20        1.0           211           10.7          585          29.5          1,165          58.8                                        1,981
     Erie .................      8        0.7           214           18.2          368          31.3            587          49.9                                        1,177
     Harrisburg ......          42        1.3           702           22.5          676          21.7          1,696          54.4                                        3,116
     Mercer ............         6        0.2           194             7.8         706          28.2          1,596          63.8                                        2,502
     Philadelphia ...           37        0.5         1,763           26.1        1,075          15.9          3,872          57.4                                        6,747
     Pittsburgh .......         36        1.1         1029            31.8          883          27.3          1,292          39.9                                        3,240
     Scranton .........         10        0.7           220           15.8          330          23.7            835          59.9                                        1,395
     Williamsport....            4        0.5           101           12.8          193          24.5            489          62.1                                          787
20




     
 June 2009
     Allentown........          40        1.5%          793           29.1%         711          26.1%         1,029          37.7%         155           5.7%            2,728
     Altoona ...........         2        0.2           312           29.5          406          38.4            276          26.1            60          5.7             1,056
     Chester...........         10        0.5           449           22.4          567          28.3            810          40.4          167           8.3             2,003
     Erie .................      7        0.6           342           28.0          426          34.9            381          31.2            66          5.4             1,222
     Harrisburg ......          41        1.3           877           28.9          824          27.1          1,060          34.9          237           7.8             3,039
     Mercer ............         2        0.1           288           10.9          879          33.2          1,310          49.5          166           6.3             2,645
     Philadelphia ...          230        3.6         1,801           27.8        1,556          24.1          2,582          39.9          298           4.6             6,467
     Pittsburgh .......         36        1.2         1,088           35.4        1,128          36.7            685          22.3          137           4.5             3,074
     Scranton .........         18        1.2           426           29.2          375          25.7            551          37.7            90          6.2             1,460
     Williamsport....            4        0.5           145           19.5          226          30.4            300          40.3           69           9.3               744
_______________
*These numbers do not include offenders who are not classified or are non-reporting, i.e., absconders and detainees.
aThe PBPP Law was recently amended by Act 2008-83 to require an eligible offender to be placed on administrative parole one year after release on parole and until the
maximum sentence date under certain conditions.
Source: PBPP Monthly Program Reports, June 2006, June 2007, June 2008, and June 2009.
       As Table 6 shows, the differences between complement (non-ASCRA) and ac-
tive agents (i.e., field office adjusted active agent total) ranged from 19 in June 2006
to as high as 53 for June 2008. Not shown on the table, however, is the numerical
breakdown by district (see Table 9). Although Philadelphia has 30 percent of the
statewide agent complement, it has about 40 percent of the vacancies and agents on
leave. Also, the enhanced level of supervision for Philadelphia for June 2009 con-
sisted of 230 offenders (59 percent of the statewide number of offenders in the en-
hanced category), which increases the applicable workload. 4

                                                                                        Table 6
                                                           Disposition of PBPP Field Agents*
                                                                                                                       Statewide
                                                                                                                         Total
                                                                                                               2009   2008   2007      2006
                           Complement .........................................................                 505    485       415    411
                           ASCRAs.................................................................               16     16         0      0
                           Non-ASCRA Complement ...................................                             489    469       415    411
                           Vacancies ..............................................................              13     30        22      9
                                                                  a
                           Not Available/On Leave ........................................                       18     17         7      5
                                                                             b
                           Re-Assignment or Suspension .............................                              0      3         3      3
                                             b, c
                           Paid Leave               ........................................................      3      3         2      2
                           Field Office Adjusted Active Agent Total...........                                  455    416       381    392
                           Filled Wage Positionsd...........................................                     13    N/A       N/A   N/A
_______________
*As of June 30.
a
 This category includes staff who are out on leave without pay for military or medical reasons or staff out on Worker’s
Compensation Act or Enforcement Officer Disability Benefits Law (Heart and Lung Act) leave.
b For June 30, 2006, 2007, and 2008, this number was estimated by taking the average of the FY 2008-09 number
for each district office and applying that percentage to calculate a number for each district each year.
cPaid leave includes paid time off such as vacation and sick leave that is usually scheduled or of brief duration.
dThis category reflects the use of annuitants.

Source: PA Board of Probation and Parole.


The Philadelphia District Office Has High Turnover Rates

       We conclude that the PBPP uses an objective, workload-based method to de-
termine the number of agents necessary to supervise the offender population.
Starting with a projected number of offenders requiring supervision in the upcom-
ing year, the Board applies a workload model that assigns time values to required
activities necessary to supervise offenders based on their degree of risk, i.e., level of
supervision. Correctional experts and national correctional agencies that research

                                                            
4
 We were informed that the Board had ordered an increase in enhanced cases in response to recent police shoot-
ings by supervised offenders in Philadelphia. 

                                                                                            21
standards setting for probation and parole agencies suggest that workload systems
generally result in a more equitable distribution of agency resources.

       It is more difficult to clearly conclude, however, that the application of the
workload model has actually resulted in an equitable distribution of resources. If
equity and fairness are the desired outcomes, then the workload model, even when
requested positions fail to materialize in the budget process, would seem to be a
useful tool in distributing agents. However, as described in other sections of this
report, the decision of where and how to distribute resources has not been the criti-
cal issue. Rather, it appears that one district in particular, Philadelphia, tends to
experience a number of personnel issues that reduces the number of active field
agents available to properly supervise its offender population. The combination of
high field agent turnover, high number of vacant positions, and high number of
agents on leave, reduces the number of active agents to levels much below the au-
thorized complement. For example, although the complement for the Philadelphia
District Office was 136 non-ASCRA agents, on June 30, 2009, due to leave and va-
cancy situations, only 121 active field agents were available to supervise offenders.
As shown on Table 7 in Finding C, with the exception of Mercer, which is in a
unique situation, Philadelphia had the highest percentage of active field agents
with workloads greater than 131, and 14 of its agents (out of 114) had caseloads of
100 or more.

     In response to our request for information on district office funding, the
PBPP reported that:

      In areas with high agent turnover, such as Philadelphia, budget ex-
      penses as a percentage of spending lag behind the equivalent staffing
      percentage as a function of the higher vacancy rate and lower time in
      position with the attendant lower pay step.

      The following points highlight some of the characteristics of this ongoing
problem (see Finding E for further discussion):

      •   Philadelphia had an average turnover rate of 18.7 percent (significantly
          higher than the statewide average of 11.4 percent) for October 2006
          through June 2009.
      •   Between October 2006 and June 2009, 30 Philadelphia field agents trans-
          ferred to another PBPP office, 13 transferred to other state agencies, 2
          were promoted, 12 retired, 11 resigned, and 4 were terminated.
      •   High turnover in Philadelphia has led to a higher concentration of agents
          with limited agent experience in the district (median years of experience
          were 3.5 for Philadelphia compared with 5.5 statewide).



                                         22
       The above personnel indicators present a picture of an organization strug-
gling to keep its key workforce in place while serving a third of the statewide
offender population. Further, the data related to levels of supervision of offenders
within the Philadelphia District suggest that this district is among a group of dis-
tricts whose agents are likely to be exposed to more difficult offenders requiring
more supervision services than offenders in other parts of the state. See Finding E
for actions taken by the Board to address these issues.




                                         23
C. Turnover and Vacancies Are Key Factors Generating Caseload/
Workload Inequities Between and Among Supervision Units


       The statewide average of caseload and workload assignments by field agents
is within those ranges identified as reasonable by probation and parole experts and
the PBPP. Individual districts, units, and field agents, however, may have signifi-
cantly higher caseloads and workloads due to available staff resources and changes
to the focus of probation and parole from one of enforcement to one of evidence-
based re-entry practices. Caseload is the number of individual offenders assigned to
an officer or a team for supervision or monitoring. Workload is the time required to
supervise a caseload and complete assigned investigations. The PBPP seeks to
maintain an average workload of 131 hours per month per field agent and a result-
ing caseload average of 66 offenders per field agent. 1 As shown on Table 7, howev-
er, on July 2, 2009, four districts had field agents with caseloads of 100 or greater
and all districts had field agents with workloads greater than 131. See Appendix D
for this information as of September 3, 2009.

                                                               Use of Caseload

        We attempted to compare the average PBPP caseloads to industry standards
but found that there is no accepted standard caseload size for field agents. Prior to
a 1991 American Probation and Parole Association (APPA) attempt to establish na-
tional caseload size standards, the Commission on Law Enforcement and the Ad-
ministration of Justice (1967) published a report that stated that 50 offenders were
believed to be an appropriate caseload size. The APPA realized, however, that de-
veloping a definitive national caseload size number was not the best approach, con-
cluding that searching “for the single magic number for the optimal caseload size is
futile” due to the diversity in local expectations for their community correctional
systems, organizational structures, local judicial decision making, and varying of-
fender risks and needs.

        The APPA recommends that caseload be thought of as a starting point for de-
veloping effective interventions as overly large caseloads with unrealistic workloads
can foster ineffective officers and agencies, noting that increasing caseloads make it
virtually impossible for offenders to receive adequate attention and interaction from
agents to have any substantial rehabilitative effect. Besides the growth in offender
to field agent ratios, they also note that the current trend of adding more conditions
of parole is potentially making it more difficult for offenders to meet all of these
conditions and equally as difficult for agents to enforce them. Because offenders



                                                            
1The    actual workload total is 131.26 hours per month.

                                                                     24
               
                                                                      Table 7
                                                     Caseload and Workload Distribution
                                                                   (July 2, 2009)

                                                 # of           Field Agents         % of    Field Agents With    % of
          District Office                        Field         With Caseloads        Field       Workloads        Field
                                                Agentsa        100 or Greater       Agents   Greater Than 131b   Agents
    Allentown ................                      52                0              0.0%            9           17.3%
    Altoona ...................                     21                0              0.0             5           23.8
    Chester ...................                     33                0              0.0            15           45.5
    Erie .........................                  23                0              0.0             4           17.4
    Harrisburg ...............                      56                1              1.8            25           44.6
                  c
    Mercer ...................                      36               10             27.8            29           80.6
                          d
    Philadelphia ...........                      114                14             12.3            53           46.5
    Pittsburgh ...............                      59                1              1.7            19           32.2
    Scranton .................                      29                0              0.0             4           13.8
    Williamsport ............                       13                0              0.0             3           23.1
       Total.....................                 436                26              6.0%          166           38.1%
_______________
aFilled field agent positions carrying a caseload. In general, this number does not include agents on extended medi-
cal or military leave and includes agents currently at the Basic Training Academy (BTA).
bThese workloads include only hours used for offender supervision. Actual workload hours would likely be higher,
resulting in more agents with workloads over 131, and include other duties such as Board of Pardons investigations.
cMercer and Venango Counties do not have county adult probation departments, therefore, the district office provides
all adult probation services. Many of the county cases are of relatively short duration.
dThe agent count does not include the FAST Unit. That unit has four agents and each has a caseload of 100 or
greater and a workload greater than 131.

Source: Pennsylvania Board of Probation and Parole.


vary in age, gender, offense, risk factors, and service needs while parole orders dif-
fer in terms of the conditions placed upon offenders, the number and complexity of
the workload demands placed upon field agents also varies. Additionally, the legal,
political, and policy environments of the jurisdictions providing probation and pa-
role services vary greatly.

       A variety of instruments is used nationwide to assess offender risk and needs
and to guide decisions about case management at intake and at regular intervals
during supervision so that more staff resources and services can be targeted at in-
tensive and moderate to high risk cases. With this is mind, in 2006 Professor Wil-
liam Burrell of Temple University proposed the following offender to field agent ra-
tio based on the level of assessed risk: 2
                                                                
                                                            
2Mr.    Burrell is currently assisting Dr. Goldkamp in his review of PBPP’s standards for caseload and workload.

                                                                          25
               
                                       Intensive supervision............           20:1
                                       Moderate to high risk ...........           50:1
                                       Low risk.................................   200:1
                                       Administrative ......................       No limit? 1,000?

      As further noted by Professor Burrell, the reduction of caseloads alone will
not produce better results. According to current research, positive results come
from a smaller agent caseload combined with the implementation of a more ba-
lanced, evidence-based approach to supervision that emphasizes working with of-
fenders on their criminogenic problems through counseling services and treatments.
The PBPP is pursuing this evidence-based approach.

Caseload Assignment

        PBPP field agent caseload assignment is dependent on the offender popula-
tion distribution throughout the state and the districts. 3 To look for imbalances, we
calculated the offender population in each region and each district as a percentage
of the total offender population of the state as of June 30th for each of the past four
fiscal years (FY 2005-06 through FY 2008-09) and found that the percentages for
the district offices and the regions have remained fairly constant. For example, the
Eastern Region, ranging between 37-38 percent over these years, had a slightly
greater percentage of the offender population than the Central Region (32 percent)
and the Western Region (30-31 percent), but on the whole the offender population
was equally distributed among the regions.

       At the district level, Philadelphia’s percentage of the offender population dur-
ing these years ranged from 29-30 percent while Harrisburg’s and Pittsburgh’s per-
centages stayed at about 13 percent. Also consistent during this time period is the
Philadelphia District Office’s percentage of the offender population being about
twice that of any other district office. See Table 8.

       Using data published monthly by PBPP, we looked at the trend in average
caseload for CYs 2006 to 2008. As can be seen in Exhibit 3, statewide average case-
loads ranged from 66 offenders per filled agent position to a high of 75. 4 Average
caseloads remained fairly level in 2006 and 2007, as did the statewide filled non-
ASCRA complement level. Beginning in September 2008, the effect of the morato-
rium can be seen with the reported average caseload falling from 71 to 66. Also im-
pacting the average caseload was an increase of 46 positions in the statewide filled
non-ASCRA complement from FY 2006-07 to FY 2007-08.



                                                            
3Agents   with special caseloads, e.g., sex offenders, generally have fewer cases assigned to them.
4A filled agent position includes field agents who are on military and medical leave, reassignment, suspension,
or other paid leave.

                                                                        26
               
                                                                                           Table 8
                                                                       Offender Population by Region/District
                                                                                   FYs 2005-06 to 2008-09*

                                                               FY 2005-06                   FY 2006-07            FY 2007-08            FY 2008-09
         Region/District                                   Offender        % of         Offender     % of     Offender     % of     Offender     % of
                                                          Population       Total       Population    Total   Population    Total   Population    Total
         Chester ...................................        2,291           7.9%         2,307        7.8%     2,504       7.8%      2,501       8.0%
         Philadelphia ............................          8,850         30.4           8,776       29.7      9,410      29.3       8,942      28.7

         Eastern Region                                    11,141         38.2          11,083       37.5     11,914      37.1      11,443      36.7

         Allentown ................................         3,254         11.2           3,399       11.5      3,578      11.1       3,439      11.0
         Harrisburg ...............................         3,752         12.9           3,792       12.8      4,162      13.0       3,983      12.8
         Scranton..................................         1,556          5.3           1,520        5.1      1,658       5.2       1,761       5.6
         Williamsport ............................           828           2.8             837        2.8        930       2.9         863       2.8
27




         Central Region                                     9,390         32.2           9,548       32.3     10,328      32.2      10,046      32.2

         Altoona ....................................       1,058          3.6           1,124        3.8      1,300       4.1       1,277       4.1
         Mercer .....................................       2,481          8.5           2,611        8.8      2,886       9.0       3,034       9.7
         Erie..........................................     1,217          4.2           1,294        4.4      1,443       4.5       1,435       4.6
         Pittsburgh ................................        3,856         13.2           3,908       13.2      4,226      13.2       3,944      12.6

         Western Region                                     8,612         29.6           8,937       30.2      9,855      30.7       9,690      31.1

           Total .....................................     29,143                       29,568                32,097                31,179




     _______________
     *Population effective June 30 each fiscal year.

     Source: Monthly Program Reports, July 2005 - July 2008, PA Board of Probation and Parole.



                
                                                                                Exhibit 3


                                                           Statewide Average Caseload*
                                                                 CYs 2006 - 2008
                     95
                     90
of Cases per Agent




                     85
 Average Number




                     80
                     75
                     70
                     65
                     60
                     55
                     50
                                 Jan       Feb      Mar         Apr       May          Jun         Jul         Aug       Sep       Oct       Nov       Dec
                     2006        69         69        69         68         70          71         71          71         72        73        73        74
                     2007        73         73        74         72         73          74         75          73         73        71        73        73
                     2008        67         68        68         67         68          68         70          70         71        69        68        66

                                                                     2006              2007              2008




                                                     Eastern Region Average Caseload
                                                              CYs 2006 - 2008
                     95
                     90
                     85
of Cases per Agent
 Average Number




                     80
                     75
                     70
                     65
                     60
                     55
                     50
                            Jan‐06


                                       Apr‐06


                                                 Jul‐06


                                                            Oct‐06


                                                                       Jan‐07


                                                                                   Apr‐07


                                                                                              Jul‐07


                                                                                                          Oct‐07


                                                                                                                     Jan‐08


                                                                                                                               Apr‐08


                                                                                                                                         Jul‐08


                                                                                                                                                   Oct‐08




                                                                          Chester                      Phila




                                                                                   28
                  
Exhibit 3 (Continued)


                                                     Central Region Average Caseload
                                                              CYs 2006 - 2008
                        95
                        90
   of Cases per Agent




                        85
    Average Number




                        80
                        75
                        70
                        65
                        60
                        55
                        50
                             Jan‐06


                                      Apr‐06




                                                           Oct‐06


                                                                       Jan‐07


                                                                                 Apr‐07




                                                                                                          Oct‐07


                                                                                                                   Jan‐08


                                                                                                                             Apr‐08




                                                                                                                                               Oct‐08
                                                Jul‐06




                                                                                          Jul‐07




                                                                                                                                      Jul‐08
                                      Allentown                     Harrisburg                 Scranton                     Williamsport




                                                         West Region Average Caseload
                                                                CYs 2006 - 2008
                        95
                        90
   of Cases per Agent




                        85
    Average Number




                        80
                        75
                        70
                        65
                        60
                        55
                        50
                             Jan‐06


                                      Apr‐06




                                                                       Jan‐07


                                                                                 Apr‐07




                                                                                                                   Jan‐08


                                                                                                                             Apr‐08
                                                Jul‐06


                                                           Oct‐06




                                                                                          Jul‐07


                                                                                                          Oct‐07




                                                                                                                                      Jul‐08


                                                                                                                                               Oct‐08




                                               Altoona                   Mercer                    Erie              Pittsburgh


_______________
Note: Data table values are rounded.
*Caseload is the total offender population divided by filled agent positions for that month.


Source: Developed by LB&FC staff using CYs 2006 – 2008 data published monthly by PBPP.

                                                                                 29
                     
        While the average caseloads for PBPP as a whole remained within a narrow
range, such was not the case for either individual district offices (as shown in Exhi-
bit 3) or for the units within each of these offices. When looking specifically at indi-
vidual supervision units based on an examination of the PBPP’s inspection reports,
six of the 32 units we reviewed had average caseloads in excess of 100, and all six of
these units were in the Philadelphia District.

       The monthly PBPP statistical reports’ caseload numbers use filled agent posi-
tions. However, agents on military or extended medical leave that are included in
these totals are not actively supervising offenders. Shown on Table 10, the filled
agent complement and the adjusted active field agent total can differ greatly.

       To determine the caseloads of the active field agents, we looked at a daily re-
port generated at the central office. This is point-in-time data that is not main-
tained historically by the PBPP information system. We looked at reports for July
2, 2009, and September 3, 2009. As can be seen in Table 9 below, the recalculated
caseloads were higher than the caseloads reported using the filled agent comple-
ment. Since the July 2, and September 3, 2009, data included new agents who were
attending the Basic Training Academy (BTA), the caseloads of the other agents in
that district would be higher than shown on the table.
                                                          Table 9
                     Comparison of Caseload Calculated Using Active Agents
                               and Using Filled Agent Positions
                                             Average                         Average
                                           Caseload per      Average       Caseload per          Average
                                           Filled Agent    Caseload per    Filled Agent        Caseload per
                                             Position      Active Agent      Position          Active Agent
                District                    June 2009      July 2, 2009    August 2009         Sept. 3, 2009
     Allentown ......................            64             66               65                  69
     Altoona .........................           58             61               58                  61
     Chester .........................           66             76               66                  71
     Eriea .............................         62             63               63                  62
     Harrisburg .....................            68             71               67                 72
     Mercer ..........................           72             84               72                  84
     Philadelphia ..................             68             76               68                  80
     Pittsburgh .....................            64             67               62                  65
     Scranton .......................            63             61               62                  62
     Williamsport ..................             54             66               53                  65

       Statewide Average ....                   66               71              65                  72
_______________
a
 Erie’s active agent complement equaled the filled agent complement for August and September 3, 2009, but the
offender population decreased by September 3 resulting in a higher average caseload for the August filled agent po-
sitions.
Source: Developed by LB&FC staff using data provided by the PBPP.

                                                            30
          
                                                                                                                                        Table 10
                                                                                                             Disposition of PBPP Field Agents*
  
                                                                                                                             Statewide                                                          Eastern Region
                                                                                                                               Total                                        Philadelphia                                      Chester
                                                                                                             2009           2008        2007        2006        2009        2008        2007        2006        2009        2008        2007        2006
                      Complement .........................................................                       505         485         415         411         140         136         119         116          39          38          33         33
                      ASCRAs ................................................................                     16          16           0           0           4           4           0           0           1           1           0          0
                      Non-ASCRA Complement ...................................                                   489         469         415         411         136         132         119         116          38          37          33         33
                      Vacancies ..............................................................                    13          30          22           9           4          16           8           2           0           0           6          1
                                                               a
                      Not Available/On Leave ........................................                            18           17           7           5           9           5           2           2           1           2           0          0
                                                                         b
                      Re-Assignment or Suspension .............................                                    0           3           3           3           0           2           2           2           0           0           0          0
                                         b, c
                      Paid Leave                ........................................................           3           3           2           2           2           1           1           1           0           1           1          1
                      Field Office Adjusted Active Agent Total ..........                                        455         416         381         392         121         108         106         109          37          34          26         31
                                                           d
                      Filled Wage Positions ...........................................
31




                                                                                                                 13          N/A        N/A         N/A            6         N/A        N/A         N/A            1         N/A        N/A         N/A



                                                                                                                                                                 Central Region
                                                                                          Allentown                                        Harrisburg                                       Scranton                                      Williamsport
                                                                             2009        2008    2007             2006        2009        2008    2007            2006        2009        2008    2007            2006        2009        2008    2007          2006
     Complement .............................................                  56           52             41          40          61          60          53          53          30          28          23          22          17          16          14    14
     ASCRAs ....................................................                 2            2              0          0           2           2           0           0           1           1           0           0           1           1           0     0
     Non-ASCRA Complement .......................                              54           50             41          40          59          58          53          53          29          27          23          22          16          15          14    14
     Vacancies .................................................                 0            2              1          1           0           3           2           3           1           1           1           0           0           0           0     0
                                          a
     Not Available/On Leave ...........................                          2            1              1          0           0           0           0           1           1           2           0           0           0           0           0     0
                                                      b
     Re-Assignment or Suspension ................                                0            0              0          0           0           1           1           1           0           0           0           0           0           0           0     0
                     b, c
     Paid Leave ............................................                     0            1              0          0           1           0           0           0           0           0           0           0           0           0           0     0
     Field Office Adjusted Active Agent
     Total ..........................................................          52           46             39          39          58          54          50          48          27          24          22          22          16          15          14    14
                                     d
     Filled Wage Positions ..............................                        2        N/A              N/A     N/A              0         N/A      N/A         N/A              2      N/A         N/A         N/A              0      N/A            N/A   N/A




                 
     Table 10 (Continued)


                                                                                                                        Western Region
                                                                                 Pittsburgh                   Mercer                       Altoona                     Erie
                                                                         2009   2008 2007     2006   2009   2008 2007    2006   2009     2008 2007   2006   2009   2008 2007   2006
       Complement ............................................            69     68     60     61     46     44    38      38     23      20    16    16     24     23    18    18
       ASCRAs ...................................................          2       2     0      0      1      1     0       0      1       1     0     0      1      1     0      0
       Non-ASCRA Complement ......................                        67     66     60     61     45     43    38      38     22      19    16    16     23     22    18    18
       Vacancies .................................................         5       4     2      1      3      3     1       1      0       0     0     0      0      1     1      0
                                           a
       Not Available/On Leave ..........................                   2       3     2      0      3      2     1       1      0       2     1     1      0      0     0      0
                                                     b
       Re-Assignment or Suspension ................                        0       0     0      0      0      0     0       0      0       0     0     0      0      0     0      0
                 b, c
       Paid Leave ...........................................              0       0     0      0      0      0     0       0      0       0     0     0      0      0     0      0
       Field Office Adjusted Active Agent
       Total .........................................................    60     59     56     60     39     38    36      36     22      17    15    15     23     21    17    18
                                       d
       Filled Wage Positions ..............................                2    N/A    N/A    N/A      0    N/A   N/A     N/A      0     N/A   N/A   N/A      0    N/A   N/A   N/A
32




     _______________
     *As of June 30.
     a
       This category includes staff who are on leave without pay for military or medical reasons or staff on Worker’s Compensation Act or Enforcement Officer Disability Benefits Law
     (Heart and Lung Act) leave.
     b
       For June 30, 2006, 2007, and 2008 this number was estimated by taking the average of the FY 2008-09 number for each district office and applying that percentage to calculate a
     number for each district each year.
     c
       Paid leave includes paid time off such as vacation and sick leave that is usually scheduled or of brief duration.
     d
       This category reflects the use of annuitants.

     Source: PA Board of Probation and Parole.




                 
        The offender population decreased less than 0.2 percent from the end of June
to July 2; the changes in average caseload were primarily due to the decrease in the
number of agents used for the calculation. Overall, average caseloads based on the
filled agent complement were 8 percent lower than the average caseloads calculated
using active agents. Similarly, the September data shows a 10 percent difference
when using active agents to calculate caseload rather than filled positions.

                                                        Use of Workload Calculation

       Offenders vary in their need for specific services and, therefore, so does the
time an agent needs to spend supervising the offender. Although some offenders
have substance abuse issues, difficulty finding and keeping a job, mental health
problems, and limited education, others do not have these problems. The compila-
tion of these characteristics along with other criminal history factors shapes an of-
fender’s relative risk of re-offending in the future. Many probation agencies realize
the need to consider these characteristics and have begun incorporating various
scales to determine an offender’s risks and needs to make better administrative de-
cisions regarding workload allocations.

      PBPP has used the Level of Service Inventory-Revised (LSI-R) to assess of-
fender risk and needs since 2004. It is an objective, quantitative instrument that
provides a consistent and valid method of predicting risk to re-offend and a reliable
means of measuring offender change over time through its re-administration. See
Finding D for a more detailed discussion of LSI-R assessment and Level of Supervi-
sion.

       Workload is the total amount of time that the required tasks and activities in
a particular caseload generate for the individual parole officer or team. The APPA
noted that the increases in the number of offenders under community supervision in
recent years makes field agent workload allocation issues a matter of concern. One
of the traditional approaches to workload issues is to conduct time analyses within
individual agencies as was done in a modified manner by the PBPP in 2002. Work-
load studies have been common in probation and parole for more than two decades 5
and are a component of the National Institute of Corrections (NIC) Model Systems
Project. Workload growth is distinct from growth in caseloads because the former
refers to the amount of work required to adequately supervise each offender’s com-
pliance with ordered conditions of supervision, whereas the latter refers to the
number of offenders supervised by each agent. Because the majority of its field
                                                            
5ACA standards (1998, third edition) include having a “Written policy, procedure, and practice provide for a

written workload formula which is implemented in the allocation of work to field staff.” 3-3126. The comment
accompanying the standard explains that this formula should consider factors such as legal requirements, goals,
character, and needs of offenders to be supervised, geographic area, administrative tasks required of the field
staff, and types of personnel to be utilized. A workload rather than caseload model is based on programs of dif-
ferential supervision ranging from intensive to minimal. Supervision tasks must be identified, measured
against a time requirement, and then translated into specific total time and staff requirements.

                                                                    33
               
agents oversees offenders at each level of supervision, PBPP developed a weighted
formula approach to determine actual workload of the agent rather than solely us-
ing caseload.

Determining Workload – PBPP

       The PBPP has defined workload as the time required to supervise a caseload,
complete assigned investigations, and conduct other duties. Workload is comprised
of several elements: direct supervision duties (face-to-face contacts, collateral con-
tacts, and field notes); other supervision duties (random urinalysis, violation/routine
paperwork, conferences, hearings, and surveillance); and miscellaneous duties
(transports, staff meetings, computer time, and vehicle maintenance). The specific
requirements to supervise an offender are based on the offender’s level of supervi-
sion. This is determined by the offender’s assessed level of risk. See Exhibit 8 in
Finding D for supervision requirements by level of supervision.

       Each task performed by an agent is assigned an average time value. The
tasks and corresponding time values were the products of focus group data collec-
tion sessions with experienced field agents in 2002.6 These time values are multip-
lied by the frequency each task is performed per month. The case type affects the
time value of direct supervision duties. A greater value is allocated to sex offender
cases and a lesser workload value is assigned to cases residing in Community Cor-
rections Centers (CCCs). Additionally, travel time is applied to each case, with
larger travel time values assigned to parole cases with rural addresses than those
with urban addresses. Each type of investigation (e.g., pre-parole, pardon board,
pre-sentence) is also assigned a workload value and is added into the total workload
value upon assignment.

       Therefore, workload is the net effect of time required to supervise individual
parole cases, accounting for travel and investigations. The PBPP has established
thirty-six (36) combinations of workload values available for assignment to individ-
ual parole cases, predicated upon case type, supervision level, and parolee residence
locale. For example, the workload value for an offender under general supervision,
assessed at medium risk, and living in an urban setting, would be 1.62 hours per
month.
                                                            
6According  to Mr. Burrell, a time study of field agent tasks would provide a more accurate assessment than the
estimated time values from a focus group. However, he also noted that a time study would be a more expensive
undertaking. In 2002, the PBPP used a focus group approach to determine workload values. Thirty-two (32)
field agents were selected and asked to estimate times necessary to perform tasks associated with supervising
offenders; e.g., contact times, travel times, other supervision tasks, investigations tasks, and miscellaneous du-
ties. Five groups of 8 – 15 agents each with an average 11 years experience were formed based on the type of
offender supervision: general, Community Corrections Center, and sex offender. The general supervision group
was further separated into sub-groups for urban, rural, and mixed locations. Criteria used for urbanity desig-
nation were population and/or population density by zip codes. Data was consolidated into logical groupings,
outliers were removed, and average workload values were calculated. Prior to 2002, time study analyses were
used to determine workload values.

                                                               34
               
       While workload represents the time required to supervise a caseload, work-
load capacity is the time available to perform these functions. Field agents with the
PBPP are forty-hour-a-week employees, which equates to 2,080 hours per year, or
173 hours per month total time on the job. Leave time and required training sub-
tract from the total time available, resulting in a time value capacity of 131 hours
available to perform the duties comprising workload.

       Agent workload figures at or below 131 hours are perceived as having enough
time to complete all duties for their cases. For those agents with workload figures
in excess of 131 hours, the opposite is true. While caseload size may grow as of-
fender populations increase, workload capacity is a set figure as there are only so
many working hours available in each day, week, month, or year for each agent. 7
Parole managers identify excess workload issues during routine staff inspections
and encourage workload equalization efforts at the local level. These figures also
drive agency complement requests during the budget cycle in response to an in-
creasing offender population (See Finding B for a discussion of the use of workload
in budget requests). Exhibit 4 is an example of the workload calculation for one
case. A field agent’s total workload is the sum of the calculations for each offender
in the agent’s caseload.
                                                                      Exhibit 4
                                                       Calculation of Weighted Workload
              Standard
              Calculation:                  Supervision contact time (level of supervision and type of case) +
                                            Standard supervision duties + other supervision duties + travel time

              Example:                      Type of Case = General Supervision Case
                                            Contact Requirements/Level of Supervision = Maximum
                                            Locality = Mixed

                          Supervision contact time (2 contacts +
                  one telephone contact + 2 collateral contacts +
                               one collateral telephone contact)                  =   86.2 minutes
                                                             Plus
                                    Standard supervision duties                   =   11.53 minutes
                                                             Plus
                                       Other supervision duties                   =   14.05 minutes
                                    Total supervision work units                  =   111.6 minutes or 1.86 hours
                                                             Plus
                                                 Locality (mixed)                 =   30 minutes or .50 hours

                                                               Total work unit    =   1.86 hours + .50 hours =
                                                                                      2.36 hours
Source: PA Board of Probation and Parole.


                                                            
7According   to the Deputy Executive Director for the Office of Probation and Parole Services, the workload capac-
ity value is reviewed annually, but fluctuates between 130 and 135 hours due to agent leave time. As a result,
they try to maintain agent workloads within 10 percent of the 131-hour standard.

                                                                         35
               
                                                    Evidence-based Re-entry Focus

       The workload calculations used by the PBPP were developed prior to the shift
to an evidence-based re-entry focus for supervision. As noted previously, the evi-
dence-based re-entry focus shifts the supervision away from a “law enforcement”
process to one that seeks to change offender behavior to enable the offender to func-
tion in society. This more contemporary view of corrections, according to APPA,
embraces strategies and services that hold offenders accountable for their criminali-
ty, provides cost-effective alternatives to incarceration, and recognizes the impor-
tance of public safety in the near- and long-term.

        Evidence-based practices designed to reduce risk of re-offending rely on risk
assessments to allow probation and parole agencies to differentiate and classify of-
fenders based upon their relative level of risk to re-offend. This allows agencies to
address criminogenic needs—anti-social behavior, anti-social personality, anti-social
values and attitudes, criminal peer groups, substance abuse, and dysfunctional fam-
ily relations—through an integrated approach of surveillance, treatment, and en-
forcement. 8

        APPA also notes that evidence-based practices are rooted in an applied scien-
tific approach to determine what interventions assist agencies in reducing recidiv-
ism levels and accomplishing various intermediate outcomes, while maximizing re-
sources. Intermediate outcomes refer to offenders completing several small pro-
social tasks, such as remaining employed, paying restitution, or completing treat-
ment. The completion of several of these goals contributes to an offender remaining
crime-free.

       According to one national expert, the evidence-based re-entry model is the
approach being used in many states. She has worked with the National Institute of
Corrections (NIC) to implement re-entry and evidence-based practices in eight
states since 2001. 9 NIC has worked with numerous research teams and agencies to
develop and implement an integrated evidence-based practices model. It should be
noted that this model focuses on those evidence-based practices that reduce risk of
re-offending and focus less on some of the other important evidence-based practices
that are more atonement-related; e.g., restitution collection and community service
work, focusing instead on substance abuse treatment, vocational education, life
skills management, and other opportunities to teach offenders to be self-sufficient.

                                                            
8Criminogenic   needs are specific factors that contribute to risk for an individual offender. Needs that are
strongly connected to criminality are: criminal thinking, anti-social attitudes, association with other offenders,
poor decision making, and substance abuse. According to research, treatment should be focused directly upon
those needs assessed as most significant for a given offender. Two offenders may be equally high risk, but may
have very different needs profiles.
9The states were Georgia, Indiana, Michigan, Missouri, New York, North Dakota, Oregon, and Rhode Island.

Recently, six additional states, Iowa, Kentucky, Minnesota, Tennessee, Texas, and Wyoming, were selected to
participate in the next phase of the Transition From Prison to Community Initiative.

                                                                 36
               
Impact of Evidence-based Re-entry
       The evidence-based re-entry model approach can be time-consuming. For ex-
ample, aiding offenders to secure appropriate treatment services, locate adequate
housing, obtain employment, and address criminogenic thinking and cognitive be-
havior issues, can involve several phone calls to or meetings with service providers
and can be particularly challenging in those geographic areas of the state where
there are few services available. The PBPP recognized that re-entry tasks would
increase agent time with each offender and impact agent workloads. To try to offset
this change, the PBPP re-instituted a “streamline committee” to investigate ways to
reduce agent hours in other areas. Several strategies were identified, including es-
tablishing hearing waivers and doing home plan investigations for only those in-
mates that have been approved for release. These two ideas alone were estimated
to save almost 22,000 agent hours annually.

       One of the expected effects of the re-entry approach to supervision is a de-
crease in the number of technical parole violations (TPV) 10 for which offenders are
reincarcerated. Evidence-based re-entry practices emphasize the use of alternative
sanctions to reincarceration when parolees can be safely supervised in the commu-
nity. According to an expert with the Center for Effective Public Policy, research
has shown that if the focus of supervision is on catching offenders committing tech-
nical violations and reincarcerating them, the offenders will not successfully com-
plete positive behavior changes sought by the applicable programs. She noted that
reincarceration for a TPV may disrupt community stability and in the long run be
less effective for the offender. The PBPP’s Chairman’s presentation at the 2009
House Appropriations Committee Budget Hearing echoed this by stating that, “pa-
rolees who are returned to prison for technical violations are more likely to fail if re-
paroled later compared to parolees who are able to maintain active ties within their
community.”

       The PBPP uses a “graduated continuum of sanctions to redirect errant of-
fender behavior.” Parole supervision staff use sanctions to address minor infrac-
tions and/or negative changes in offender behavior. The sanctions are progressive
and can range from a warning letter to curfew imposition to electronic monitoring to
incarceration. A Violation Sanctioning Grid was developed to assist staff with their
response to violations. The grid lists the various violations, allows the agent to ca-
tegorize the seriousness of the violation, and identifies possible sanctions. In addi-
tion to the severity of violation, staff considers the offender’s risk level and adjust-
ment while on supervision when deciding the appropriate response. In September
2009, the PBPP introduced and began training staff to use a refined sanctioning
grid that includes identifying criminogenic needs as well as additional sanctions
that address those needs.


                                                            
10A   TPV is a failure to comply with parole conditions.

                                                               37
               
      As Table 11 shows, formal Board action to re-incarcerate for a TPV has de-
creased between FY 2005-06 and FY 2008-09 by 719 offenders even though the of-
fender population increased. Formal action for TPVs as a percentage of the state
sentenced population also decreased during that time. 11 This is not to suggest that
the TPVs did not occur, only that they are being addressed through means other
than reincarceration.

       Also shown on Table 11, is the number of criminal parole violations (CPV)
that resulted in formal Board action over the same time period. 12 In this case, the
CPVs increased by 298 from FY 2005-06 to FY 2008-09. The rate of CPVs has
grown from 0.63 percent of the state sentenced population in FY 2005-06 to 0.72
percent in FY 2008-09.
                                                               Table 11
                  Comparison of State Sentence Parole Violator Field Case Closures
                                  Over the Past Four Fiscal Years
               

                                                               FY 2005-06   FY 2006-07   FY 2007-08   FY 2008-09

State Sentence TPV Field Case Closuresa
Annual State Sentence TPVsb .........................            3,808        3,247        2,846          3,089
Average State Sentence Populationb ...............              21,567       21,688       22,310         22,274
TPVs as % of State Sentence Population........                  1.47%         1.25%        1.06%         1.16%

State Sentence CPV Field Case Closuresa
Annual State Sentence CPVsb .........................            1,626         1,648       1,738          1,924
Average State Sentence Populationb ...............              21,567       21,688       22,310         22,274
CPVs as % of State Sentence Population .......                  0.63%         0.63%        0.65%         0.72%
_______________
a
 Closures refers to formal action by the PA Board of Probation and Parole.
b
 This includes only state-sentenced offenders. Special county probation and parole and other states’ probation and
parole offenders are not included in this number. A CPV may include a TPV.

Source: PA Board of Probation and Parole.


       In her FY 2009-10 Budget Presentation before the House Appropriations
Committee, the PBPP Chairman notes that Pennsylvania’s one-year recidivism
rate has decreased from 28 percent in 2003 to 21 percent by 2007 and the three-year
rate of 49 percent is below the national average of 50 percent. 13 Also reported in
the presentation are an increasing offender employment rate and an increase in the
number of successful completions of parole.
                                                            
11In both cases, the actual number of offenders and percentage of offenders, the numbers increased from FY
2007-08 but showed an overall decline from FY 2005-06. Board officials attribute the recent increase, in part, to
the police officer homicides in Philadelphia that resulted in the moratorium on parole in September 2008.
12A CPV is a new crime committed by the parolee.
13Recidivism is defined by the PBPP as when offenders are returned to prison for new crimes or technical viola-

tions.

                                                                  38
               
Workload Assignment
       We reviewed the workloads by district office over time using the monthly re-
ports produced by the PBPP. See Exhibit 5. Looking at these numbers we deter-
mined that during CYs 2006 to 2008, the statewide average workload, when calcu-
lated using filled agent positions, decreased and is now at or below workload capaci-
ty. Of note is the sharp decline in the workload starting in September 2008. This
decrease coincides with an increased agent complement and the moratorium on of-
fender release called for by the Governor and agreed to by the Board from Septem-
ber 2008 through December 2008.

       As with caseload, the average workload for the district offices can vary signif-
icantly from the statewide average. Overall, however, the districts each showed a
decline in the average workload to at or below workload capacity. The impact of the
moratorium in 2008 is even more pronounced for the individual districts.

       We also looked at the workloads of individual supervision units as calculated
by PBPP during the annual inspections of these units. As noted earlier, the trend
line for statewide and district average workloads has been decreasing and is now
below the monthly capacity level of 131 hours. However, among individual units
within the districts, there is considerable variation in workload levels. For exam-
ple, 14 of the 32 inspection reports we examined had units with average workloads
in excess of 131 hours. This was especially true of Philadelphia, where 8 of 15 units
had average workloads greater than 131, and for 6 of these units, the average work-
load was more than 10 percent greater than the workload capacity.
       One unit in the Western Region with a six-agent complement had one agent
on military leave, one agent on medical leave for almost one year, one agent on a
short (two month) medical leave, and one agent on a two-week maternity leave dur-
ing the 12 month time frame of the inspection. The caseload for the agent on mili-
tary leave was distributed among the other five agents, and the caseloads for the
agents on medical leave were each divided between two other agents. Additionally,
one agent serves as a firearms instructor and is frequently away providing training.
As a result, the workloads for the unit’s agents ranged from 150 to 180 hours.

       The Philadelphia District experienced many of the same problems as the oth-
er districts and has had to deal with both reduced agent numbers and much higher
workload levels. For example, one inspection report identifies a unit that at the
time of the audit had only three active field agents due to one agent on extended
medical leave, one agent resignation, and one agent transfer. The average workload
for these three remaining agents was 225. All field contacts for this unit were sus-
pended about a month prior to the inspection, and all offenders were told to report
to the office for supervision. Also at this time, approximately 500 of the unit’s cases
were being transferred to other units. For the eight supervision units in the Phila-
delphia District with high average workloads, the inspection reports noted a total of
four vacancies, two medical leaves, two military leaves, and two agent transfers.

                                          39
       
                                                                                   Exhibit 5


                                                               Statewide Average Workload
                                                                            CYs 2006-2008
Average Workload Hours




                         160
                         150
                         140
                         130
                         120
                         110
                         100
                                    Jan       Feb       Mar         Apr      May          Jun         Jul       Aug        Sep       Oct       Nov       Dec
         2006                      129        129       132         128       132        133         132        133        133       137       136       137
         2007                      135        133       134         130       134        136         137        135        132       132       135       136
         2008                      124        124       123         123       124        123         126        126        127       122       118       118
         Capacity                  131        131       131         131       131        131         131        131        131       131       131       131

                                                             2006             2007                 2008                Capacity




                                                             Eastern Region Average Workload
                                                                     by District Office
                                                                          CYs 2006 - 2008
Average Workload Hours




                         160
                         150
                         140
                         130
                         120
                         110
                         100
                               Jan‐06


                                          Apr‐06


                                                    Jul‐06


                                                                Oct‐06


                                                                          Jan‐07


                                                                                     Apr‐07


                                                                                                 Jul‐07


                                                                                                            Oct‐07


                                                                                                                       Jan‐08


                                                                                                                                 Apr‐08


                                                                                                                                           Jul‐08


                                                                                                                                                     Oct‐08




                                                                    Capacity                  Chester                 Phila




                                                                                      40
                          
Exhibit 5 (Continued)



                                                               Central Region Average Workload
                                                                       by District Office
                                                                           CYs 2006 - 2008
   Average Workload Hours




                            160
                            150
                            140
                            130
                            120
                            110
                            100
                                  Jan‐06


                                            Apr‐06


                                                      Jul‐06


                                                                  Oct‐06


                                                                           Jan‐07


                                                                                    Apr‐07


                                                                                                Jul‐07


                                                                                                         Oct‐07


                                                                                                                     Jan‐08


                                                                                                                              Apr‐08


                                                                                                                                           Jul‐08


                                                                                                                                                    Oct‐08
                                    Capacity                   Allentown            Harrisburg                    Scranton               Williamsport




                                                               Western Region Average Workload
                                                                       by District Office
                                                                           CYs 2006 - 2008
   Average Workload Hours




                            160
                            150
                            140
                            130
                            120
                            110
                            100
                                  Jan‐06


                                            Apr‐06


                                                      Jul‐06


                                                                  Oct‐06


                                                                           Jan‐07


                                                                                    Apr‐07


                                                                                                Jul‐07


                                                                                                         Oct‐07


                                                                                                                     Jan‐08


                                                                                                                              Apr‐08


                                                                                                                                           Jul‐08


                                                                                                                                                    Oct‐08




                                           Capacity                  Altoona                 Mercer                Erie                Pittsburgh



Source: Developed by LB&FC staff using CYs 2006 – 2008 data published monthly by the PBPP.

                                                                                     41
                             
      Nevertheless, the average workload for all of the units in the inspection re-
ports we examined was 133 hours, with a median value of 128. Similarly, caseloads
averaged 86, with a median value of 85. See Exhibit 6.

              Use of Caseload and Workload Data to Balance
                        Field Agent Assignments
       Supervisors, deputy district directors, district directors, and regional direc-
tors, as well as certain central office personnel, have access to a daily computer-
generated report that shows the caseload and workload levels of every field agent
currently supervising a caseload for purposes of addressing caseload and workload
balance. It is point-in-time data that is not maintained for trend purposes and does
not include temporary adjustments that may be made to address workload bal-
ances. It is reviewed monthly at the central office. In addition, the district direc-
tors meet with their deputies or supervisors to discuss caseload and workload on a
regular basis. High caseloads and workloads can be the result of an increase in of-
fenders in a particular geographic area or the leave status of an agent or a vacancy.
In the case of an extended leave or vacancy, the increase in cases can be dramatic,
with little opportunity to prepare for the leave or vacancy. For example, one of the
regional directors cited a situation where the office had one day’s notice that the
agent was leaving for a six-week military assignment.

        As potential problems are identified, cases may be reassigned to another
agent. The district directors are primarily responsible for making these decisions,
but if the leave is for an unknown period of time, the decision to transfer cases is
more difficult since it may only be for a short duration. Adding to the difficulty of
transferring cases is that prior to transfer, cases may need to be brought up to
date with, for example, the appropriate sign-offs. It can also be difficult to transfer
cases since most of the districts assign cases based on geographic location. In rural
districts it may be especially difficult to reassign a caseload since it may involve
significant travel time for the receiving agent, negatively impacting that agent’s ex-
isting caseload and workload. Additionally, if it’s a specialized caseload, e.g., sex-
offenders, it may be difficult to reassign. A PBPP official noted that certain high-
risk cases may be covered temporarily by the supervisor or directed to report to the
office (versus a field contact).

       One of the district directors pointed out that while he tries to balance the
caseload and workload distribution among his field agents, he is aware that some
agents are capable of handling higher workloads than others, and takes that into
consideration when making those decisions. In general, the regional director and
district directors find it less disruptive to the offenders and agents to readjust as-
signments among field agents when new staff are assigned rather than each time
the situation occurs.



                                           42
       
                                                       Exhibit 6
                                 Average Caseload and Workload by Unit*

                                                          Average     Case         Average         Workload
       District                   Unit           Agents   Caseload    Range        Workload         Range
Chester .............      Delaware Unit           8         75       55 - 93         109         82.3 - 139.8
Philadelphia ......        West Unit 3             5         78       65 - 87         116         96.0 - 144.7
Philadelphia ......        Northeast Unit 2        6         91       84 - 98         127        121.8 - 141.8
Philadelphia ......        Northeast Unit 3a       3        225       34 - 191        326         37.5 - 285.0
Philadelphia ......        Northeast Unit 4        4         85       34 - 166        115         37.5 - 232.8
Philadelphia ......        Northeast Unit 5        7         90        6 - 129        129          6.0 - 187.3
Philadelphia ......        Northeast Unit 6        5         87       46 - 127        122         61.3 - 188.1
Philadelphia ......        South Unit 2            4        104       97 - 112        152        138.5 - 172.3
Philadelphia ......        Northwest Unit 5        4        104       90 - 124        154        129.8 - 186.2
Philadelphia ......        West Unit 1             6         95       55 - 119        132         85.2 - 158.4
Philadelphia ......        West Unit 2             4        124       92 - 138        169        124.9 - 189.1
Philadelphia ......        CCC Unit 2              5         88       77 - 116        138        110.7 - 189.5
Philadelphia ......        Northwest Unit 4        5        102       72 - 120        143        108.9 - 172.4
Philadelphia ......        Northwest Unit 3        3         42       22 - 79          65         39.0 - 113.0
Philadelphia ......        Northwest Unit 1        5        125      113 - 138        175        158.8 - 199.7
Philadelphia ......        CCC Unit 1              7         66       42 - 77         107         76.9 - 127.4
Allentown ..........       Allentown Unit 4        4         56       48 - 67          94         75.3 - 110.5
Allentown ..........       Allentown Unit 3        7         85       61 - 105        125         84.9 - 148.1
Harrisburg .........       Chambersburg SO         7         60       50 - 75         100         86.3 - 111.5
Harrisburg .........       Dauphin Unit 4          6         90       21 - 144        144         23.2 - 193.5
Williamsport ......        Williamsport Unit 2     7         63       45 - 75         111         83.9 - 136.5
Williamsport ......        Williamsport Unit 1     8         61       17 - 76         104         27.9 - 138.5
Altoona .............      North General           6         67       31 - 80         134         58.7 - 167.5
Erie ...................   Erie West               6         91       86 - 96         144        132.4 - 150.0
Erie ...................   Erie East               7         69       42 - 86         119         91.6 - 141.0
Mercer ..............      Franklin SO Unit 2      5         72       65 - 78         128        110.4 - 136.6
Mercer ..............      Mercer North            5         75       52 - 85         126        114.6 - 136.0
Mercer ..............      Mercer South            6         83       72 - 95         140        120.3 - 154.1
Pittsburgh .........       Mon Valley South        4         95       94 - 98         152        140.8 - 160.0
Pittsburgh .........       East Allegheny          5         52       30 - 65          81         53.1 - 94.5
Pittsburgh .........       Greater Washington      5         59       54 - 66         109         98.3 - 116.7
Pittsburgh .........       Beaver Falls SO         4         87       81 - 97         158        138.0 - 174.8




_______________
*As calculated by PBPP for unit inspections conducted July 2007 through May 2009.
a
 The caseload and workload ranges included an agent on long-term medical leave. The average caseload and work-
load for the unit was calculated using three active agents although the range included the caseloads of all four
agents.

Source: Developed by LB&FC staff using data from a sample of PBPP Unit Inspection Reports.


                                                          43
            
Caseload/Workload Distribution in Other States

       LB&FC staff surveyed the probation and parole agencies in the contiguous
states and in states the PBPP identified as having comparable organizations to de-
termine how agent caseload and workload distribution was managed. No consistent
approach is used by the other states we contacted. For example, in Delaware cases
are assigned on a rotating basis by level of supervision and the number of cases
each officer is currently supervising. The average caseload per officer is 81 for Level
2 supervision (standard supervision) and 41 for Level 3 supervision (intensive su-
pervision). For officers supervising sex offenders, the average caseload per officer
for Level 2 is 69, and the average caseload for Level 3 is 35.

       In Georgia, work is assigned to the officers by the total number of cases and
the geographic area supervised by each officer. The Board does not have a uniform
standard for assigning work citing the many factors affecting the officer’s ability to
effectively supervise a designated caseload, for example, offender risk, local condi-
tions such as unemployment rates, and other duties that the officer may be as-
signed. The Board’s statewide average caseload per officer is currently 72, although
the average is over 100 cases per officer in some offices. The goal is an average
caseload of 60 per officer. In West Virginia, work is assigned to the agents based on
caseload. Currently, agents have an average caseload of 57.

        Michigan agents carry special caseloads based on the type of offender, with
work assigned to agents based on a combination of caseload and workload factors.
Generally, the average caseload carried by each agent is 90 cases per agent. 14 For
agents supervising sex offenders who have been sentenced to GPS supervision, the
caseload limit is 25 cases per agent. Under the State Parole Board in New Jersey,
work is assigned to the parole officers on a caseload basis, apportioned based on
overall unit responsibilities, risk factors, geography, and available resources. Cur-
rently, the average caseload per officer is 47 offenders. New York has no overall ca-
seload or workload standard for assigning cases to supervising agents for proba-
tion. 15 Each of the state’s 58 counties has its own criteria for assigning cases to
agents. The only statewide standard relates to offenders assigned to the intensive
supervision program. Under this ISP program, there is a caseload limit of 21 cases
per agent.

        Ohio’s Department of Rehabilitation and Correction has a written policy that
sets out its workload formula for the allocation of work assignments to its parole of-
ficers. Supervisors are required to evaluate parole officer workload within each re-
gion quarterly to maintain an equitable distribution of workload among units and
parole officers. Factors in the workload formula include:

                                                            
14TheMichigan Department of Corrections administers the parole and the probation function for the state.
15New York has a state agency responsible for supervising parole. We were unable to determine their caseload
and workload standards.

                                                               44
               
      •   offender risk level and need;
      •   offender release type;
      •   geographic area of supervision;
      •   training hours needed;
      •   special assignments of field staff;
      •   type of investigations assigned and completed;
      •   number and frequency of violation hearings attended;
      •   caseload size;
      •   caseload specialization; and
      •   number of active parole officers per unit.

      The Maryland Department of Public Safety and Correctional Services assigns
work to its agents with a caseload model that is based on evidence-based practices.
The Department stated that 99 is the average caseload per agent for general cases
supervised using a case plan. For special offenders such as sex offenders, however,
the Department limits caseload size to 30 cases per officer. These special offenders
are supervised based on a containment model and contact standards are required.




                                            45
       
D. The Standards for Offender Supervision and Frequency of
Contacts Are Uniform Among District Offices, but Staffing Levels
Affect Actual Practice


       The PBPP uses standard screening tools to determine the likelihood an of-
fender will re-offend. That likelihood translates into the level of supervision appli-
cable to the offender and, with it, the frequency of contacts and other standards for
supervision. The actual supervision an offender receives, however, is dependent on
the individual field agents. PBPP internal audits have identified instances where
certain supervision standards were not met by the field agents. These included
both administrative (e.g., failure to record start and stop times for contacts) and op-
erational (e.g., insufficient offender contacts) deficiencies. In many of these in-
stances, the inspection reports also noted staffing constraints due to vacancies in
the supervision unit which, in turn, created high caseload and workload levels.
These difficulties occurred throughout the state but were more significant in the
Philadelphia supervision units.

                  Assessment Tools Used to Determine Risk of Reoffending

        According to The Institute on Crime, Justice and Corrections (ICJC) at
George Washington University, assessments can provide an accurate prediction of
how individuals with similar characteristics might behave in the future. These as-
sessments are predicated on the assumption that individual criminal behavior can
be predicted by assigning offenders to groups that have explicit re-offending proba-
bilities. The authors noted that the PBPP selected the Level of Service Revised
(LSI-R) instrument as its risk classification tool because it introduces dynamic and
more current factors into the risk assessment process, beyond the conventional use
of static criminal history and demographic factors.1

       In its Risk and Needs Assessment Guidelines, the Pennsylvania Department
of Corrections defines the LSI-R as “an objective, quantifiable instrument that pro-
vides a consistent and valid method of predicting risk to re-offend, and is a reliable
means of measuring an offender’s change over time through re-administration. The
LSI-R provides insight into which offenders should receive the highest priority for
treatment regardless of their specific problem areas.” According to the PBPP’s pro-
cedures manual, the LSI-R assessment helps predict parole outcome, success in cor-
rectional halfway houses, institutional misconducts, and recidivism. The results of



                                                            
1TheWisconsin Risk Assessment tool was used by PBPP for at least 10 years prior to the adoption of the LSI-R.
The Wisconsin assessment tool was based on static factors that did not change.


                                                               46
               
the LSI-R assessment establish the level of supervision and identify offender needs
for the development of a supervision plan. 2

       The LSI-R is comprised of 54 static and dynamic items across ten sub-scales.
Through an interview process prior to release, offenders are rated on items requir-
ing either a “yes/no” response or the use of a scale ranging in value from 0 to 3.
Based on the responses, the interviewer scores the offender on each item, totals the
item scores, and determines the offender’s overall risk level using the scale adopted
by the PBPP. Following a validation study by ICJC in 2003, the Board approved
new decision-making guidelines with LSI-R as its risk assessment instrument, and
in September 2003 began training supervision field staff to use the LSI-R. The
PBPP reviewed the application of the LSI-R in 2008. 3

       Based upon the results of the assessment and/or any special conditions that
have been imposed, a level of supervision is determined and an initial supervision
plan is developed. As of December 2008, all offenders are supervised at the Maxi-
mum level for the first 90 days of their parole. 4 Level of supervision is routinely re-
viewed by the field agent and is required to be reviewed at least once every six
months with a supervisor. The LSI-R is also re-administered annually to validate
agent impressions and as a measure of offender change.

Assessment Tools Used in Other States

        We surveyed the probation and parole agencies in the contiguous states and
in states the PBPP identified as having comparable organizations to determine
what screening tools were being used in these states. See Exhibit 7. For example,
Georgia’s Department of Corrections Probation Operations Office uses a “risk-based
provision assessment” developed in-house to determine the appropriate level of
supervision for each offender. The Office uses COMPAS 5 to develop case plans
for offenders that have been assessed at “high risk.” In New Jersey, the State Pa-
role Board uses the LSI-R and the Static-99 assessment tools to determine level of

                                                            
2Current  risk assessments are able to assess the overall likelihood to re-offend, but not specifically the like-
lihood of violent re-offending. PBPP is working with external experts to develop a violent risk assessment tool.
3Undertaken in October 2008, the in-house study looked at contemporary case management methods and re-

sults over the preceding three years and examined the outcomes of 25,688 offenders released to state supervi-
sion between January 2005 and October 2007. This study found that the distribution of cases within risk level
had shifted from 41 percent high risk in 2005 to 30 percent high risk in 2007; there was a direct relationship
between ascending LSI-R scores and increasing recidivism rates; violent and non-violent offenders recidivated
at similar rates; and that the aggregate LSI-R scores were normally distributed with 99 percent of the scores
within three standard deviations of the mean.
4In response to recommendations made by Dr. John Goldkamp in his December 1, 2008, report, Restoring Parole

and Related Processing for Categories of Violent State Prisoners: Findings and Recommendations II, the PBPP
requires those offenders in the “most violent” category to be released to a community corrections center (with
certain individual case exceptions) and supervised at maximum supervision with curfew for 90 days and all
state-sentenced offenders to be supervised at maximum supervision with curfew for 90 days.
5COMPAS - Correctional Offender Management Profiling for Alternative Sanctions.




                                                               47
               
supervision for offenders. It reports that it uses an evidence-based approach to su-
pervision that uses risk assessment, case planning, and diversionary programming
to deal with technical parole violations (TPV) and has been successful in reducing
the number of TPVs returned to state prison.

       All counties in New York (Probation), except New York City (NYC), use a
modified version of COMPAS to determine classification, case planning, and level of
supervision. The tool used by NYC is strictly to assess risk of recidivism. Ohio’s
Adult Parole Authority uses the Static Risk Assessment and the Dynamic Needs
Assessment tools to determine the level of supervision based upon the risk posed by
the offender to the community and the needs of the offender. West Virginia uses an
evaluative tool that it developed in-house to determine the level of supervision. De-
laware’s Bureau of Community Corrections uses the Sentencing Accountability
Commission (SENTAC) classification system to determine the level of supervision
for each offender. The Delaware SENTAC was created by statute and directed to
establish a system that emphasizes accountability of the offender to the criminal
justice system and accountability of the criminal justice system to the public. In
addition to the LSI-R and Static – 99 tools, the Maryland Department of Public
Safety and Correctional Services uses an instrument developed in-house to deter-
mine an offender’s appropriate level of supervision.

                                                             Exhibit 7
                                           Assessment Instruments Used
                                         to Determine Level of Supervision
  
                           State                                Assessment Instruments Used
            Delaware .........................       SENTAC
            Georgia ............................     COMPAS, In-Housea
            Maryland ..........................      LSI-R, Static-99, and In-House Instrument
            Michigan ..........................      COMPAS
            New Jersey......................         LSI-R, Static-99, In-House, APPA
            New York .........................       COMPAS
            Ohio .................................   Static-99, Criminal History/Risk Score
                                                     Sheet/Dynamic Needs Assessment
            Pennsylvania ...................         LSI-R, Static – 99 used for sex offenders
            West Virginia ...................        In-House
_______________ 
aLevel of supervision is determined by risk assessment that automatically generates a score on every released in-
mate. Overrides on some parolees upward are based on certain criteria such as sex offenses. All sex offenders are
considered high risk.

Source: Developed by LB&FC from information provided by probation and parole agencies for the identified states.




                                                                48
         
                              Level of Supervision

       The PBPP has established minimum requirements for each of level of super-
vision. As would be expected, as the level of supervision increases, so does the fre-
quency of contact. See Exhibit 8. The increasing contact requirements are reflected
in the values used to calculate agent workload (discussed in Finding C). Each of-
fender added to an agent’s caseload also adds to the agent’s workload with higher
level of supervision offenders adding more significantly to the workload.

Geographic Distribution of Offender Population by Level of Supervision

       To determine if the supervised offender population is distributed equitably
across the state, we examined offender population by level of supervision for each
district office for CYs 2006 to 2008. Other than the enhanced level of supervision,
each district office has maintained approximately the same percentage of the popu-
lation at each supervision level during this time period. See Exhibit 9.

        The enhanced level of supervision is imposed by the Board at its discretion
and, as such, shows the most volatility. This is evident with the Philadelphia Dis-
trict’s percentage of the enhanced population, which ranged between 23 percent and
36 percent of the offender population during CY 2006, between 26 and 41 percent
during CY 2007, and between 14 and 26 percent during CY 2008. The increase in
the enhanced level in 2009 (see Finding B) is attributed to the Board implementing
Dr. Goldkamp’s recommendations and responding to the concerns in Philadelphia
due to parolee violence.




                                         49
       
                                                  Exhibit 8
                           PBPP Supervision Contact Requirements
 
     Level of Supervision                   Contact Requirement                         Condition
                                                                              One of which may be in the office
                                                                              and one of which must be at the
                                       4 Face-to-Face Contacts per Month      approved residence. These face-
                                                                              to-face contacts must be propor-
                                                                              tionately spaced throughout the
            Enhanced
                                                                              month so that they do not all oc-
                                                                              cur during the same week.

                                       2 Collateral Contacts per Month        One must be face-to-face.


                                                                              No more than three of which may
                                       6 Face-to-Face Contacts per Quarter    be office contacts, two must be at
                                                                              the approved residence and, at
            Maximum                                                           least one face-to-face contact
                                                                              each calendar month.

                                       2 Collateral Contacts per Month        One must be face-to-face.

                                                                              One of which may be an office
                                       3 Face-to-Face Contacts per Quarter    contact, one must be at the ap-
                                                                              proved residence and no more
                                                                              than one month lapsing without a
             Medium                                                           face-to-face offender contact.

                                       3 Collateral Contacts per Quarter      One must be face-to-face, with
                                                                              no more than one month lapsing
                                                                              without a collateral contact.

                                       One Face-to-Face Contact per Quarter   At least every other face-to-face
                                                                              contact must be at the approved
                                                                              residence.
            Minimum
                                       One Face-to-Face Collateral Contact    This collateral contact is not to
                                       per Quarter                            occur during the same month as
                                                                              the face-to-face offender contact.

                                       One Face-to-Face Contact per Six       At least every other face-to-face
                                       Months                                 contact must be at the approved
    Administrative/Special                                                    residence.
       Circumstance                    One Face-to-Face Collateral Contact    The collateral contact is not to
                                       per Six Months                         occur during the same quarter as
                                                                              the face-to-face offender contact.
 
 
 
 
 
Source: PA Board of Probation and Parole.


                                                     50
         
                                                      Exhibit 9
                                Percent of Offenders by Level of Supervision as a
                                  Percent of Statewide Total, by District Office*
                                                (CYs 2006 – 2008)
 
                                               2006                     2007                2008
    Allentown ......                    Average    Range      Average      Range     Average    Range
                             Enhanced     5%        2 - 8%         8%      4 - 13%    12%       9 - 16%
                             Maximum     10%       9 - 11%        11%     9 - 14%     12%      11 - 14%
                             Medium      11%      10 - 12%        13%     12 - 13%    11%      10 - 12%
                             Minimum     11%      11 - 11%        11%     10 - 11%    11%      10 - 12%

                                               2006                     2007                2008
    Altoona ..........                  Average    Range      Average      Range     Average    Range
                             Enhanced     6%      2 - 10%         3%       2 - 6%      2%       1 - 5%
                             Maximum      3%       2 - 3%         3%       3 - 4%      4%       4 - 5%
                             Medium       5%       4 - 5%         6%       5 - 6%      6%       6 - 7%
                             Minimum      4%       4 - 5%         4%       4 - 4%      4%       3 - 4%

                                               2006                     2007                2008
    Chester ..........                  Average    Range      Average      Range     Average    Range
                             Enhanced     9%      5 - 14%          7%      4 - 11%    10%      7 - 12%
                             Maximum      9%       9 - 9%          8%      6 - 9%     4%        4 - 6%
                             Medium       9%       8 - 9%         10%      9 - 11%    10%      9 - 11%
                             Minimum      7%       6 - 7%          7%      6 - 7%     9%       8 - 9%

                                              2006                  2007                   2008
    Erie .................              Average   Range       Average   Range        Average   Range
                             Enhanced     5%       3 - 8%         2%       0 - 5%      3%       1 - 5%
                             Maximum      3%       3 - 4%         4%       3 - 4%      5%       4 - 6%
                             Medium       4%       4 - 5%         5%       5 - 5%      6%       5 - 6%
                             Minimum      5%       5 - 5%         5%       5 - 5%      4%       4 - 5%

                                              2006                  2007                   2008
    Harrisburg .....                    Average   Range       Average   Range        Average   Range
                             Enhanced    11%       5 - 15%        12%      9 - 17%    20%      11 - 26%
                             Maximum     12%      11 - 13%        13%     12 - 14%    13%      12 - 14%
                             Medium      11%      10 - 11%        12%     11 - 12%    12%      12 - 14%
                             Minimum     15%      15 - 16%        13%     11 - 15%    12%      12 - 13%




                                                         51
              
Exhibit 9 (Continued)

                                           2006                       2007                      2008
 Mercer ............                 Average   Range            Average   Range           Average   Range
                       Enhanced         3%         1 - 4%          2%        1 - 3%          3%         1 - 5%
                       Maximum          4%         4 - 4%          4%        3 - 5%          3%         3 - 4%
                       Medium          11%        11 - 11%        11%       10 - 11%        11%        11 - 12%
                       Minimum         13%        13 - 14%        13%       12 - 13%        12%        12 - 13%

                                           2006                       2007                      2008
 Philadelphia ..                     Average   Range            Average   Range           Average   Range
                       Enhanced        29%        23 - 36%        35%       26 - 41%        20%        14 - 26%
                       Maximum         40%        39 - 41%        36%       30 - 40%        33%        31 - 35%
                       Medium          26%        23 - 29%        21%       19 - 23%        19%        17 - 21%
                       Minimum         20%        19 - 22%        25%       23 - 29%        29%        28 - 29%

                                              2006                      2007                       2008
 Pittsburgh......                    Average       Range        Average       Range       Average       Range
                       Enhanced        25%        23 - 30%        25%       22 - 32%        21%        16 - 30%
                       Maximum         14%        14 - 15%        16%       14 - 18%        19%        19 - 20%
                       Medium          14%        14 - 15%        15%       14 - 15%        15%        15 - 16%
                       Minimum         11%        11 - 12%        11%       10 - 11%         9%         9 - 10%

                                              2006                      2007                       2008
 Scranton ........                   Average       Range        Average       Range       Average       Range
                       Enhanced         6%          4 - 8%         4%         2 - 5%         7%         3 - 14%
                       Maximum          3%          3 - 4%         3%         3 - 4%         4%          4 - 5%
                       Medium           6%          5 - 6%         6%         6 - 7%         6%          5 - 6%
                       Minimum          8%          7 - 8%         7%         6 - 7%         6%         6 - 6%

                                              2006                      2007                       2008
 Williamsport ..                     Average       Range        Average       Range       Average       Range
                       Enhanced         1%          0 - 2%         1%         0 - 3%         1%         0 - 3%
                       Maximum          1%          1 - 2%         2%         1 - 2%         2%         2 - 2%
                       Medium           3%          2 - 3%         3%         3 - 3%         3%         3 - 3%
                       Minimum          5%          5 - 6%         4%         4 - 5%         4%         4 - 4%

_______________
*District office percentages were calculated by dividing the number of offenders supervised by that office at each level
of supervision by the total statewide classified offender population at the same supervision level for each month of the
calendar year. The average value is over the 12 month period. The range of values is the minimum and the maxi-
mum monthly percentage for the 12 month period.

Source: Developed by LB&FC staff using data published monthly by the PBPP.


                                                          52
          
Distribution of Levels of Supervision Over Time by District

        Analysis of the classified offender population over time statewide shows a
significant increase in the number of offenders being supervised at the minimum
level of supervision beginning in 2004. See Exhibit 10. According to the PBPP
staff, this coincided with the adoption of the LSI-R assessment tool and the belief
that, up until this point, the offender population had been over-supervised.

                                                                     Exhibit 10
                                                 Classified Offender Population as of June
                                                           by Level of Supervison
                                                              CYs 2002 - 2009
                                16,000
                                14,000
   Classified* Offenders 




                                12,000
        Number of 




                                10,000
                                 8,000
                                 6,000
                                 4,000
                                 2,000
                                    0
                                          2002      2003      2004        2005    2006   2007     2008       2009

                                     Enhanced       Maximum            Medium      Minimum      Administrative


_______________
*The classified offender population does not include absconders, detainees, or offenders who have not been classi-
fied as to level of supervision.

Source: Developed by LB&FC staff from data published by the PBPP.


Changes to Level of Supervision

        According to PBPP staff, the offender’s level of supervision is routinely re-
viewed by the supervising agent. It is not unusual for an offender’s level of supervi-
sion to change. PBPP policy states that a change of one level requires the approval
of the agent and the agent’s supervisor. A change of two levels requires the approv-
al of the district director as well. We were unable to obtain data on the frequency
that changes occur, but have been told by staff that it occurs often.




                                                                        53
                             
      In 2008, a new level of supervision, administrative, was created by statute.
Act 2008-83 provides that eligible offenders 6 be placed on administrative parole one
year after release on parole and until their maximum sentence date. To be granted
such parole, it must be determined that:

              •       the eligible offender has not violated his terms and conditions of parole;
              •       the eligible offender has not been subject to extensive use of sanctions
                      while on parole; and
              •       placement of administrative parole will not compromise public safety.

      The Administrative level of supervision has the least number of contact re-
quirements—one supervision contact each year. See Exhibit 8.

       Each district office conducted case reviews from February through April 2009
to determine which of the offenders currently being supervised by them would be
eligible for the administrative level of supervision. A total of 2,190 cases were re-
viewed and 961, or 44 percent, were deemed eligible. The Eastern Region had the
highest percentage of eligible cases (49 percent) and the Western Region the lowest
(37 percent). The Western Region, however, also had the district offices with the
highest percentage of eligible cases (Mercer, 41 percent) and the lowest (Altoona, 16
percent). The reason most often cited for ineligibility was failure to pay supervision
fees (terms and conditions of parole). Administrative supervision was 6.7 percent of
the reporting population in April, and as of August 2009 it was 4.6 percent.

PBPP Efforts to Ensure Agents Supervise Offenders as Required

        As discussed in Finding C, Central Office staff conduct annual inspections of
all of the field supervision units to ensure that field agents are complying with su-
pervision standards. Each unit is audited annually and re-audits may be ordered.

       Use of Annual Inspections. All supervision units are responsible for imple-
menting the standards with regard to compliance with PBPP procedures. This is
monitored through annual unit inspections by the Bureau of Offender Re-entry
Coordination. A sample of offenders from each level of supervision is selected by
the inspecting parole managers, and, the case files and field books for these offend-
ers are examined to determine whether they are current and contain all materials
required by PBPP procedures. The units are advised to expect an inspection by a

                                                            
6Defined  as those with: (1) no history of present or past violent behavior; (2) no deadly weapon enhancement or
firearms offense conviction; (3) no personal injury crime conviction; (4) no conviction of certain enumerated of-
fenses, including sex and drug offenses committed with firearms (or an equivalent offense under federal, for-
eign, or state laws); (5) no pending charges which would make the offender ineligible; and (6) no high volume
drug trafficking conviction.


                                                               54
               
parole manager about a month prior to the conduct of the audit. The individual pa-
role managers inspect different units each year.

       If a specific deficiency is found in 10 percent or more of the cases and field
books reviewed, it is documented in the final inspection report. Inspectors docu-
ment the deficiencies found in the previous year’s inspection and compare these to
the current results to identify trends. The inspectors do not, however, review the
corrective action plan from the prior year’s audit to ensure that the corrective action
was taken. 7 Supervisors are given a draft checklist identifying any deficiencies
found and given 24 hours to initiate corrective actions. The district director pre-
pares an after-action report that details the corrective steps taken, and this be-
comes part of the final inspection report. The regional director has responsibility to
ensure corrections are implemented.

       LB&FC Analysis of the Deficiencies Noted in Audit Reports. We examined
supervision unit inspection reports from each region. We requested eight reports
each from the Western and Central Regions, all available reports from the Phila-
delphia District, and at least one additional audit report from the Eastern Region.
The Bureau of Offender Re-entry Coordination provided reports for 32 inspections
conducted between July 2007 and May 2009, with 21 of these occurring in CY 2008.
Inspection reports are retained by PBPP for only 12 months so trend analysis of
greater than two years was not possible. Due to CHRIA confidentiality concerns, all
reports were selected by PBPP staff, and prior to our receipt of the reports, PBPP
staff and offender names were redacted.

       Statewide Results. The deficiencies noted in the inspection reports we re-
viewed can be classified as either operational or administrative. Statewide, the
most common operational deficiencies identified were an inappropriate number of
offender and/or collateral contacts for level of supervision and an inappropriate
number of urine screens administered for level of supervision. The most frequent
administrative deficiencies found during these inspections were supervision plans
not signed by offender, supervision plans submitted late, and the start/stop times
not recorded for contacts.

       Seven of the 32 inspections we reviewed did not identify enough deficiencies
to meet the 10 percent criteria for triggering corrective actions. As expected, five of
these units had caseload/workload numbers, at the time of the inspections, at or be-
low the PBPP goal levels. Interestingly, however, one Philadelphia District Office
division with three of its four units having no reported deficiencies, had two units
with caseload/workload numbers well in excess of these same levels.


                                                            
7Inthe past six months the PBPP has acknowledged this deficit and is reportedly implementing an audit
process that closes this loop.


                                                               55
               
       Five of the ten units in the Western Region we reviewed had high caseload/
workload numbers and three of these units were either carrying vacant agent posi-
tions or had agents on leave during the inspection period. One unit in Erie had a
recent graduate of BTA who did not yet have a caseload, and as a result, all of the
other agents in the unit exceeded the 131 hour workload standard. The ten defi-
ciencies for this unit were evenly divided between administrative and operational.
Another unit with two vacancies, had 44 deficiencies (30 of which were for failure to
record start/stop times of contacts). Only one of the six units reviewed from the
Central Region had an average workload greater than 131, but is responsible for all
program cases and this may account for the higher numbers.

       Philadelphia Results. As with the statewide results, the most common defi-
ciencies noted in the inspection reports for the Philadelphia units were inappro-
priate number of contacts for level of supervision, inappropriate number of urine
screens for level of supervision, unsigned supervision plans, LSI-R assessments not
completed, and the start/stop times not recorded for contacts.

       Eight of the 15 units from the Philadelphia District had average workloads in
excess of the 131 hour standard. All but one of these units had vacant agent posi-
tions or agents on leave—one unit had only three active agents at the time of the
inspection. Those units with agent shortages also had numerous deficiencies identi-
fied in the inspection reports.

       The inspection results for each unit in one particular division in Philadelphia
were uniformly poor, and a re-inspection of these units was conducted in January
2009, approximately six months after the initial inspections. The results of the re-
inspection show that progress had occurred over this relatively short time period.
Caseload and workload numbers for all but one unit decreased, and the active agent
complement numbers varied slightly for individual units, but overall remained the
same. While many of the same deficiencies were found again, overall the number of
deficiencies reported by the inspectors were about one-third of those found during
the initial inspections conducted in 2008, and a majority of the re-occurring defi-
ciencies were administrative.




                                         56
       
E. The PBPP Is Acting to Address Issues With Hiring and Retention
of Field Agents Affecting the Philadelphia District Office


       The PBPP has had difficulty filling vacant field agent positions and retaining
field agents in the Philadelphia District Office. Thus, as noted already, the turn-
over rate for the district is consistently higher than the statewide average and has
resulted in that office having field agents and supervisors with limited job expe-
rience when compared to other district offices. Further, the Philadelphia District
Office has the highest percentage (7.6 percent) of recommended applicants who fail
the background checks during the hiring process adding to the difficulties in filling
the vacant positions in a timely manner. Beginning in the fall of 2007, when the
PBPP initiated an analysis of the Philadelphia workload, the PBPP has taken sev-
eral steps to address the issue of hiring and retention of field agents in the Phila-
delphia District Office.

                            Turnover of Field Agent Positions

       The Philadelphia District Office had a significantly higher turnover rate than
the statewide average for the last two fiscal years. As shown on Table 12, the
statewide average for FY 2007-08 was 11.5 percent and was just under 12 percent
in FY 2008-09. The Philadelphia turnover rate was over 25 percent for FY 2007-08
and over 16 percent for FY 2008-09. As Table 12 shows, however, the Williamsport
District Office had the highest turnover rate in FY 2008-09 at about 24 percent,
which is unusual for that district. 1 For the period October 2006 through June 2009,
the Philadelphia District Office had an average agent turnover rate of 18.7 percent,
significantly higher than the statewide average of 11.4 percent. This high ongoing
turnover rate is one factor in producing higher caseloads and workloads for the
agents who remain in the Philadelphia District Office.

       In October 2006, the Board began tracking six categories of agent transfers,
promotions, and separations (retirement, resignation, and termination) to other
agencies and inter-district transfers for the Philadelphia District. Between October
2006 and June 2009, 72 field agent equivalent staff 2 in the Philadelphia District Of-
fice took one of these actions. Of those, 30 transferred to another Board office, 13
transferred to other state agencies, 2 were promoted, 12 retired, 11 resigned, and 4
were terminated. During this same period, the turnover of field agent equivalent




1In FY 2008-09, the Williamsport office had two field agents promoted to supervisor positions and two agents
retire.
2This includes field agents and ASCRAs. Philadelphia reportedly had significant turnover due to a retirement

incentive and the creation of specialized positions.

                                                     57
                                                  Table 12
                                      Field Agent Equivalent Turnover
                                        (FY 2007-08 and FY 2008-09)

                                      FY 2007-08                            FY 2008-09
                         Authorized     Actual    As % of      Authorized      Actual      As % of
 District Office        Complementa    Turnover Complement    Complementa     Turnover   Complement
 Allentown......            52           1             1.9%       56             5          8.9%
 Altoona .........          20            0            0.0        23             1          4.3
 Chester .........          38           2             5.3        39             4         10.3
 Erie ...............       23            0            0.0        24             2          8.3
 Harrisburg ....            60           4             6.7        61             8         13.1
 Mercer ..........          44            2            4.5        46             4          8.7
 Philadelphia..            136           35           25.7       140            23         16.4
 Pittsburgh .....           68           8            11.8        69             6          8.7
 Scranton .......           28            4           14.3        30             2          6.7
 Williamsport..             16           0             0.0        17             4         23.5
   Total ..........        485           56           11.5%      505            59         11.7%

_______________
a
 Includes ASCRAs.

Source: Pennsylvania Board of Probation and Parole.


staff in all districts ranged from the high of 72 in Philadelphia to a low of 3 in Al-
toona. The next highest turnover after Philadelphia was the Pittsburgh District Of-
fice, with 19 field agent equivalent staff leaving that district. The turnover data for
the rest of the state for this entire period does not include the basis for the turnover.

      In July 2008, the Board expanded the tracking of the six categories of agent
transfers, promotions and terminations to all district offices. Table 13 shows the
actions taken in this regard in all district offices from July 2008 through June 2009.
During this time period, the transfers from Philadelphia to other PBPP offices ac-
counted for six agents, and the field agent equivalent turnover in total for Philadel-
phia represented almost 40 percent of the comparable total statewide turnover dur-
ing that time period.

      One of the effects of high turnover is the reduced experience of field agents in
those district offices with higher turnover. For example, the higher turnover rate in
the Philadelphia District Office has led to a higher concentration of agents with li-
mited experience in that district. According to PBPP human resource records, as of
the end of 2007, 46 percent of the Philadelphia District Office agents had less than


                                                       58
three years experience in the parole agent position. At the same time, 75 percent of
Philadelphia District Office supervisors had less than three years supervisory expe-
rience. 3

                                                        Table 13
                                   Basis for Field Agent Equivalent Turnover*
                                                (July 2008 - June 2009)

                                Other     Other
                                PBPP      State
      District Office           Office   Agencies   Promoted    Retirement   Resignation   Termination     Total

Allentown .............            1        1          1            2            0               0           5
Altoona ................           1        0          0            0            0               0           1
Chester ................           1        0          1            1            1               0           4
Erie ......................        2        0          0            0            0               0           2
Harrisburg............             3        0          2            2            1               0           8
Mercer .................           3        1          0            0            0               0           4
Philadelphia .........             6        3          1            3            7               3          23
Pittsburgh ............            1        1          2            2            0               0           6
Scranton ..............            1        1          0            0            0               0           2
Williamsport .........             0        0          2            2            0               0           4
      Total .................     19        7          9           12            9               3          59
_______________
*Includes ASCRAs.

Source: PA Board of Probation and Parole.


       We looked at the seniority of field agents statewide at the end of 2008. 4 It
ranged from less than two months to almost 31 years statewide. As shown on Table
14, the median years of experience statewide was almost 5.5 years. 5 In the Phila-
delphia District Office, however, the median years of experience was 3.5 years, the
lowest median years among the district offices. This discrepancy in experience be-
comes more apparent when looking at field agents with one year or less experience. 6
Statewide, 17 percent of field agents had one year or less experience. This can be
explained, in part, by the addition of 74 agents to the complement since FY 2006-07.
However, as Table 14 shows, four district offices (Allentown, Chester, Philadelphia,

3Itwas stated frequently during this review that talented field agents are unwilling to seek a promotion to the
supervisor position since it requires increased responsibility with a decrease in benefits. Unlike field agents,
the supervisors are not assigned a Commonwealth vehicle and cannot schedule their own hours.
4We used December 30 to minimize the affect of recent BTA graduates and retirements on the years of expe-

rience.
5Median is the middle or the half way point in a ranking from highest to lowest. The median number has as

many things above it as below it.
6PBPP officials have stated that it takes about one year for a new agent to handle a full caseload.


                                                           59
and Pittsburgh) had a larger than the statewide percentage of field agents with less
than one year experience. Of those, only the Philadelphia District has median years
of experience lower than the statewide average. Additionally, as shown on Table
12, of those districts, only the Philadelphia District had a turnover rate higher than
the statewide average for the last two years. The Philadelphia and Allentown Dis-
trict Offices were the only offices that had units where half or more than half of the
agents in a unit had one year or less experience. Allentown had one such unit and
Philadelphia had three. 7 This lack of experience affects the ability of the agent to
handle a full caseload and for the supervisor to redistribute caseload due to a leave
or vacancy situation.

                                                        Table 14
                                              Seniority of Field Agents
                                                    December 30, 2008

                                             Agent        Average       Median       One Year       % of
       District Office/Region                Counta        Years        Years        or Lessb       Total
    Chester ...........................        36            8.47         6.04           7         19.4%
    Philadelphia ....................         117            5.07         3.42          28         23.9
     Eastern Region .............             153            5.87         4.17          35         22.9%

    Allentown ........................         51            6.83         6.00           15        29.4%
    Harrisburg ......................          53            6.03         4.42            6        11.3
    Scranton .........................         29            5.00         4.42            3        10.3
    Williamsport ....................          16            8.96         9.08            1         6.2
     Central Region ..............            149            6.42         4.83          25         16.8%

    Altoona ...........................        20           10.52       10.88            1          5.0%
    Mercer ............................        40            6.95        6.71            3          7.5
    Erie .................................     23            6.09        4.75            1          4.3
    Pittsburgh .......................         63            7.80        6.75           11         17.5
     Western Region ............              146            7.67         6.75          16         11.0%

       Statewide ....................         448            6.64         5.46          76         17.0%
_______________
a
 Filled field agent positions. These numbers do not include ASCRAs.
b
 Field agents with one year or less time in the position.

Source: Pennsylvania Board of Probation and Parole.




7The Allentown unit and one of the units in Philadelphia with half or more than half of the agents with one year

of experience or less were newly created in late 2007, which may explain, in part, the lack of seniority of the
agents in those units.

                                                           60
Basis for Agent Transfers From the Philadelphia District Office
        We spoke with several of the agents who transferred to other districts from
the Philadelphia District Office since 2007 in an effort to identify what actions could
be taken by the PBPP to retain field agents in this district.8 The agents’ length of
service in the Philadelphia District Office ranged from less than two years (transfer
to an ASCRA position) to over thirteen years, and they served a number of different
offender populations. 9 In virtually all cases, the agents we spoke with viewed the
Philadelphia District Office as an excellent training opportunity due to the number
and variety of situations encountered in a relatively short period of time in that of-
fice. Several of these agents specifically sought employment in the Philadelphia
District Office due to the opportunities associated with working in a large, urban
environment with a large offender population. Others sought the positions in this
office because those positions were vacant, and they wanted to work with the PBPP.
Only one of the agents we interviewed stated that her intent in accepting a field
agent position in Philadelphia was to transfer to another district when the opportu-
nity became available, although this was the most frequently stated scenario anec-
dotally.

        The majority of the agents we spoke with transferred either for family con-
siderations (e.g., a desire to return to a home area to be closer to family) or as a re-
sult of the increased cost-of-living and commuting time. Two, however, specifically
cited difficulties with working in the Philadelphia District Office, and several consi-
dered the caseload and workload to be difficult to adequately maintain. Most of the
agents we spoke with were aware of large caseloads and workloads in the district
office but were not subject to them. The experience of the agents regarding case-
load and workload generally appears to be unit specific, with some units having
fewer vacancies, being better able to shift resources around to respond to vacancies
or having specialized populations that had lower caseloads.

        In general, the agents described similar working conditions in their new dis-
tricts, although certain aspects of the function, e.g., lack of cooperation from other
agencies, tended to be more of an issue in the Philadelphia District. The opinion of
the agents varied regarding other aspects of the job, e.g., type of offender, with sev-
eral saying the offenders were generally the same and others finding one or the oth-
er district’s offenders more difficult to supervise. Several of the agents attributed
any noted differences to the size of the Philadelphia District Office and the city en-
vironment. A few also commented that any plan for daily activities in Philadelphia
would frequently be disrupted by events such as offender arrests. Some viewed this
favorably, since the unpredictability of the job was one of the factors that attracted
them to it. Others noted that this made it difficult to complete work activities.

8We spoke with agents who remained employed with the PBPP in district office positions. This included agents
now employed as supervisors and district directors, although the district directors had transferred from Phila-
delphia prior to 2007.
9Several of the agents had general caseloads, others at some point served as a FAST agent, had CCC caseloads,

worked in a mental health unit, and worked in a sex offender unit.

                                                      61
                  Process for Filling Field Agent Positions

     Field agents are hired through the Civil Service Commission process. Mini-
mum requirements for the position are:

      One year as a parole investigator; OR a bachelor’s degree; OR four
      years of paraprofessional or technical experience in probation, parole,
      corrections, or other criminal justice agency; OR four years of parapro-
      fessional interviewing, counseling, or case management experience in a
      social services or human services setting; OR an equivalent combina-
      tion of experience and training which includes two years of paraprofes-
      sional or technical experience in a probation, parole, corrections, or
      other criminal justice agency, or two years of paraprofessional inter-
      viewing, counseling, or case management experience in a social servic-
      es or human services setting.

      Related education can be substituted for the required experience, ex-
      cept for the two years of paraprofessional or technical experience with
      a criminal justice agency or the two years of specified paraprofessional
      experience in a social or human services setting. Appropriate expe-
      rience can be substituted for the required education. Unrelated expe-
      rience or education will not be accepted.

      When an opening occurs in a district office the following steps are taken:

      •   The open position is posted in the seniority unit for five days. Philadel-
          phia and Chester districts are a joint seniority unit. Every other district
          is its own seniority unit.
      •   If no current district agent applies for the position, the opening is for-
          warded to the regional office.
      •   The open position is posted publicly statewide for 15 days.
      •   After the posting closes, the district director receives an interview packet
          from the PBPP Human Resources Division through the regional office.
      •   The district director interviews the candidates and makes a recommenda-
          tion to hire conditioned on the candidate passing the background, medical,
          and psychological tests.

       Applicants are subject to a criminal history background check. A felony con-
viction results in an automatic disqualification since the applicant would be prohi-
bited from carrying a firearm. A history involving domestic violence, other criminal
conduct, or termination from another state agency for negative reasons would re-
ceive additional review although not an automatic disqualification. Once selected,
the applicant is required to attend the Basic Training Academy (near Harrisburg)
                                           62
for an up to eight-week training course. Applicants are also advised that a new
employee hired or transferred into a parole agent position will be required to re-
main within that seniority unit for a minimum period of two years from the date of
hire or transfer in. Exceptions are made for hardship, Americans with Disabilities
Act (ADA) accommodation, and a promotion outside the seniority unit.

       To ensure that applicants are aware of the specific requirements concerning
the eight-week training and two-year restriction on transfers, the PBPP uses a Con-
dition of Employment Advisory to Prospective Employee that provides this informa-
tion to the prospective employee at the time of the interview. The prospective em-
ployee is required to sign the document indicating his agreement to comply with the
conditions stated in the advisory. The Civil Service posting for the Parole Agent 1
position similarly includes these conditions as job requirements. The posting also
states that the candidate will need to pass a background investigation and undergo
a medical examination, a urinalysis screening, and a psychological examination.
According to Board officials, it takes approximately 170 days from the posting of a
vacancy until the new employee is hired and completes the Basic Training Academy
(BTA).

Review of Position Description and Minimum Standards

       Due to concerns that the position description for parole agent may not ade-
quately reflect the appropriate skills needed for the position, the PBPP has recently
contracted with a consultant for the development of updated medical and psycholog-
ical guidelines; identification of knowledge, skills, and abilities required for effective
job performance; and the generation of the minimum qualifications needed for the
parole agent job. For example, according to Board officials, because the current job
description allows for “social services and human services” experience, a number of
the applicants for the positions have not actually been qualified for the enforcement
aspects of the position. The PBPP is seeking a more accurate job description and
minimum qualification standard that will result in less attrition by attracting and
hiring individuals more suited for the work of an agent.

       Based on the consultant’s recommendations, the PBPP plans to revise the
Minimum Education and Training (MET) requirements and update the standards
for the medical and psychological evaluations. In addition, the PBPP is meeting
with field staff to develop agent interview questions based on job requirements. A
draft of the consultant’s report was provided to the Board in August 2009, and the
final report is expected in the fall.

      The Center for Effective Public Policy also has recognized that a new field
agent skill set is needed for the evidence-based re-entry approach to supervision.
The Center is in the process of developing a new Integrated Case Management
Handbook for the NIC, and this has been identified as an area for review. Although
they have not specifically identified the new skill set needed by field agents, inter-

                                           63
personal skills have become more important for an agent in order to, for example,
develop relationships with providers and offenders’ families. This change may af-
fect the requirements for the job.

Results of Recent Hiring Processes

        Between September 2005 and April 2009, the PBPP hired 234 individuals to
fill parole agent (including ASCRA) positions. The district offices recommended 323
(23.2 percent) of the applicants interviewed for the positions available. 10 As Table
15 shows, the Philadelphia District Office interviews the highest percentage of per-
sons received from the requested hiring lists. According to Board officials, this is
due to the district having a larger number of vacancies to fill than other districts.

       One of the factors affecting the number and percentage of applicants hired is
whether, once recommended, the applicants are able to pass the background, psy-
chological, and physical checks as well as their continued interest in the position.
Since applicants often are applying for more than one Civil Service position, they
may not be available for the field agent position once the hiring process concludes.
These factors have been cited as some of the reasons the Philadelphia District Office
has difficulty filling its vacant positions.

       As Table 15 shows, the Philadelphia District Office has a higher number of
applicants who fail the background, psychological, and physical checks as well as a
higher number of applicants who withdraw from consideration. The percentage of
applicants affected, however, is highest in the Philadelphia District Office only for
failed background checks, 7.6 percent of the recommended applicants, compared to
rates ranging from 3 percent to 4.8 percent for those offices that had applicants who
failed this check. In the other categories, the Philadelphia District Office has a
comparable, or lower, percentage of applicants affected. For example, 11.4 percent
of the recommended applicants withdrew in Philadelphia compared to 28.6 percent
in the Chester District Office. During this same time period, one applicant failed
the BTA and returned to his prior state agency position.

       Between February 2009 and June 2009, 36 field agents have been hired and
started their employment with the district offices. 11 This represented a net increase
of 25 agents, with the others filling more recent vacancies. An initial group of 20
agents completed BTA at the end of May 2009. Of the remaining 16 new agents,
15 graduated from BTA in early September 2009, and one is expected to complete
the course in October. 12 A majority of the new agents, 20, are assigned to the

10Since additional positions may have become vacant after the initial posting of the position, it was not possible
to identify how many vacant positions were available for the applicants. According to a PBPP official, however,
experience has shown that only the Philadelphia District Office has difficulties hiring for all vacancies.
11Depending on the BTA schedule, a field agent may be hired and assigned to a district office prior to completing

BTA training. While waiting to attend training, the field agents assist with paperwork and learn office proce-
dures as well as shadow current field agents to learn the job.
12This agent missed a couple of days of defensive tactics training which is being rescheduled for her in October.


                                                       64
                                                                                      Table 15
                                                                        Results of Hiring Process*
                                                                        (September 2005 – April 2009)

                                                          % of             #                               Failed                                                         % of
                              Names           #          Names           Recom-              % of           Back-      Failed      Failed                                Recom-
     District Office         Provided   Interviewed     Provided         mended          Interviewed       ground      Psych.      Phys.      Withdrew a     Hired       mended
     Allentown .....          334           208             62.3%            56              26.9%             2          0           4             1           47          83.9%
     Altoona ..........       171           110              64.3            11              10.0              0          0           0             1            8          72.7
     Chester ........         146           103              70.5            21              20.4              1          1           1             6           11          52.4
     Erie ................    149            65              43.6             8              12.3              0          0           0             1            7          87.5
     Harrisburg .....         260           120              46.2            20              16.7              0          1           1             1           16          80.0
     Mercer ...........       242           124              51.2            18              14.5              0          0           0             1           13          72.2
     Philadelphia ..          534           390              73.0          132               33.8             10          3           3           15            88          66.7
65




     Pittsburgh ......        358           135              37.7            33              24.4              1          0           0             4           26          78.8
     Scranton .......         214           117              54.7            19              16.2              0          0           0             2           13          68.4
     Williamsport ..           87            23              26.4             5              21.7              0          0           0             0            5        100.0




     _______________
     *The district office data includes the data for the sub-offices in that district.
     a
       Includes applicants who withdrew during the hiring process, declined a job offer, accepted a different position, accepted a position in another PBPP district office or were
     no longer interested in the position.

     Source: Pennsylvania Board of Probation and Parole.
Philadelphia District Office, 6 are assigned to the Harrisburg District Office, 4 are
assigned to the Chester District Office, 3 are assigned to the Williamsport District
Office, 2 are assigned to the Allentown District Office, and one is assigned to the
Pittsburgh District Office. With these additional field agents, the statewide vacan-
cy rate at the end of June 2009 was 2.9 percent (13 vacancies).

Training Standards for Field Agents

        As noted above, an up to eight-week training period at the BTA is required
as a condition of employment. Included in this training are 68 hours of firearms
training and 37 hours of agent safety skills. A minimum of 40 hours of training
each year is also required. Twenty hours are devoted to annual firearms requalifi-
cation and 8 hours to agent safety skills. The remaining hours are used for updat-
ing case management, assessment, or training pertaining to evidence-based super-
vision strategies.

       The PBPP implemented several changes to the BTA training in late fall 2008
through early winter 2009. These included the addition of motivational interview-
ing training to provide agents with the skills to facilitate behavioral change with
their offenders; weekly and final exams to test on the subject matter covered; the
addition of an elementary performance test regarding the practical skills of the de-
fensive tactics techniques taught during that course; a combined case planning and
LSI-R training; and gangs overview along with security threat groups training.

       The annual training for field agents is offered by the PBPP throughout the
state and covers a variety of topics including computer skills, report writing, first
aid, and evidence-based practice in community corrections. PBPP trainers, staff,
and others provide the training. Last year, the PBPP purchased the “Laser Shot”
System to provide use-of-force training through computerized scenarios that allow
the agent to interact with the events occurring on screen. 13 In addition to being in-
corporated into the BTA, the system has been used in all ten district offices for an-
nual training. The plan is to use it in the districts twice a year to provide rein-
forcement of use-of-force training as well as assisting with firearm proficiency. The
PBPP is also using an air soft system that also simulates shooting and can be used
with “simunition” (similar to paint balls) to indicate to the trainee where on the “of-
fender” his shots hit (this can be used with more than one person participating). Al-
though this has been used in some of the districts over the past two years, space re-
quirements limit its use in all districts at this time. 14



13The PBPP recently approved a proposal to implement the use of Electro Muscular Disruption Technology (tas-
ers).
14Due to the nature of the soft projectiles as well as the “simunition,” the location for this training must be one

that accepts damage from the projectiles and is well-removed from the public to avoid unintended injuries. In
Mercer County, a house has been donated for law enforcement training, and has been used for this training.

                                                        66
       PBPP is working with the National Curriculum Training Institute (NCTI) to
train ASCRAs and field agents to act as co-facilitators of cognitive groups to provide
services to offenders. According to a PBPP official, the majority of vendors have ex-
perience with mental health or dependency, but not with life skills or violent beha-
vior prevention. This has allowed PBPP to provide programming to the offenders
through its own staff. The ASCRAs and the TCs frequently serve as the group
counseling providers.

        PBPP Efforts to Address Field Agent Turnover and Its Effects

       The PBPP has implemented several changes recently in an effort to address
the turnover and subsequent staffing problems in the Philadelphia District Office.
This has included restricting field agent transfers, using annuitants to reduce the
burden on field agents by performing administrative functions, and creating a re-
tention incentive pay increase for the Philadelphia District Office. The PBPP is also
planning to establish wage field agent positions in the Philadelphia District Office
to quickly fill vacant and agent-on-leave positions in an effort to reduce the length
of time a position remains vacant and maintain caseload and workload levels within
their guidelines.

Applicability of the Two-Year Minimum Policy

       As noted above, new employees hired or transferred into field agent positions
(the policy is not applicable to ASCRAs) are required to remain in the hiring unit for
a minimum of two years from the date of hire. The Board implemented this policy
prior to 2002 in an attempt by the PBPP to address the turnover rate in the Phila-
delphia District Office. 15 Requests to transfer to other locations are considered af-
ter the two-year period. As noted above, Philadelphia frequently has vacancies due
to turnover, so there was a concern that applicants would seek a position in Phila-
delphia and then transfer to their chosen district when a position became available
in that other district. This policy provides some control over those transfers.

        The Board makes exceptions to this two-year minimum policy in cases of
hardship, ADA accommodation, and promotion outside the hiring unit. These re-
quests are processed through Human Resources to confirm that they meet the re-
quisite standards. Between May 2004 and February 2009, the Board received 26
hardship requests for transfer from one Board office (district, central, SCI) to
another. Fourteen of the requests were from agents in the Philadelphia District
Office seeking transfers to other districts. The other requests were from eleven dif-
ferent locations with only one district, Erie, accounting for two requests. The major-
ity of the requests for transfers from Philadelphia, 64 percent, were denied. Of the


15The PBPP could not identify the actual date the policy was adopted. The earliest date of a signed advisory
from an applicant the PBPP could provide is dated November 16, 2001.

                                                      67
remaining requests from other districts, five were denied, six were approved, and
one was withdrawn. Two of the approved requests were for transfers to Philadel-
phia.

        The Department of Corrections (DOC) has a similar restriction on transfers
for its corrections officers. Corrections officers are required to remain at their State
Correctional Institution (SCI) for two years from their date of hire. Transfer re-
quest lists are developed and the approval for a transfer is based on seniority with
SCIs required to allow a certain number of transfers from the SCI each quarter. 16
According to a DOC official, the policy was adopted to regulate the impact of the
movement of employees from one work location to another.

Use of Annuitants
       The PBPP has also used annuitants (retired PBPP field agents) to address
staffing needs and assist with certain non-field workload activities to reduce the
workload for the active agents and allow them more time to supervise their caseload
in the field. For example, the annuitants may perform various field agent func-
tions, such as:

        •   receive new offenders, review their conditions of parole, and/or conduct a
            risk assessment;
        •   provide urinalysis testing;
        •   take offender photographs, as required of all new offenders;
        •   assist with tracking unconvicted parole violator cases to determine the
            disposition of new charges and take proper action on parole violations;
        •   assist agents with closing cases by reviewing the tracking of maximum
            sentence dates and ensuring public safety by checking to see if an offender
            is in good standing and can be discharged from parole;
        •   review case files to ensure compliance with Quarterly Case Reviews and
            Unit Audits and ensure deficiencies have been addressed;
        •   monitor “Alpha” lists – the information on each offender that must be ve-
            rified and updated at the end of the month, e.g., address, supervision le-
            vels, and supervision fee payments;
        •   assist with special reviews of cases;
        •   assist with preparing cases for hearings; and
        •   mentor new agents.

       According to the PBPP, the use of the annuitants to perform field agent office
duties allows the field agent to spend more time in the field making contacts or

16One per quarter for each SCI with the Graterford and Camp Hill SCIs required to allow an additional transfer

every six months for an annual total of six employees.

                                                     68
conducting investigations. Annuitants are restricted to working no more than 95
days in a calendar year.

      During FY 2008-09, as many as 15 annuitants were used to supplement field
agents. This included seven in the Philadelphia District who moved between units
based on need, two in the Allentown District, three in the Scranton District, and one
each in the Pittsburgh, Mercer, and Chester Districts.

Board-initiated Retention Incentive for Philadelphia Field Agents

       In an effort to reduce the turnover in Philadelphia, the PBPP has recently
implemented a retention incentive program for the Philadelphia District that pro-
vides approximately the equivalent of a two-step increment in pay after two years of
continuous service in the district. Another two-step increment is available after
four years of continuous service in the district. These increases are in addition to
all other longevity increases and general pay increases. This is also applicable to
incumbent agents in the district office. If an agent transfers out of the Philadelphia
District, any increments awarded under the incentive program will be taken away.
An experienced agent voluntarily transferring into the Philadelphia District will
immediately receive the applicable increments depending on his or her years of ex-
perience.

        In a letter to the Office of Administration (OA) seeking the incentive for the
Philadelphia District Office, the Chairman of the PBPP noted that the adjusted cost
of living in Philadelphia is 11 percent higher than Harrisburg or Pittsburgh and
their surrounding areas. In addition, the agents who work in Philadelphia are sub-
ject to the city’s 4 percent wage tax. The effect of the increased cost of living
coupled with the wage tax significantly reduces the “take home” pay received by
those agents in comparison to other areas of the state.

       The PBPP estimated the cost of the program to be $845,000 in FY 2009-10,
with associated savings in terms of reduced overtime and recruiting, hiring, and
training of new staff in the amount of $166,000, resulting in a net cost of $679,000
in FY 2009-10. This program was approved for implementation by OA and the un-
ion representing the field agents and went into effect in September 2008. Recruit-
ment and training costs are reported by the PBPP to be $5,508 per agent, including
human resources staffing expenses and BTA costs.

       The DOC similarly has a pay differential for the corrections officer position
in the Chester (Chester County) and Graterford (Montgomery County) SCIs. Ac-
cording to a DOC official, it was established to offset the higher cost of living in
Southeastern Pennsylvania in order to retain employees. Although the DOC does
not have statistics on whether this pay differential has had the intended effect of



                                          69
retaining employees, according to a DOC official, the DOC has not had difficulty fill-
ing positions in these SCIs.

       OA has also approved other exceptions to the Commonwealth classification
and compensation schedule to address specific agency staffing needs. These in-
clude:

      •   PennDOT had difficulty attracting CDL applicants in about half the coun-
          ties for temporary and full-time equipment operator positions. OA autho-
          rized the agency to appoint above the minimum pay grade.
      •   The Departments of Public Welfare, Corrections, and Military and Veter-
          ans Affairs had difficulty attracting Registered and Licensed Practical
          Nurses. OA authorized the agencies to appoint above the minimum pay
          grade.
      •   The Department of Labor and Industry had difficulty attracting and re-
          taining vocational rehabilitation counselor interns due to the stringent
          educational requirements and the higher compensation in New Jersey.
          OA authorized the agency to advance the employee in range to compete
          with the higher salary in New Jersey.

        Since the program has only been in place for approximately one year, it is too
soon to tell whether it is having the intended effect in reducing turnover. Anecdo-
tally, however, the officials in the Philadelphia District Office noted that fewer
agents are asking when they can transfer out of that district. In our interviews
with agents who transferred from Philadelphia, those who transferred from Phila-
delphia due to family considerations, length of commute, and the high cost of living
in the area stated that they would not have been influenced to stay by the retention
incentive. Of those agents who transferred for other reasons, one of the agents
stated he would have remained in Philadelphia but for the difficulties he was expe-
riencing in his employment unit, and another left after she became eligible for the
incentive since her original intent was to transfer to a specific district when an
opening became available. Both indicated that the incentive would not have had, or
had, any influence on their decision.

        Since many of the transferring agents have family connections in the area of
the state to which they transfer, the decision to transfer for these agents may not be
influenced by the increase in salary. Most of the agents we spoke with, however,
expressed their opinion that the incentive may help retain agents who live in the
city and new agents who move to the area since it will help to offset the higher cost
of living and the city wage tax.




                                         70
Use of Wage Positions

       To further address the ongoing vacancy issue in Philadelphia, the Board has
established a cadre of ten wage positions for Philadelphia. As regular complement
positions become vacant, they can be filled immediately by drawing on the filled
wage positions. The PBPP has estimated that it takes approximately one year
before a new agent can manage a full caseload, including four months for the hiring
process, eight weeks for the basic training, and six months of on-the-job-training.
During this time the current staff picks up the additional cases until the new em-
ployee is ready to handle a full caseload. The goal in using wage positions is to have
agents ready to pick up a full caseload when a transfer or other vacancy (e.g., mili-
tary leave) occurs. Depending on the FY 2009-10 budget, the PBPP plans to begin
hiring for the wage positions with the intent to use them to fill vacancies as they
arise.

       The wage employees will attend the BTA and be able to function as full-time
field agents. Until a vacancy becomes available, the wage position will assist other
agents or take responsibility for the caseload and workload of a field agent on ex-
tended leave. It is anticipated that the use of the wage positions will reduce the
need for annuitants, discussed above, in the Philadelphia office.




                                         71
F. The Board Has Identified Technological Needs for the Parole
Supervision Function


      The PBPP has recognized that automation and the implementation of new
technologies may enhance the Board’s capacity to track and supervise offenders.
The Board has been pursuing improved automation and technological enhance-
ments statewide in a number of areas including: (1) the development, in coopera-
tion with the Pennsylvania Department of Corrections (DOC), of an Integrated Of-
fender Case Management System; (2) the purchase of electronic devices to automate
parole agent data collection and reporting; and (3) the expanded use of electronic
monitoring (focusing on GPS—global positioning satellite systems) to monitor the
movement and behavior of certain types of offenders.

       Two of the initiatives—IOCMS and the purchase of electronic devices for
agent data collection and reporting—are currently in progress, but implementation
target dates are either indefinite or several years away. Also, for the near future,
the expanded use of electronic monitoring through the purchase and application of
GPS is not being pursued, due largely to the unfavorable outcomes of two pilot stu-
dies.

Integrated Offender Case Management System (IOCMS)
        The PBPP and the DOC are in the process of acquiring an integrated offender
case management system to replace or partially replace the agencies’ current of-
fender management systems. When fully implemented, the IOCMS will ensure that
all offender-based information from both agencies is available for management of
offenders in one integrated system.

        Timely access to accurate, complete, and current offender information is criti-
cal to the Commonwealth’s ability to fulfill its public safety responsibilities. DOC
and PBPP have historically maintained separate systems with limited capability for
electronic exchange of information. Since the two organizations have a similar
workflow and a common client population, a single integrated system is expected to
eliminate inefficiencies/duplication and promote data integrity as offenders move
across the two entities.

       In 2004, the PBPP proposed a project to develop a Pennsylvania offender
management system to fully integrate offender case management across business
functions critical to the agency. That initial project vision was subsequently ex-
panded to develop an integrated offender case management system in partnership
with the DOC. Beginning July 1, 2006, the PBPP’s Bureau of Information Technol-
ogy was merged with the DOC. Since that time, all ongoing operational hardware
and software replacement and maintenance have become the responsibility of the
DOC.


                                          72
 
       According to PBPP officials, an integrated offender case management system
will benefit the agency by providing a platform for greater access to information (in-
cluding a mobile solution for parole agents in the field) and increased efficiency for
all PBPP business processes, from initial incarceration, to parole decision-making,
to parole supervision. Specifically for the Board, the IOCMS will:

      •   provide an integrated flow of offender data with the DOC and across key
          parole decision, supervision, and revocation/recommit business processes;
      •   provide an effective information interface with business partners;
      •   provide real-time point-of-contact data for field supervision;
      •   improve data quality and access for analysis and research; and
      •   enable implementation of outcome-based performance processes.

       The project has moved through two stages involving the services of two ven-
dors under contract. The first contract was issued in October 2005 with Deloitte to
develop the high level business requirements necessary to support a newly designed
workflow process. The requirements designed by Deloitte represent an outline for
the second stage, during which the offender case management system will be de-
signed, developed, and implemented. This subsequent stage also necessitated ob-
taining vendor services, but DOC and PBPP were unable to obtain funding approval
for a Request for Proposal (RFP) until 2008.

       When funding was approved, a second contract was issued in April 2008 with
TaTa Consulting Services (TCS). This is a five-year contract (with an option for five
one-year renewals). Its purpose is to finalize the design of the IOCMS and to im-
plement it. This process involves documenting the detailed business and technical
requirements, developing the application functions, developing training materials,
conducting system testing, providing end-user help desk support, defining security
roles and access rights, and providing for ongoing IOCMS maintenance, updates,
and enhancements. The cost of the TCS contract is $19,621,293, plus costs for
maintenance and enhancements. The IOCMS project is operationally managed by
an executive steering committee (which meets quarterly, or as required), an opera-
tional steering committee (meeting bi-weekly, or as required), and a project team
leadership committee (which meets weekly). Each of these committees includes
representatives from the Office of Administration (Office of Information Technolo-
gy), the DOC, and the PBPP.

      The overall project time frame anticipates final implementation by 2013. Ac-
cording to PBPP, IOCMS is on schedule with the project plan, and PBPP has been
involved in providing subject matter experts in various areas that are under devel-
opment including, for example, development of a comprehensive, integrated busi-
ness workflow model that reflects the business processes performed by both DOC
and PBPP.

                                          73
 
Automating Parole Agent Record-Keeping

       One aspect of the IOCMS project may directly impact certain operational
activities of parole agents. Currently, PBPP’s parole agents track over 31,000
offenders statewide through the use of paper forms and hard files, and agents
record all contacts with parolees and probationers on the Record of Interview Form
PBPP-259 by hand. This includes primary contacts (directly with the subject) and
collateral contacts (with employer, family members, and associates). The results of
this manual recordkeeping process are not linked with offender history in a central
database.

       Beginning in January 2009, senior agency officials met to examine options
and decide on a strategy to automate this recordkeeping function. Among several
available technological products (e.g., laptop computers, pen tablets, and the Black-
berry®), the committee decided to further investigate the application of the Black-
berry®, which appeared to be the most cost-effective solution with a number of de-
sirable features.

       A Blackberry® is a mobile communications device designed to fit into a pocket
or clip into a belt holster. The benefits of the Blackberry® for the PBPP included:

      •   device is small and portable;
      •   low cost;
      •   easily replaced;
      •   updates are automatic;
      •   transmissions are encrypted;
      •   device security—passwords and JNET certificate;
      •   can remotely erase the unit if lost; and
      •   gives parole agents access to other resources, e.g., email and JNET.

The goal of this project, according to information provided by the Board, is to:

      •   create a backend database that will hold the basic data collected on the
          Record of Interview Form PBPP-259;
      •   create an electronic version of the Record of Interview Form PBPP-259 for
          use on a Blackberry® that will be used for data entry and that will up-load
          said data to the backend database in real time; and
      •   create basic reports and notes to review the information on a given case or
          by a given agent, including metrics needed by supervisory personnel.



                                          74
 
      In effect, parole agents will fill out an electronic Form PBPP-259 using a
Blackberry®. The data will then be transferred to a central database for storage.
Data transfer can be done immediately if network connectivity is available or up-
loaded in a batch once the parole agent has network access.

       The Blackberry® has been tested in two pilot projects. The first pilot study
was conducted from March 8, 2009, to May 15, 2009, and included 25 agent partici-
pants from five units in northwest Philadelphia and one unit in Harrisburg. The
second pilot ran from July 20, 2009, through August 14, 2009, and included 50
agents, selected from each of the ten PBPP districts. The results of these evalua-
tions have not been positive, as many of the agents in the first group ultimately dis-
continued their participation because of operational difficulties. In the second pilot,
additional operational difficulties were encountered that have called into question
the device’s ability to increase agent efficiency in recording and reporting their in-
terview contacts.

       Due to the anticipated launch of the IOCMS and its integration into PBPP
operations by 2012-13, any electronic device that is adopted for parole agent record-
keeping use would necessarily be acquired on a short-term, interim (i.e., three-
fiscal-year) basis, and would need to be quick and inexpensive to implement. Com-
monwealth officials have estimated that the Blackberry® Project (including the pilot
study) will result in a net amount of “new” funding required, through FY 2011-12, of
approximately $850,000. The PBPP anticipates that any electronic device acquired,
if funding becomes available, and used by PBPP parole agents during this period
will eventually be replaced when the IOCMS is fully implemented.

       As of September 2009, PBPP officials report that, due to a combination of re-
duced productivity of agents, coupled with connectivity and operability problems of
the Blackberry® and its required portable keyboard, the Blackberry® was found to
be unsuitable, and research on it was suspended. However, the PBPP reports it is
now planning a pilot study using 30 of the new “netbook” class of portable comput-
ers to determine their effectiveness to improve agent productivity and provide real
time data.

Global Positioning Systems (GPS)

       Since 2005, the PBPP has considered the use of global positioning satellite
systems (GPS) to track offenders. If implemented, this technology would be an
expansion of an existing electronic monitoring program that uses radio frequency
electronic monitoring equipment as a sanction for some offenders who violate condi-
tions of their parole. The Board’s policy is to employ electronic monitoring, as
appropriate, to comply with Board/court imposed conditions or as an adjunct to of-
fender supervision. The PBPP owns between 700 and 800 electronic monitoring
equipment units, about 300 of which are in use around the state at any given time.

                                          75
 
       GPS Defined. GPS refers to global positioning technology or a system which
uses orbiting satellites to pinpoint the physical location of objects or persons. These
satellites continuously transmit position and time signals that can be detected by a
GPS receiver. In the corrections field, GPS technology can enable probation and pa-
role officers to track offenders in their communities. GPS is one form of electronic
monitoring that can be used in the supervision of offenders (other forms of electron-
ic monitoring include radio frequency devices, pagers, and voice verification sys-
tems).

       Offender tracking using GPS can be either active or passive. Under active
GPS tracking, a probation or parole officer may locate an offender’s current location
at any time of day as well as review their location for any previous time frame.
Passive GPS, on the other hand, records and stores the offender’s movements, but
does not immediately transmit the data. At a later time the offender, usually wear-
ing an ankle bracelet and miniature tracking device, connects the tracking device to
a base station from which the recorded log data can be reviewed by the supervising
officer after the fact. A hybrid tracking system can also be used that generally op-
erates in passive mode unless the offender enters a restricted area, which automati-
cally triggers a conversion to active mode and typically an alert signal.

       Statutory Provision. Act 2006-178 amended the Crimes Code to enact cer-
tain provisions of Megan’s Law and Jessica’s Law. Pursuant to this act, the PBPP
was given authority to impose upon sex offenders supervision conditions that in-
clude offender tracking through global positioning system technology. Further, Act
178 directed the Board to apply for federal funding as provided in the Adam Walsh
Child Protection and Safety Act of 2006 to support and enhance programming using
satellite global positioning system technology.

        Pilot Studies. In 2005, the PBPP undertook research into the tracking of of-
fenders using GPS technology. An eight-month pilot study was conducted from Oc-
tober 2005 through June 2006. Five companies made their equipment available for
the study; each company agreed to provide 10 units for a period of 60 days. Re-
search on these GPS units was conducted with the participation of offenders resid-
ing at five community corrections centers located in Harrisburg, Erie, Allentown,
York, and Philadelphia. Offenders in the pilot were either on pre-release status or
on parole status. An analysis of each company’s GPS system was conducted by the
research team, together with an analysis of questionnaire responses from on-site
staff involved in the project. Also, at the conclusion of each pilot, a brief interview
was conducted with the offenders who were selected as participants.

       Each of the systems was shown to have strengths and weaknesses. For ex-
ample, the units all demonstrated the ability to locate an offender in real time and
to provide a notification of violations within a reasonable time frame (some of the
systems also provided the ability to communicate directly with the offender).

                                          76
 
However, the research also demonstrated the technology to have a number of prob-
lems. All of the GPS units or systems:

      •   required significant human resources to set up, maintain, and monitor;
      •   generated a significant amount of nuisance alerts (that required signifi-
          cant staff time to assess);
      •   relied heavily on the offender recharging the battery; and
      •   failed to maintain a GPS signal when the unit was inside certain types of
          structures.

       A second pilot study, conducted between July 28, 2008, and October 1, 2008,
was, in contrast with the first pilot, more directed to determining GPS applicability
in PBPP monitoring of sex offenders. This pilot program consisted of eight vendors,
each issuing 10 GPS units (total of 80 GPS units tested). Testing was conducted in
two phases (four different vendors for each phase) in the same four district offices
for each phase (Philadelphia, Allentown, Harrisburg, and Altoona).

      The pilot concluded that GPS cannot be viewed as a panacea for the safe
management of sex offenders, but it does have the potential to be used in conjunc-
tion with other sex offender containment strategies. Other results of the pilot study
included the following:

      1. No vendor was able to complete their phase without some technology mal-
         functions such as:

          •   offender equipment would suddenly cease to function;
          •   home bases would interfere with the individual’s home landline;
          •   batteries would die on handheld transmitters; and
          •   at times, offenders were not being tracked for days. This was due to
              the equipment not being able to establish cellular or satellite reception.

      2. The functioning of GPS technology in rural and urban areas continues to
         experience problems due to poor cellular and satellite coverage (offender
         tracking in such areas may be mediocre at best due to the location limita-
         tions).
      3. The drift, which is a temporary inability to provide a location fix, which
         had been experienced during the earlier pilot, continues to be of concern.
         Some areas experienced less drift than others.
      4. The manpower intensity level of PBPP’s agents would increase if active
         GPS tracking is adopted, in comparison to passive and hybrid tracking.
         However, all forms of this supervision tool will increase the workload

                                           77
 
                      substantially for all agents involved, because agents will be required to
                      immediately respond to alert notifications received by active systems.
                      Passive and hybrid systems will require agents to review positional data
                      on a daily basis for all supervised offenders and take appropriate sanc-
                      tioning actions.

              5. The equipment, software, and tracking ability on the whole have improved
                 from the previous pilot findings. The tracking has become more precise,
                 the equipment less bulky, and the software more user friendly.
              6. GPS tracking can be a time saving device for agents in some aspects of
                 their work but this savings will be consumed by the time required to re-
                 spond to alerts, as referenced in point 4 above, and result in a net increase
                 of the time required for supervision.

        The second pilot study also concluded that GPS technology, despite improve-
ments, was still limited in functionality and reliability for PBPP use, and the PBPP
researchers urged caution in pursuing the purchase of this technology until it is
more reliable and can be used with confidence. The PBPP further indicated in this
pilot report its intent to regularly review GPS technology’s progress to determine
whether at a future time, the complex geographical challenges of the Common-
wealth could be addressed. The report noted that, if GPS tracking were considered
in the future, its use should include sexually violent predators, special populations
(e.g., sex offenders, domestic violence, and high profile cases), as well as its use as a
sanctioning/violation tool. In the meantime, the Board would continue to use its ex-
isting electronic monitoring system (radio frequency).
       Recent Reports on Use of GPS for Tracking Sex Offenders. A report on
GPS monitoring of sex offenders issued by the International Association of Chiefs of
Police (IACP) summarizes the advantages and disadvantages of this technology. 1
The reported advantages of GPS include flexibility (the ability of GPS to be pro-
grammed to a particular offender’s risk level and offense patterns); its ability to en-
sure successful reintegration by promoting compliance with conditions and treat-
ment; and its ability to exert a greater degree of control over an offender’s move-
ments, thereby increasing accountability. Further, GPS monitoring increases the
ability to investigate current and past behavior to understand where an offender is
in the relapse/offense cycle.

       Disadvantages, according to the report, vary depending upon the nature of
GPS used (i.e., whether active or passive). High daily cost, labor intensiveness, and
the requirement for immediate agency response are potential disadvantages under
the active system. The disadvantages under the passive system include “after-the-
fact” tracking of data and the lack of immediate notification of zone violation.

                                                            
1Tracking Sex Offenders with Electronic Monitoring: Implications and Practical Uses for Law Enforcement, In-
ternational Association of Chiefs of Police, August 2008.

                                                               78
 
              Other IACP concerns about GPS monitoring of sex offenders included:

       •      False Sense of Security. Because GPS monitoring is a relatively new technol-
              ogy, the public may not realize that offenders can tamper with ankle devices
              or render them inoperable. Additionally, there may be a time lapse between
              an alert notification and agency response.
       •      Legal Concerns. Since the use of GPS technology for offender monitoring is
              somewhat new, agencies using GPS technology should be prepared to encoun-
              ter legal challenges to situations not having significant court precedent. How
              might the court respond, for example, in cases: (1) involving a failure to re-
              spond to an alert that results in a new crime; and (2) involving a new crime
              committed when the radio signal is lost or during equipment malfunction.
       •      Increased Workload. Especially in states where lifetime GPS monitoring is
              required for certain convicted sex offenders, the workload of law enforcement
              and community corrections officers can be anticipated to increase (due, e.g.,
              to monitoring GPS equipment, responding to alerts, reviewing GPS data, and
              maintaining and inventorying the equipment).

      The Department of the Auditor General (DAG) has recommended the use of
GPS technology by the PBPP to track sex offenders. 2 The DAG recommended that
the Pennsylvania State Police and the PBPP request the General Assembly to
amend Megan’s Law to require five years of GPS monitoring for: (1) offenders who
are caught after failing to register their whereabouts as required; and (2) for all
sexually violent offenders whose victims are children.

       The Auditor General’s recommendation was based on findings that 14 Penn-
sylvania counties were using GPS to monitor sex offenders and officials from 11 of
the 14 said that their programs were effective supervision tools. The DAG report
also determined that GPS is the highest level of supervision available short of in-
carceration; that GPS data can be used in court; that GPS can serve as a deterrent
to criminal behavior; and that GPS is cost-effective (because offenders are paying a
daily fee for their tracking as opposed to taxpayers paying costly fees for incarcera-
tion).

       GPS in Other States. According to the National Conference of State Legisla-
tures (NCSL), at least 39 states have enacted GPS or electronic monitoring provi-
sions specific to sex offenders. States vary in their statutory requirements. Some
states mandate GPS tracking of certain sex offenders, while others allow law en-
forcement officials to use their own discretion. See Exhibit 11.



                                                            
2See Using GPS Technology to Track Sex Offenders: Should Pennsylvania Do More?, PA Department of the Au-
ditor General, July 2008; and an observation in Supervision and Monitoring of Parolees, PA Department of the
Auditor General, June 2009, p.31.

                                                               79
 
                                                       Exhibit 11

                       Statutory Provisions of Selected States on the Use of
               Electronic Monitoring (e.g., GPS Technology) to Monitor Sex Offenders

                       Requirement or
       State            Authorization                                        Description
Georgia             A sexually dangerous       A person who is classified as a sexually dangerous predatora is re-
                    predator is required to    quired to wear an electronic monitoring system for the remainder of
                    wear an electronic         his or her natural life. The sexually dangerous predator is further
                    monitoring (EM) de-        required to pay the cost of the system to the Department of Correc-
                    vice for life.             tions if that predator is on probation, to the Board of Pardons and
                                               Paroles if the predator is on parole, or to the sheriff after the predator
                                               completes his or her term of probation or parole or if the predator has
                                               moved to Georgia from another state, territory, or country.
Maryland            GPS technology may         Specific conditions of extended supervision must begin upon the re-
                    be included in sex of-     lease of a sexual offender from incarceration or imposition of proba-
                    fender’s conditions of     tion. One condition of release may include monitoring the sex of-
                    release.                   fender through global positioning satellite tracking technology.
Michigan            Lifetime EM required       A crime of criminal sexual conduct committed by an individual 17
                    for those convicted of     years old or older against an individual less than 13 years of age car-
                    criminal sexual con-       ries a mandated sentence of lifetime electronic monitoring. If this
                    duct with a child under    individual commits certain actsb the individual is subject to imprison-
                    age 13.                    ment up to 2 years and/or a fine up to $2,000.
New Jersey          Continuous EM for          The State Parole Board is required to establish a program for the
                    high risk sex offend-      continuous, satellite-based monitoring of sex offenders in the state
                    ers.                       whose risk of re-offense is high; or a person who has been deemed
                                               appropriate for continuous satellite-based monitoring and who is
                                               classified as a sexually violent predator and who has been sentenced
                                               to community or parole supervision for or has been convicted of or
                                               adjudicated delinquent for a sex offense and the victim of the offense
                                               was under 18 or over 60 years old.c
New York            EM allowed on certain      A court may impose as a condition of probation electronic monitoring
                    sex offenders as con-      when, in the court’s discretion, such will advance public safety, pro-
                    dition of release.         bationer control, or probationer surveillance. As such, the Division of
                                               Probation and Correctional Alternatives is required to develop uni-
                                               form procedures for the use of electronic monitoring.
Ohio                GPS may be autho-          If an offender convicted of any sexually oriented offense is consi-
                    rized for certain sex      dered to be a Tier III sex offender/child-victim offender and the of-
                    offenders.                 fender does not serve a prison or jail term, the court may require that
                                               the offender be monitored by means of a global positioning device. If
                                               the court orders such monitoring, the costs are to be borne by the
                                               offender. However, if the offender is indigent, the cost shall be paid
                                               by the crime victims reparations fund.
West Virginia       EM required for sex-       A sexually violent predator that is on probation, parole, or supervised
                    ually violent predators.   release is required to be subject to electronic monitoring and to pay
                                               the cost of such monitoring unless the court declares person is una-
                                               ble to pay.d


                         Footnotes for this Exhibit are on the following page.
                                                           80
    
                                      Footnotes for Exhibit 11
a
  Georgia statute requires any sexually dangerous predator to wear an electronic monitoring system that
has at a minimum the following capabilities:
      • to locate and record the location of a sexually dangerous predator by a link to a global positioning
          satellite system;
      • to timely report or record a sexually dangerous predator’s presence near or within a crime scene
          or in a prohibited area or the sexually dangerous predator’s departure from specific geographic
          locations; and
      • to include an alarm that is automatically activated and broadcasts the sexually dangerous preda-
          tor’s location if the global positioning satellite monitor is removed or tampered with by anyone
          other than a law enforcement official designated to maintain and remove or replace the equip-
          ment.
b
  Intentionally removes, defaces, alters, destroys, or fails to maintain the electronic monitoring device in
working order; fails to notify the Department of Corrections that the electronic monitoring device is dam-
aged; or fails to reimburse the Department of Corrections or its agent for the cost of monitoring.
c
  This monitoring system is required to provide, at a minimum: time-correlated or continuous tracking of
the geographic location of the monitored person using a global positioning system based on satellite and
other location technology; and an automated monitoring system that can be used to permit law enforce-
ment agencies to compare the geographic positions of monitored persons with reported crime incidents
and whether the subject was in the proximity of such reported crime incidents.
d
  Additionally, any person required to register as a sex offender may, as a condition or probation, parole,
or supervised release, be subject to electronic monitoring. The statute defines electronic monitoring as
including one or any combination of the following technologies: voice verification; radio frequency; video
display/breath alcohol test; global positioning satellite or global positioning satellite-cellular. According to
the statute, each offender is initially placed on radio frequency monitoring. Within 30 days of being
placed on monitoring, the offender will be assessed in order to determine which level and type of monitor-
ing is necessary in order to protect the public.




Source: Developed by LB&FC staff from other state statutes.

                                                      81
 
III. Background Information
       The Pennsylvania Board of Probation and Parole (PBPP), an independent
agency of the Commonwealth, was created to oversee the Commonwealth’s proba-
tion and parole laws. The General Assembly declared in the Pennsylvania Board of
Probation and Parole Law, Act 1941-323, as amended, that:

      The parole system provides several benefits to the criminal justice sys-
      tem, including the provision of adequate supervision of the offender
      while protecting the public, the opportunity for the offender to become
      a useful member of society and the diversion of appropriate offenders
      from prison. In providing these benefits to the criminal justice system,
      the Board shall first and foremost seek to protect the safety of the pub-
      lic. In addition to this goal, the Board shall address input by crime vic-
      tims and assist in the fair administration of justice by ensuring the
      custody, control, and treatment of paroled offenders.

      The Board has exclusive power regarding all parole matters as to any person
sentenced by any court to a maximum term of two years or more in a state or county
prison. Additionally, the Board is to supervise any person placed on probation when
the court by special order directs such supervision.

       Parole is the release of certain inmates from incarceration to continue serving
their sentences in the community under varying degrees of supervision. Inmates
sentenced to incarceration for more than two years become eligible for parole con-
sideration after serving the minimum sentence imposed by the sentencing court.

      Probation is a court ordered sanction that allows a person to remain in the
community under the supervision of a probation officer. The conditions of this
community-based supervision can vary. It could include jail time, fines, restitution,
community service, or other sanctions. Probation can also require a predetermined
amount of time that an offender must report to his or her probation officer. If the
person does not follow the rules of their probation, they could go back before the
judge and be sent to jail or prison. The goal of probation is to ensure public safety.

Board Composition and Authority

       The PBPP consists of nine members appointed by the Governor, by and with
the advice and consent of a majority of all the members of the Senate. To be eligible
to serve on the Board, an individual must have at least six years of professional ex-
perience in parole, probation, social work, or related areas, including one year in a
supervisory or administrative capacity and a bachelor’s degree. Any equivalent
combination of experience and training shall be acceptable. Each member serves a
term of six years, or until a successor shall have been duly appointed and qualified,

                                          82
 
but in no event more than 90 days beyond the expiration of the appointed term.
The members of the Board shall not hold any other public office or employment, nor
engage in any business, profession, or employment during their terms of service as
members of the Board.

      The Board shall have the power and its duty shall be:

      •   to supervise and make pre-sentence investigations and reports as pro-
          vided by law;
      •   to collect and maintain copies of all pre-sentence investigations and re-
          ports;
      •   to collect and maintain a record of all persons who are placed on probation
          and parole;
      •   to collect, compile, and publish statistical and other information relating
          to probation and parole work in all courts and such other information the
          Board may deem of value in probation service;
      •   to establish, by regulation, uniform statewide standards for (i) pre-
          sentence investigations; (ii) the supervision of probationers; (iii) the quali-
          fications for probation personnel; (iv) minimum salaries; and (v) quality of
          probation service;
      •   to adopt regulations establishing specific composition, functions, and
          responsibilities for the Citizens Advisory Committees, and to receive
          reports, recommendations, or other input concerning parole policies and
          parole-related concerns from these committees on a regular basis;
      •   to adopt regulations establishing criteria for Board acceptance of cases for
          supervision and pre-sentence investigations from counties that on Decem-
          ber 31, 1985, maintained adult probation offices and parole systems;
      •   to enter into contracts for purchasing community services to assist paro-
          lees and to supplement existing programs;
      •   to pay the cost of pre-parole drug screening tests for inmates within the
          parole release jurisdiction of the Board who are confined in a state or local
          correctional facility;
      •   to enter into contracts that provide for the continuous electronic monitor-
          ing of parolees;
      •   to establish and provide for intensive supervision units and day reporting
          centers for the supervision of parolees; and
      •   to provide information as required under 42 Pa.C.S. §2153(a)(14) (relating
          to powers and duties) as requested by the Pennsylvania Commission on
          Sentencing.


                                           83
 
       In addition, the Board administers a grant-in-aid to counties that provide ad-
ditional probation staff for pre-sentence investigations and for improved probation
supervision and programming. The grant-in-aid is to help cover the additional
staffing costs for staff who meet the qualifications and standards established by the
Board. The grant-in-aid shall provide 80 percent of the personnel salary costs in-
curred by a county to administer these additional services and programs. If insuffi-
cient funds are appropriated, each county shall receive a prorated reduction in the
grant-in-aid.

       In compliance with the statutory requirement to have specific office locations,
the PBPP maintains a central office in Harrisburg and statewide field and institu-
tional offices divided into three geographical regions: Eastern, Central, and West-
ern. The Philadelphia and Chester areas comprise the Eastern region; Allentown,
Harrisburg, Scranton, and Williamsport areas comprise the Central region; and
Erie, Pittsburgh, Mercer, and Altoona comprise the Western region. See Exhibit 1
on page 5 for a map indicating the locations of these offices.

       The legislation also created an Advisory Committee on Probation to assist the
Board. The Advisory Committee is to consist of nine members, seven of whom shall
be appointed by the Governor, with the consent of a majority of all the members of
the Senate. At least two members shall be judges of courts of record of this Com-
monwealth, at least one shall be a county commissioner, at least one shall be a chief
county probation officer, and the remaining members shall be qualified in the field
of probation and parole either by training or experience. The President pro tempore
of the Senate and the Speaker of the House of Representatives shall each appoint a
member of their respective houses to serve as members of the committee. Members
serve four-year terms and until their successors have been appointed and qualified,
but in no event more than 90 days beyond the expiration of their appointed term.
Members of the General Assembly, judges, and the county commissioner member
continue only so long as he or she remains in that office. Each member of the Advi-
sory Committee shall be paid all reasonable and necessary travel and other ex-
penses incurred in the performance of his duties. The Advisory Committee is
tasked with aiding the chairman and the Board in formulating and reviewing stan-
dards for probation personnel and probation services in the counties. See Exhibit
12 for an organization chart of the PBPP.

       The Board has established special programs aimed toward offender rehabili-
tation and re-integration into the community. Participation in these programs is
determined based on the individual circumstances and needs of each parolee. The
programs include:




                                         84
 
                                                                                                              Exhibit 12
                                          Pennsylvania Board of Probation and Parole Organization Chart

                                                                                                                       CHAIRMAN
                                               Office of Budget                                                                                                                     Office of the
                                                 Comptroller                                                                                                                      Victim Advocate
                                                                                                                      Board Members



                                          Office of General Counsel                                                                                                               Sexual Offenders
                                                Chief Counsel                                                                                                                    Assessment Board




                                           Governor’s Policy Office
                                                Policy Office

                                                                                      Office of Policy, Legislative                            Office of
                                                                                     Affairs, and Communications                      Professional Responsibility
85




                                           Office of Probation                                                                                                      Office of Board                                      Office of
                                          and Parole Services                                                                                                         Secretary                                   Administrative Services




             Bureau of          Western                        Bureau of                        Bureau of Offender                  Research and                    Case Management                     Bureau of                     Bureau of Budget
         Probation Services     Region                      Central Services                   Re-Entry Coordination             Development Division                   Division                     Human Resources                  & Office Services




               Grants and       Central                          Interstate Parole               Specialized Services &                                                Case Analysis                       Personnel
                                                                                                  Community Outreach                                                                                                                        Budget Division
           Standards Division   Region                           Services Division                                                                                       Division                           Division
                                                                                                        Division



             Court Services     Eastern                      Interstate Probation                 Transition Services &                                               Hearing Officers                 Equal Employment
                                                                                                                                                                                                                                      Office Services Division
               Division         Region                        Services Division                   Staff Liaison Division                                                 Division                      Opportunity Office




                                                                                                                                                                                                           Training
                                                                                                                                                                                                           Division




                                                                                                                                                                                                       County Firearms
                                                                                                                                                                                                        Commission




     Source: PA Board of Probation and Parole.



      
      •   York Re-entry Court Program. The Re-entry Court Program is a joint ef-
          fort between the PBPP and specific court of common pleas judges that is
          designed to provide parolees with a history of drug and/or alcohol abuse
          with drug and alcohol treatment services, face-to-face attention from the
          “re-entry court team,” and other special resources. Parolees periodically
          meet with the re-entry court team consisting of a common pleas court
          judge, a Board member, parole supervision staff, drug and alcohol treat-
          ment providers, and mental health providers rather than meeting only
          with a parole agent. Parolees in the Re-entry Court Program participate
          for approximately 12 to 18 months before “graduating” from the program
          and moving to traditional supervision under the Board.
      •   Philadelphia Re-entry Program. The Philadelphia Re-entry Program is
          designed to help parolees succeed by providing assistance with employ-
          ment opportunities or requiring them to participate in community service
          when parolees do not find or maintain employment. In addition to em-
          ployment opportunities, parolees are linked with other services such as
          PA CareerLink and faith-based and community organizations to provide
          valuable resources. Parolees in the Philadelphia Re-entry Program par-
          ticipate for approximately six months before moving to traditional super-
          vision under the Board.
      •   Pennsylvania Community Alternatives to Prison Program (Penn CAPP).
          Penn CAPP is an alternative to state prison for certain technical parole
          violators, with a focus on addiction and relapse issues. Penn CAPP partic-
          ipants are transported to a community corrections center/contract facility
          for a period not to exceed 90 days. The parole agent and the contract facil-
          ity will develop an action plan based on the offender’s needs, and upon
          successful completion of the action plan, the inmate will be returned to the
          community to continue on parole. If the inmate incurs a significant mis-
          conduct in Penn CAPP or otherwise exhibits poor adjustment, he may be
          returned to a state correctional institution to continue with parole viola-
          tion proceedings.
      •   Community Parole Center Program (Halfway Back). The Community Pa-
          role Center (CPC) Program is an alternative to recommitment to confine-
          ment for offenders who are violating technical conditions of their supervi-
          sion. This program is a viable alternative for lower-risk offenders. The
          length of the CPC program is 90 days, however, offenders in need of addi-
          tional time in the program for treatment reasons can be extended for up to
          an additional 90 days.

                     Program Funding and Expenditures

       The operations and programs of the PBPP are funded through a combination
of state, federal, and other funding sources. Approximately 85 percent of the total

                                         86
 
funding that is allocated to the appropriations that comprise the agency is derived
from general state tax revenues. The second largest funding source is derived from
court-imposed fees on offenders under state or county supervision. Federal sources
(e.g., grants from the U.S. Department of Justice) have typically supplied less than
1 percent of the remaining funds available.

      Four General Fund appropriations are available to the PBPP for expenditure
purposes:

      •   General Government Operations,
      •   Improvement of Adult Probation Services,
      •   Sexual Offenders Assessment Board, and
      •   Drug Offender Work Program.

      PBPP expenditures from these appropriations (including funding from feder-
al and other sources) for FY 2005-06 through FY 2007-08 are presented on Table 16.
As shown, the major source of funding for operation of the state parole system
comes from the General Government Operations appropriation. Between 82 and 85
percent of the PBPP’s expenditures from the General Government Operations ap-
propriation has been for personnel services during the three-year period. Table 17
provides a detailed breakdown of personnel expenditures from that appropriation
for FY 2007-08.

       As previously described, the Board administers a grant-in-aid program to
support county adult probation personnel and services. Funding for this program is
provided through the Improvement of Adult Probation Services appropriation and
the county portion of the supervision fees. The grant-in-aid program (covering up to
80 percent of personnel costs for pre-sentence investigations and for improved pro-
bation services) provides these funds to counties based on a historically-designated
staffing level established in 1991.

       Act 1991-35 mandates that a monthly supervision fee be paid by all offenders
under state supervision unless a waiver is granted. Act 35 also requires that the
sentencing judge of the court of common pleas impose upon county offenders, as a
condition of supervision, a monthly supervision fee of at least $25 unless reduced,
waived, or deferred. County supervision fees are to be used to supplement federal,
state, and county appropriations for the county adult probation and parole depart-
ments and are to be used to pay the salaries, benefits, and operational expenses of
those departments. The act provides for the establishment of County Offender Su-
pervision Funds and a State Offender Supervision Fund. Fifty percent of the mo-
nies collected are deposited in the county fund, and the remaining 50 percent are
sent to the state for deposit into the state fund. The 50 percent of county fees sent
to the state are returned to the counties for use in the grant-in-aid program.

                                         87
 
                                                                          Table 16
                                      PA Board of Probation and Parole Expenditures,
                                              by Appropriation and Purpose
                                             (FY 2005-06 Through FY 2007-08)
 
    General Government Operations                                                 FY 2005-06       FY 2006-07       FY 2007-08
    Personnel Services .........................................              $ 65,343,810     $ 66,939,239     $ 72,839,755
    Operational Expenses .....................................                    13,472,519       11,455,182       14,549,194
       (F) DCSI – Sex Offender Treatment ...........                                 248,827               0                0
       (F) DCSI Client Identification .......................                        435,750           64,500               0
       (F) JAG Client Identification Maintenance...                                        0               0           96,386
       (F) JAG-Parole Guidelines ..........................                                0           17,404          59,335
       (F) JAG Advanced Re-entry Training ..........                                       0           17,604         130,000
    Fixed Assets ....................................................                280,766           64,087         982,036
       (F) DCSI Client         Identificationa .....................                       0         144,750                 0

           Appropriation Total ..................................             $ 79,781,672     $ 78,702,766     $ 88,656,706



    Improvement of Adult Probation Services
    Operational Expenses ........................................ $                   91,246   $     146,597    $     172,000
    Grantsb ...............................................................       32,306,167       34,933,402       35,017,484

           Appropriation Total .....................................          $ 32,397,413     $ 35,079,999     $ 35,189,484



    Sexual Offenders Assessment Board
    Personnel Services .........................................              $    1,742,481   $    1,981,665   $    2,181,000
    Operational Expenses .....................................                     1,198,914        1,214,947        1,466,976
       (F) Sex Offender Work Program .................                                13,331           49,383         181,596

           Appropriation Total ..................................             $ $2,954,726     $    3,245,995   $    3,829,572



    Drug Offender Work Program
    Personnel Services .........................................              $      126,048   $     189,101    $     122,510
    Operational Expenses .....................................                        18,712           17,758           35,771

           Appropriation Total ..................................             $      144,760   $     206,859    $     158,281
 
 
                                                   



                                                                              88
 
Table 16 (Continued)
 
    Firearm Education and Training Commission                         FY 2005-06         FY 2006-07       FY 2007-08
    Personnel Services .....................................      $        199,517   $     186,950    $     196,623

    Operational Expenses .................................                 132,517         217,919          251,525
          Appropriation Total ..............................      $        332,034   $     404,869    $     448,148


    Federally Seized Forfeited Propertyc
    Operational Expenses .................................        $             0    $        7,000   $            0
          Appropriation Total ..............................      $             0    $        7,000   $            0


            Grand Total .......................................   $ 115,610,605      $ 117,647,488    $128,282,191
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
_______________ 
aThe amount for FY 2006-07 represents the state matching share of a PCCD grant towards the purchase of CPIN
(Commonwealth Photo Imaging Network) equipment for PBPP offices. This equipment provided real time ability to
use high definition digital photo capability and Livescan fingerprinting capacity to identify offenders through the JNET
(Justice Network) system. The expenses for this purchase appear under both Operational Expenses and Fixed As-
sets, because the costs were divided between equipment costs over $5,000, software, and annual maintenance.
bThis is a grant-in-aid program, which allows the PBPP to reimburse counties for approximately 80 percent of total
eligible salary costs. In addition to the state appropriation, offender supervision fees are collected and returned to the
counties dollar-for-dollar.
cFunding provided as a result of the PBPP’s share of the proceeds of federally seized or forfeited property (Asset
Forfeiture Program, U.S. Department of Justice) acquired through Drug Enforcement Agency (DEA) raids of offender
residences, during which the Board’s parole agents provided assistance.

Source: Developed by LB&FC staff using information obtained from PA Board of Probation and Parole.

                                                                      89
 
                                                               Table 17
                           PA Board of Probation and Parole Breakdown of
                                 Personnel Services Expenditures
                                           (FY 2007-08)

                                                 Description                                      Amount
                       Salaries (Regular Hour) ....................................              $47,709,347
                       Health Benefits ..................................................          8,622,102
                       Hospital Insurance.............................................             6,223,409
                       Social Security Contributions ............................                  3,119,189
                       Retirement Contributions ..................................                 2,247,007
                       Rewards/Bonusesa............................................                1,144,250
                       State Workmen Insurance Premiums ...............                            1,129,484
                       Overtime Hours (Straight Time Rate)................                          795,967
                       Medicare ............................................................        733,259
                       Wages (Regular Hour) ......................................                  458,831
                       Leave Payout ....................................................            298,070
                       Employee Group Life Insurance........................                        124,776
                       General Pay Increment (Cash Payment) ..........                              118,487
                       Allowances ........................................................           63,310
                       Unemployment Compensation ..........................                          29,390
                       Salaries (Higher Class Pay) ..............................                    13,844
                       Salaries (Shift Differential) ................................                12,150
                       Repayments (Salary Overpayment) ..................                            (3,119)
                           Totalb ............................................................   $72,839,755
 
 
 
 
 
 
 
 
 
 
 
 
 
 
_______________ 
aThis amount is not a bonus but represents the lump sum payment of $1,250 that state employees received in July
2007, as the first pay increase of the new union contract covering the period July 1, 2007, through June 30, 2011.
bMay not add due to rounding.

Source: Developed by LB&FC staff using information obtained from PA Board of Probation and Parole.

                                                                    90
 
       The Sexual Offenders Assessment Board (SOAB) was established through
Act 1995-24, as amended by Act 1996-46. The SOAB is to provide detailed assess-
ments regarding offenders convicted of the offenses enumerated in the act, common-
ly referred to as Megan’s Law. The assessments are provided upon request to dis-
trict attorneys to assist in the determination, prior to sentencing, of those offenders
who may be found by the court to be sexually violent predators (SVPs). In coopera-
tion with the SOAB, the Pennsylvania State Police (PSP) maintains the official re-
gistry of all sexual offenders and SVPs in the Commonwealth.

       The Drug Offender Work Program (DOWP) appropriation is a line item ap-
propriation, which began in FY 1989-90, to divert a limited number of drug offend-
ers in the Philadelphia area from receiving sentences to state correctional institu-
tions. The court sentenced these offenders to conditional probation under the
supervision of the Board with a court mandated number of hours to be served in
community service, such as the clean-up of the underpass of I-95. State funding for
this program was discontinued in FY 2008-09.

       Additionally, the Firearm Education and Training Commission (established
by Act 1994-158) provides firearms education and training to county probation and
parole officers who have been authorized by their president judge to carry a firearm
in connection with the performance of their duties. Once certified, all officers are
required to be re-certified annually. The Commission is funded by a restricted re-
ceipts account within the General Fund. Monies for the account are derived from a
$5 fee imposed on any person who accepts Accelerated Rehabilitative Disposition
(ARD), pleads guilty or nolo contendere, or is convicted of a felony or misdemeanor
under the laws of the Commonwealth.




                                          91
 
    92
 
IV. Appendices




      93
                           APPENDIX A

PRIOR PRINTER'S NO. 2321                        PRINTER'S NO.   2369
             THE GENERAL ASSEMBLY OF PENNSYLVANIA



SENATE RESOLUTION
      No. 367                        Session of
                                         2008
INTRODUCED BY KITCHEN, TARTAGLIONE, STACK, WASHINGTON, LOGAN,
ERICKSON, FONTANA, O'PAKE, TOMLINSON, COSTA, C. WILLIAMS, KASUNIC
AND EARLL, AUGUST 4, 2008
SENATOR GREENLEAF, JUDICIARY, AS AMENDED, SEPTEMBER 17, 2008
                            A RESOLUTION

1 Directing the Legislative Budget and Finance Committee to study
2 the Statewide complement of probation and parole officers.
3 WHEREAS, The Pennsylvania Board of Probation and Parole
4 focuses available resources to promote community stability and
5 reduce recidivism in this Commonwealth; and
6 WHEREAS, Effective supervision under the Pennsylvania Board
7 of Probation and Parole includes appropriate agent-to-offender
8 ratios in our counties; and
9 WHEREAS, According to Federal standards and recommendations
10 of the American Probation and Parole Association, the agent-to-
11 offender ratio for moderate to high-risk offenders should not
12 exceed 1 to 50; and
13 WHEREAS, The Pennsylvania Board of Probation and Parole
14 Statewide agent-to-offender ratio is now holding at 1 to 66; and
15 WHEREAS, The probation and parole agent-to-offender ratio in
16 Philadelphia has reached 1 to 148; therefore be it
17 RESOLVED, That the Senate direct the Legislative Budget and


                                94
Appendix A (Continued)


1 Finance Committee to study and report on the Statewide
2 complement of probation and parole officers; and be it further
3 RESOLVED, That the Legislative Budget and Finance Committee
4 identify relevant inequities in funding distribution, caseload
5 assignment, hiring, training and retention, required levels of
6 offender supervision and frequency of contacts; and be it
7 further
8 RESOLVED, That the Legislative Budget and Finance Committee
9 make a report of its findings and recommendations to the Senate
10 no later than October 31, 2008 2009; and be it further <
11 RESOLVED, That a copy of this resolution be transmitted to
12 the Pennsylvania Board of Probation and Parole, 1101 South Front
13 Street, Suite 5100, Harrisburg PA 17104-2517.




                               95
                                       APPENDIX B
                             Selected Pending Legislation
                            Relating to Probation and Parole
                                    (As of September 30, 2009)



Senate Bill 827: Requires the use of global positioning satellite units to supervise sexual
offenders. (Orie)
House Bill 35: Provides for business of the Pennsylvania Board of Probation and Parole and
for power to parole. (Perzel)
House Bill 36: Amends the Prisoner Pre-release Plan Law relating to release plans. (Dally)
House Bill 209: Amends Title 42 and codifies Title 61 relating to adoption of guidelines for
resentencing; for adoption of guidelines for parole and for adoption of recommitment ranges
following revocation of parole; temporary release from county correctional institutions; sentence
of total confinement; parole without board supervision, for judicial power to release inmates and
for transfers of inmates in need of medical treatment; state intermediate punishment; probation
and parole generally; the Pennsylvania Board of Probation and Parole; among other
amendments. (Caltagirone)
House Bill 241: Requires convicted arsonists to register with the State Police; requires the
PBPP to gather certain information relating to convicted arsonists and to devise a form for
notifying arsonists that they must register. (Conklin)
House Bill 306: Relates to search and seizure in relation to supervision by county probation
and parole officers. (Keller)
House Bill 356: Restricts the travel of certain persons who are on parole or probation or who
are serving a sentence of intermediate punishment. (Payne)
House Bill 770: Provides for the power to parole. (Petri)
House Bill 1271: Provides for binding arbitration for parole agents and provides for the ability
to decertify AFSCME. (Dally)
House Bill 1567: Relates to sentencing guidelines and provides for resentencing and parole
and recommitment ranges following revocation and post-prison intensive restrictions. (Boyle)
House Bill 1958: Provides that the courts may require the use of global positioning satellite
units. (DePasquale)
House Bill 1994: Amends the Pennsylvania Board of Probation and Parole Law, providing for
jurisdiction of the Parole Board in certain cases. (Matzie)




 
Source: Developed by LB&FC staff.

                                               96
 
               APPENDIX C
     District Office Organization Charts
97
     APPENDIX C (Continued)
98
     APPENDIX C (Continued)
99
      APPENDIX C (Continued)
100
      APPENDIX C (Continued)
101
      APPENDIX C (Continued)
102




      Source: PA Board of Probation and Parole, as of September 2009.
                                              APPENDIX D
                                   Caseload and Workload Distribution
                                             (September 3, 2009)

                                    # of      Field Agents          % of       Field Agents With         % of
      District Office               Field    With Caseloads         Field          Workloads             Field
                                   Agentsa   100 or Greater        Agents      Greater Than 131b        Agents
  Allentown ................         51              0               0.0%               12               23.5%
  Altoona ...................        21              0               0.0                  7              33.3
  Chester ...................        35              3               8.6                14               40.0
  Erie .........................     23              0               0.0                  4              17.4
  Harrisburg ...............         55              0               0.0                27               49.1
            c
  Mercer ...................         36            10              27.8                 30               83.3
                    d
  Philadelphia ...........          105            21              20.0                 61               58.1
  Pittsburgh ...............         60              1               1.7                18               30.0
  Scranton .................         28              0               0.0                  2               7.1
  Williamsport ............          13              0               0.0                  4              30.8
    Total.....................      427            35                8.1%              179               41.9%




_______________
aFilled field agent positions carrying a caseload. In general, this number does not include agents on extended
medical or military leave and includes agents currently at BTA.
bThese workloads include only hours used for offender supervision. Actual workload hours would likely be higher,
resulting in more agents with workloads over 131, and include other duties such as Board of Pardons investigations.
cMercer and Venango Counties do not have county adult probation departments, therefore, the district office provides
all adult probation services. Many of the county cases are of relatively short duration.
dThe agent count does not include the FAST Unit. That unit has four agents and each has a caseload of 100 or
greater and one has a workload greater than 131.

Source: Pennsylvania Board of Probation and Parole.


                                                         103
104
    APPENDIX E

Response to This Report




          105

								
To top