THE DISENFRANCHISEMENT OF JUSTICE-INVOLVED COLLEGE STUDENTS FROM STATE FINANCIAL AID By Bradley D. Custer A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Higher, Adult, and Lifelong Education – Doctor of Philosophy 2019 THE DISENFRANCHISEMENT OF JUSTICE-INVOLVED COLLEGE STUDENTS FROM ABSTRACT STATE FINANCIAL AID By Bradley D. Custer The collateral consequences of a criminal conviction have long-lasting and devasting effects on people involved in the criminal justice system. From losing the ability to find employment, to being banned from public housing, to losing access to most federal benefits, “justice-involved” people face a life of punishment, even after their sentences are completed. One rarely discussed collateral consequence is the inability of justice-involved people to get state financial aid grants for college. It is well known that incarcerated people cannot get federal Pell Grants for college, but some states impose even more restrictive barriers on state-funded scholarships for students who were previously convicted of crimes or who are currently incarcerated. Through statutory or regulatory eligibility rules, policymakers in over half of states deny aid eligibility to one or more populations of justice-involved student, according to this study’s findings. The eligibility rules of these programs, often decades old, have never been investigated, leaving researchers without an understanding of the scope of the disenfranchisement of justice-involved students from state financial aid. This two-phase dissertation explores this policy issue. Because there was no existing research on this topic to build on, this dissertation begins with a descriptive study that illustrates the financial aid policy landscape for justice-involved students (Phase I). For the first time, I identified all the state grant programs that currently deny eligibility to incarcerated students and students with criminal convictions. The descriptive analysis also identifies when the policies were adopted and explores patterns in where the policies exist across the country. This accounting of policies unveils how commonly states deny aid to justice-involved students, something that policy researchers and advocates should scrutinize. Results from Phase I raised questions about the policymaking process, chiefly: how and why did state legislators adopt these eligibility rules to deny aid to justice-involved students? This is the pursuit of the policy adoption study (Phase II). Using qualitative case study methodology and framed in a conceptual model of state policy innovation and diffusion in higher education, I interviewed state policymakers and studied historical documents to understand the rationales for denying state grant aid to justice-involved students. Findings have implications for the study of state policy adoption as well as for policymakers and advocates seeking to restore state financial aid to justice-involved students. Copyright by BRADLEY D. CUSTER 2019 Dedicated to the memory of Dr. Jamie Dulle (1978-2019), a mentor and friend, passionate advocate for students, higher education scholar, and much more. v ACKNOWLEDGEMENTS I will remember my time in the Higher, Adult, and Lifelong Education program at Michigan State University as a time of personal and professional growth as an interdisciplinary scholar. The HALE curriculum introduced me to new higher education topics that stretched beyond my previous student affairs training, like policy, adult learning, and organizations. Importantly, I also had the flexibility to take courses in criminal justice and public policy. Now and forever, my scholarship will be grounded in these fields, and my work will be relevant to professionals as diverse as policymakers, lawyers, higher education administrators, researchers from multiple fields, and beyond. The members of my guidance committee represent these interdisciplinary fields with which I now identify. I sincerely thank Drs. John Dirkx (advisor and chair), Dongbin Kim, Jennifer Cobbina, and Sarah Reckhow for their expert guidance on my dissertation. Good research is grounded in the work of other scholars. The nearly 30 pages of references in this dissertation are acknowledgements of the people whose excellent scholarship inspired me to conduct this study. Of those I cited, I am especially grateful to a few who were kind enough to talk with me via email, phone, or Zoom about their research and about my dissertation. I thank to Drs. Kyle Ingle, Erik Ness, James Hearn, Adam Wellstead, and Lora Cohen-Vogle for their insights and encouragement. No historical research is conducted without the assistance of others, and this dissertation was no exception. I relied on librarians, state agency officials, and others from across the country to provide the documents I needed to complete this project. A few went above and beyond who deserve acknowledgement. At MSU, Aaron Tomak in the library processed countless interlibrary vi loan requests for policy documents, often trying multiple times on a single document to find what I needed. Brent Domann in the law library searched legal databases for court cases and old state laws, helped me to find hardbound laws in the law library stacks, and taught me much about how state statutes are created, indexed, and archived in Michigan. Outside of MSU, Sue Kleemann at the Illinois Student Assistance Commission and Kristi Carman at the Georgia Student Finance Commission provided documents from their archives that proved critical to my research. I thank these four individuals for their patience and generosity. vii TABLE OF CONTENTS LIST OF TABLES ................................................................................................................... x LIST OF FIGURES ................................................................................................................. xi CHAPTER 1: INTRODUCTION .............................................................................................. 1 Statement of Problem ........................................................................................................... 3 Purpose of the Study ............................................................................................................. 3 Research Questions .............................................................................................................. 4 Significance ......................................................................................................................... 5 Definitions ........................................................................................................................... 7 Conclusion ........................................................................................................................... 8 CHAPTER 2: STATE POLICY REVIEW (PHASE I) ............................................................. 10 Literature Review ............................................................................................................... 10 U.S. Criminal Justice System .......................................................................................... 11 Student Financial Aid ..................................................................................................... 18 Financial Aid for Justice-Involved Students ..................................................................... 23 Conclusion ..................................................................................................................... 32 Theoretical Framework ....................................................................................................... 33 Methods ............................................................................................................................. 35 Sample ........................................................................................................................... 35 Data Collection ............................................................................................................... 37 Analysis ......................................................................................................................... 40 Limitations ..................................................................................................................... 42 Results ............................................................................................................................... 42 Tables and Infographics .................................................................................................. 43 State Profiles .................................................................................................................. 49 Policy Adoption Events................................................................................................... 89 Discussion .......................................................................................................................... 91 Implications ....................................................................................................................... 92 An Evaluation Agenda .................................................................................................... 92 Seeing and Supporting Justice-Involved Students ............................................................ 94 An Advocacy Agenda ..................................................................................................... 97 Inclusive Policymaking ................................................................................................... 97 Theory of Social Construction of Target Populations ..................................................... 101 Conclusion ....................................................................................................................... 103 CHAPTER 3: POLICY ADOPTION STUDY (PHASE II) .................................................... 105 Literature Review ............................................................................................................. 105 State Policy Adoption ................................................................................................... 105 Policy Diffusion and Innovation in State Financial Aid Studies ...................................... 108 Conceptual Model of State Policy Innovation and Diffusion in Higher Education ........... 113 Conclusion ................................................................................................................... 118 viii Methods ........................................................................................................................... 119 Assumptions & Positions .............................................................................................. 120 Methodology ................................................................................................................ 121 Data Collection ............................................................................................................. 122 Sample ......................................................................................................................... 124 Identifying and Contacting Participants ......................................................................... 126 Instrument .................................................................................................................... 128 Interviews..................................................................................................................... 128 Data Analysis ............................................................................................................... 130 Quality ......................................................................................................................... 133 Limitations ................................................................................................................... 135 Results ............................................................................................................................. 137 Pennsylvania ................................................................................................................ 138 Michigan ...................................................................................................................... 147 Illinois .......................................................................................................................... 150 Georgia ........................................................................................................................ 166 Indiana ......................................................................................................................... 173 Oklahoma ..................................................................................................................... 179 Kentucky ...................................................................................................................... 181 Delaware ...................................................................................................................... 185 Wyoming ..................................................................................................................... 188 Utah ............................................................................................................................. 190 Discussion ........................................................................................................................ 192 Policy Adoption Factors ................................................................................................ 192 Two New Factors.......................................................................................................... 204 Other Findings .............................................................................................................. 214 Implications ..................................................................................................................... 216 An Updated Conceptual Model ..................................................................................... 217 For Policy Researchers.................................................................................................. 220 For Policymakers and Advocates ................................................................................... 221 CHAPTER 4: CONCLUSION .............................................................................................. 226 The State Policy Landscape .............................................................................................. 227 State Policy Adoption ....................................................................................................... 230 Policy Recommendations .................................................................................................. 232 Conclusion ....................................................................................................................... 235 APPENDICES ..................................................................................................................... 237 APPENDIX A: State Grant Policies with Rules for Justice-Involved Students .................... 238 APPENDIX B: Brief Recruitment Email ........................................................................... 245 APPENDIX C: Interview Protocol for State Policymakers ................................................. 246 APPENDIX D: Interview Protocol for Other Policy Actors................................................ 247 REFERENCES .................................................................................................................... 248 ix LIST OF TABLES Table 1: Excluded State Financial Aid Programs ..................................................................... 36 Table 2: Five Categories of Eligibility Rules ........................................................................... 44 Table 3: Within State Proportions of Eligibility Rules ............................................................. 45 Table 4: Primary State Policy Adoption Events affecting Justice-Involved Students ................. 90 Table 5: State Financial Aid Programs for Exonerees ............................................................ 100 Table 6: State Policy Actors Contacted for Interviews ........................................................... 127 Table 7: Participant Profiles .................................................................................................. 127 Table 8: Analytical Framework ............................................................................................. 132 Table 9: Degree Program Enrollments and Completions in Indiana State Prisons ................... 177 Table 10: Influential Factors on State Policy Adoption from the Cases ................................... 193 Table 11: Layering and Patching of Michigan Financial Aid Policies ..................................... 213 Table 12: State Grant Policies with Rules for Justice-Involved Students ................................. 238 x LIST OF FIGURES Figure 1: Federal financial aid policy events for justice-involved students. ............................... 23 Figure 2: States prohibiting aid to incarcerated students in at least one program. ...................... 46 Figure 3: States prohibiting aid to students with criminal convictions in at least one program. .. 47 Figure 4: States prohibiting aid to students with drug involvement in at least one program........ 48 Figure 5: All states that deny grant eligibility to at least one type of justice-involved student. ... 49 Figure 6: A conceptual model of state policy innovation and diffusion in higher education. .... 114 Figure 7: An updated conceptual model of state policy innovation and diffusion in higher education. ............................................................................................................................ 219 xi CHAPTER 1: INTRODUCTION The U.S. criminal justice system is arguably framed with two broad philosophical premises: punishment and rehabilitation (see De Luca, Miller, & Wiedemann, 1991; Mills, 1996). Punishment serves to deter crime and reprimand those who committed offenses; rehabilitation serves to correct antisocial behaviors and restore individuals to their communities. It is easy to observe which philosophy reigns. People who have experienced the criminal justice system (“justice-involved”) are among the most oppressed in American society. Those swept into the criminal justice system are more likely to be poor, unemployed, disabled, mentally ill, substance abusing, from minoritized ethnic groups, and less educated than the average American (see Alexander, 2012; Rampey et al., 2016). In prison, they suffer physical and mental abuse, receive limited medical, mental health, and educational services, and lose relationships with friends and family on the outside (see Johnson & Toch, 2000). After prison, the collateral consequences of having a criminal record are so pervasive and systematic that volumes have been written about them, including the loss of employment, occupational licensure, housing, welfare benefits, voting rights, parental rights, privacy, or the ability to serve in the military, participate on juries, or hold public office (Love, Roberts, & Klingele, 2013). Justice-involved people attending U.S. colleges and universities are not immune from the effects of collateral consequences. In fact, U.S. higher education policies at the federal, state, and institutional levels sanction additional consequences on students with criminal records, including restrictions in the areas of admissions, financial aid, campus housing, student employment, athletics eligibility, and more (Custer, 2018a). Notably, the federal government and some states deny financial aid to varying categories of justice-involved students, including incarcerated students or students with prior drug convictions. For example, incarcerated students across the 1 country are ineligible for federal Pell Grants, and students with certain drug convictions are ineligible for the well-known Georgia HOPE Scholarship. For the estimated 2.1 million people in jails and prisons (Kaeble & Glaze, 2016) and the 20 million people with felony convictions in the U.S. (Jacobs, 2015), these eligibility rules in financial aid programs are significant barriers to higher education. The disenfranchisement of justice-involved people from financial aid for college fits within the philosophy of punishment. For decades, higher education has been shown to be a powerful rehabilitative tool for people in prisons; prisoners who completed higher education programs are more likely to find employment upon release and are less likely to commit subsequent crimes, according to a robust meta-analysis (Davis et al., 2013).1 And financial aid programs are understood to be critical policy tools in improving students’ access to and completion of college programs (Bettinger, 2012). If rehabilitation is truly a goal of the U.S. criminal justice system, denying financial aid to justice-involved college students is a contradiction in policy goals. The disenfranchisement of justice-involved students from state financial aid is also a contradiction of higher education policy goals. The goals of state financial aid include increasing access to higher education, rewarding strong academic performance, and incentivizing residents to stay in the state so they contribute to the economy (Heller, 2002; Orfield, 2002). By denying aid to justice-involved students, who are typically less educated than average (Rampey et al., 2016), a state may not achieve its goals of creating a more education populace. 1 A 2018 study by some of the same authors offered an updated meta-analysis that included more recent studies, finding positive effects of postsecondary education on recidivism but null effects on post-release employment (Bozick, Steele, Davis, & Turner, 2018). 2 Although some attention has been paid to issues of federal financial aid for justice- involved students (Gehring, 1997; Page, 2004; Taylor, 2008; Ubah, 2004; Welsh, 2002), there has been no research to explore state financial aid policies that deny aid to justice-involved students or to understand the reasons why these eligibility rules were adopted. The purpose of this study is to fill these gaps by conducting a descriptive analysis of state financial aid programs, their histories, and their eligibility rules, and by conducting an empirical analysis of the factors that led policymakers to deny aid to justice-involved college students. Statement of Problem Although state financial aid programs are designed to improve student access and completion of higher education, some states specifically and intentionally deny financial aid to justice-involved people. For the millions of people in the U.S. who are incarcerated or who have a criminal record, some of whom may seek to enroll in higher education, the inability to get state financial aid is a significant barrier for accessing higher education. Justice-involve students could stand to benefit a great deal from higher education, because a large body of research suggests that higher education attainment is associated with a decrease in criminal recidivism and an increase in employability (Davis et al., 2013; see also Bozick, Steele, Davis, & Turner, 2018). Yet, this population-in-need is disenfranchised from opportunities for state financial aid. Prior to this study, there was no available information on which states deny aid to this population, the history of these policies, or explanations for why state lawmakers adopted them, leaving scholars, administrators, and students without an understanding of the scope of this problem. Purpose of the Study The purpose of this study is to explore the landscape of state financial aid policies that deny eligibility to justice-involved college students. For the first time, this study offers 3 descriptive analysis of state-funded grant programs that deny eligibility to justice-involved students. Then, a deeper investigation into the history and policy adoption process of a sample of states’ financial aid programs offers explanations for how and why justice-involved students were disenfranchised. Research Questions To address the stated problem, two broad research questions are posed: 1. What is the state financial aid landscape for justice-involved students across the U.S.? 2. How and why did state policymakers in some states adopt eligibility rules for justice- involved students in financial aid grant programs? Answering the first research question constitutes the “Phase I” study. Framed in the theory of social construction of target populations (Schneider, Ingram, & deLeon, 2014), Phase I required a careful review of financial aid grant programs in every state to identify: (a) in what states justice-involved students are excluded from eligibility, (b) what types of justice-involved students are excluded, and (c) when the policies were adopted, which altogether constitutes the policy “landscape.” State laws, legislative records, program manuals, state financial aid agency websites, and other policy sources were examined. The product of Phase I is a historical, descriptive analysis of the identified policies, displayed in textual (written state profiles) and graphical (tables and infographics) representations to show patterns and variations in policies. From Phase I results, implications for justice-involved students, future research, and policymakers are discussed. Answering the second question constitutes the “Phase II” study. Where Phase I described the national landscape of state grant policies that disenfranchise justice-involved students, Phase II takes a case study approach to investigate how and why these policies were adopted in a 4 sample of states identified in Phase I. A conceptual model of state policy innovation and diffusion for higher education framed this study by offering a series of factors that were expected to influence policy adoption (Hearn, McLendon, & Linthicum, 2017). Historical documents and interviews with state policymakers were used to answer the research questions. The product of Phase II is a historical narrative of how and why policymakers adopted eligibility rules for justice-involved students in state grant programs, analyzed against the existing state policy adoption conceptual framework. From Phase II results, an updated conceptual model and implications for future research and policymaking are presented. Significance Since states began creating financial aid programs, thousands of justice-involved people across the country may have been denied eligibility for financial aid, possibly ruining their opportunities for enrolling in higher education. This study is significant in that it is the first to expose each state’s policies that contain these disenfranchising eligibility rules and is the first to investigate how and why state lawmakers adopted them. For justice-involved college students, this information unveils the barriers to receiving financial aid they will face in some states, which is important for them to know when deciding on which college to attend. Some states will offer more financial aid opportunities to these students than others. For student advocates working in higher education, this information is a necessary starting point for assisting students in finding alternative funding solutions. State financial aid programs are widely evaluated for impact on college access, choice, retention, completion, debt reduction, return on investment, and other outcomes (see Bettinger, 2012; Bettinger, Gurantz, Kawano, & Sacerdote, 2016). Likewise, the effects of denying aid to justice-involved students deserve evaluation, but before designing evaluations, background 5 information is required. This study presents in-depth, historical descriptions of these policies, their patterns, and the problems they present, which sets the stage for additional policy research and impact evaluations (Loeb et al., 2017). This study is among the first to apply a new conceptual model of state policy innovation and diffusion in higher education (Hearn, McLendon, & Linthicum, 2017). Based on decades of policy research, the model presents a series of factors theorized to influence the adoption of higher education policies in American states. This dissertation examines the extent to which the model identifies the factors that state policymakers reported were influential in their decisions to adopt eligibility rules for justice-involved students in state grant programs. Thus, this study is significant in that it builds a literature base by applying a new conceptual model in the field of higher education. Recommendations for improving the model will be presented based on evidence from the study. This study is also significant for its qualitative case study approach. The policy diffusion literature is critiqued for its reliance on quantitative methods and for merely showing that diffusion exists, whereas developing a richer understanding of how and why policies are adopted is rarer (McLendon & Cohen-Vogel, 2008). Phase II employs qualitative data from interviews, policy data, and archival data to describe the conditions and influences in a selection of states that led to the adoption of certain policies. Thus, this method yields new insights into the policy adoption process previously undiscovered by conventional quantitative analyses. Finally, learning about the policymaking process in past cases can yield new insights for future policymaking. Before this study was conducted, no published research was available to explain why policymakers disenfranchised justice-involved students from state financial aid. Having this knowledge can prepare advocates to defend against future attacks on aid eligibility. 6 Definitions 1. Criminal record: the documentation of a person’s arrests and convictions (misdemeanor and/or felony), drawn from many records across governmental agencies (e.g., courts, law enforcement) and private companies (e.g., background checking companies) (Jacobs, 2015). 2. Eligibility rules: the written requirements that define who qualifies to participate in a program. Eligibility rules can exist in state statutes, state regulations, state agency rules, state websites, program handbooks, program application forms, and more. 3. Felony: generally, “a serious crime punishable by more than a year in prison or by death,” though this varies by jurisdiction (Felony, n.d., n.p.). 4. Justice-involved student: a person enrolled at a higher education institution who has had some involvement with the U.S. criminal justice system. The person could be currently incarcerated, formerly incarcerated, or never incarcerated. Generally, a “justice-involved student” has been convicted of a felony, but the term may also include people convicted of misdemeanors or people who were arrested or charged with crimes (Custer, 2018a; U.S. Department of Education, 2016). 5. Misdemeanor: generally, “an offense for which a person cannot be sentenced to more than one year of incarceration,” though this varies widely by state (Natapoff, 2018, p. 47). Examples of lower-level misdemeanors include loitering, trespassing, jaywalking, disorderly conduct, petty theft, and certain traffic offenses; more serious misdemeanors include certain drug offenses, domestic violence, and drunk driving (Natapoff, 2018). 6. Policy or program: “the decisions (including both actions and nonactions) of a government or an equivalent authority” (Weible, 2014, p. 4). In the context of this dissertation, a policy or program refers to procedures created by state policymakers aimed at providing financial aid 7 to college students. Generally, the terms policy and program are used interchangeably, but perhaps a distinction can be made in this way: written procedures constitute a policy (i.e., statutes, regulations, rules, program handbooks), and the procedures that are enacted from the written policies constitute the program (i.e., scholarships, grants, loans). 7. Policymaker: a person involved in the creation of state-level policy. This might include governors who create executive orders and state statutes, state legislators who create state statutes, state agency officials who promulgate rules or regulations under the authority of state laws, the staff who contribute to those efforts, and others. 8. State financial aid grant program: procedures created by a state government to administer non-repayable, state-funded monies to students to pay for the costs of college. This term is further defined and bounded as the object of analysis in this study (see Methods, Phase I). Conclusion Contrary to state policymakers’ goals of encouraging state residents to attend colleges by providing financial aid grant programs, and despite the likely benefits that flow to justice- involved students who enroll in college, some states explicitly prohibit justice-involved students from receiving financial aid. This study is the first to identify, quantify, and describe all the state grant programs that contain eligibility requirements pertaining to justice-involved students. From this accounting of policies, cases were selected for deeper investigation into how and why state policymakers created the eligibility rules for justice-involved students, offering new insights into the rationales of policymakers and the factors that influence state policy adoption. The results of this two-phase dissertation has implications for state policymakers, researchers, and justice- involved students and their advocates. 8 The remainder of the dissertation is divided among three chapters. Chapter 2 is the “Phase I” study, complete with a literature review, methods, results, and discussion of implications. Likewise, Chapter 3 is the “Phase II” study. Chapter 4 is a conclusion section that summarizes key findings from both phases and outlines the most important takeaways for research and practice. 9 CHAPTER 2: STATE POLICY REVIEW (PHASE I) Phase I of this dissertation takes a national perspective on the problem of denying financial aid to justice-involved college students. To begin, the literature review provides the historical background necessary for situating the study within the federal crime policy and student financial aid policy contexts. The literature review also identifies a total absence of literature addressing state financial aid policies that deny eligibility to justice-involved students. The purpose of Phase I is to fill this gap, using the theory of social construction of target populations to frame the identification of justice-involved students who are denied state financial aid (Schneider, Ingram, & deLeon, 2014). The descriptive results of Phase I lay the foundation for Phase II of this dissertation, which investigates how and why state policymakers denied those students financial aid. Literature Review The state financial aid policies for justice-involved students under review in this study are situated in several historical contexts. The first section describes the context of the U.S. criminal justice system, including its central philosophies, the tough on crime politics era, the prison boom, and the collateral consequences of having a criminal record. The second section describes the context of financial aid policy in the U.S., including the history of federal and state financial aid programs and programs for justice-involved students. Special attention is paid to where these contexts overlap, and I argue that tough-on-crime policymaking likely influenced the gradual erosion of federal financial aid opportunities for justice-involved students. I offer historical evidence from the literature as to how and why federal policymakers denied financial aid to justice-involved students, which informs this study of state financial aid policies. 10 U.S. Criminal Justice System A complete history of the U.S. criminal justice system is beyond the scope of this dissertation, but a brief overview beginning in the 20th Century is necessary for context setting. This section focuses on the two competing philosophies of the modern criminal justice system– rehabilitation and punishment–and provides an explanation for the major changes in federal crime policymaking that led to a massive expansion of the prison population. Finally, I make an accounting of the lasting effects of a criminal record on people and on college students. Rehabilitation. Philosophies for the goals and purposes of the correctional system in the U.S. have changed over time, but rehabilitation has remained central in the minds of most Americans in recent decades. For the first half the 20th Century, the medical model dominated correctional philosophy. Crime was believed to be caused by inherent flaws in an individual as predetermined at birth or through significant mental impairment (Pratt, 2009). As social science improved, rehabilitation became the prominent strategy for addressing criminal behavior. It favored an individual assessment and intervention with a person to change his/her criminal thinking and behavior. In this model, administrative professionals “assumed to be trained experts at behavior modification” were the practitioners primarily responsible for administering rehabilitation programs (Pratt, 2009, p, 19). Rehabilitation reigned for about 20 years, but when U.S. crime rates began to rise in the late 1960s, people began to doubt the effectiveness of rehabilitation. Robert Martinson’s landmark study in 1974 on “what works” with rehabilitation fueled the fire of doubt when he concluded there is “very little reason to hope that we have in fact found a sure way of reducing recidivism through rehabilitation” (p. 49). Immediately after Martison’s paper (see Palmer, 1975) and in the decades since, scholars (including Martinson himself; 1979) have critiqued Martinson’s overly broad conclusion that 11 rehabilitation simply does not work. And despite losing popularity briefly beginning in the 1980s, rehabilitation has remained important to Americans. In multiple studies over time, despite people having generally punitive attitudes towards criminals, the majority still gives broad support to rehabilitation being “an integral purpose of the correctional enterprise” even over punishment (Cullen, Pealer, Fisher, Applegate, & Santana, 2002, p. 131; Pratt, 2009). Thus, rehabilitation deserves a closer look. Rehabilitation is a “planned correctional intervention that targets for change internal and/or social criminogenic factors with the goals of reducing recidivism and, where possible, of improving other aspects of an offender’s life” (Cullen & Jonson, 2011, p. 295). This broad definition allows for many kinds of activities; even in the 1970s, Martinson (1974) found multiple types of rehabilitation, including education and vocational training, individual and group counseling, transforming the environment, medical treatment, sentencing, decarceration (or the release of people from custody), and parole and probation. As such, literature reviews and meta-analyses increasingly focus on specific types of programs, but some scholars attempt to capture the effectiveness of rehabilitation generally. Thousands of studies and over 50 meta-analyses of rehabilitation programs have been conducted (Lipsey & Cullen, 2007; Wormith et al, 2007). Lipsey and Cullen (2007) conducted a “review of systematic reviews” to summarize the findings of meta-analyses, finding positive and large effects on recidivism. They concluded, “The preponderance of research evidence, therefore, supports the general conclusion that rehabilitation treatment is capable of reducing the re-offense rates of convicted offenders and that it has greater capability for doing so than correctional sanctions” (Lipsey & Cullen, 2007, p. 314). Meta-analyses have consistently supported the premise that when psychological and psychosocial treatment programs respond to 12 the needs and risk-levels of clients, they are effective in reducing recidivism (Andrews et al., 1990; Dowden & Andrews, 2000; Hanson, Bourgon, Helmus, & Hodgson, 2009). Relevant to this dissertation, recent reviews have found consistent support for prison higher education programs to reduce recidivism and improve employability post-release (Davis et al., 2013; Mastrorilli, 2016). With this body of evidence, there is little room left for skepticism on the merits of rehabilitation. Punishment. Despite its apparent effectiveness as a correctional approach, rehabilitation is not the dominant force in the U.S. criminal justice system. Rather, punishment is the long- standing thread in the history of American penal philosophy, and punishment came to a new kind of prominence in the “tough on crime” era. As described next, the actions of policymakers beginning in the 1960s caused a dramatic boom in the prison population from about 1980 to 2008, which is now known as the era of mass incarceration (The Sentencing Project, 2017). There are several reasons that societies punish people who violate their rules and norms (Banks, 2005). First, from a moral stand point, it is argued that offenders deserve to be punished and must be held accountable, called retribution. Second, the punishment received by an offender is intended to deter the individual from committing future crimes, called specific deterrence. Third, incapacitating offenders via incarceration is intended to literally prevent them from committing additional crimes, known as the incapacitation effect (Sweeten & Apel, 2007). Fourth, for society in general, punishing offenders also demonstrates that a society’s rules will be enforced, reinforces the actions that the society disapproves of, and discourages others from committing crimes, called general deterrence (Banks,2005). These perspectives on punishment are manifested in U.S. crime policies. The following subsections address the rise in punitive policymaking and the subsequent prison boom. 13 Tough on crime politics. Events of the 1960s and 1970s provide the backdrop for what would become a shift towards punitive policymaking. The assassinations of President John F. Kennedy and Dr. Martin Luther King, white backlash to the activism of African-Americans during the Civil Rights Movement, and steadily increasing crime rates made crime control a hot political issue (Newell, 2013; Oliver & Hilgenberg, 2006). Barry Goldwater, the Republican presidential candidate in 1964, is often credited for starting the “law and order” political rhetoric at the national level (Newell, 2013; Pratt, 2009). Richard Nixon would further capitalize on this theme during his campaign and presidency, arguing that American society was breaking down and needed to be restored through intensified law enforcement (Newell, 2013). These sentiments, as veiled distain for the race riots, proved popular, and voters increasingly came to view crime as a top national issue (Newell, 2013; Pratt, 2009). Later, in the 1980s, Ronald Reagan also campaigned on the racially-charged law-and-order platform, and during his two terms as president, he greatly expanded the federal involvement in crime control (Newell, 2013). Drug use, another veiled racial target, would become the favorite crime problem of the era, with both Nixon and Reagan declaring a “war on drugs” in 1972 and 1982, respectively (Banks, 2005). Then, George H. W. Bush and Bill Clinton both kept up the image of being tough on crime in their successful presidential campaigns, particularly with their public support for the death penalty. Spanning over 40 years, Republicans and Democrats alike in Congress joined in passing major crime bills, including the Omnibus Crime Control and Safe Streets Act of 1968, the Comprehensive Drug Abuse Prevention and Control Act of 1970, the Anti-Drug Abuse Acts of 1986 and 1988, and the Violent Crime Control and Law Enforcement Act of 1994. Contained in these laws were harsh mandatory minimum sentencing laws, three-strikes laws, truth-in- 14 sentencing laws, felon disenfranchisement laws, massive funding for increased law enforcement and prison building, and much more. These laws were the policy instruments that created the prison boom. Prison boom. The U.S experienced a massive boom in the prison population; in 1980, the number of prisoners per 100,000 people was less than 150 compared to over 500 by 2005 (Pratt, 2009). At its peak around 2008, over 2.3 million people were incarcerated, with over 7.3 million under correctional supervision, including those in prison, on parole, and on probation (Kaeble & Glaze, 2016). Although many explanations for the prison boom exist, including population changes, increasing crime rates, shifting economic conditions, and others (Levitt, 2004; Pfaff, 2012; Pratt, 2009), tough-on-crime era policies constitute the best explanation. The drug war policies were especially harmful. According to Alexander (2012), “drug offenses alone account for two-thirds of the rise in the federal inmate population and more than half of the rise in state prisoners between 1985 and 2000. Approximately a half-million people are in prison or jail for a drug offense today, compared to an estimated 41,100 in 1980 – an increase of 1,100 percent” (p. 75). Alexander (2012) outlined the many law enforcement and prosecutorial policies associated with the war on drugs that contributed to the prison boom, including loose search and seizure procedures, citizens misunderstanding their rights, blatant police abuses, cash incentives for drug raids, police use of military tactics and equipment, poor legal representation, coerced plea deals, and much more. These led to more individuals, particularly African-Americans, to be apprehended by law enforcement, many of whom were subsequently sent to prison. Still in 2017, 46% of all people in federal prison are convicted of drug offenses (Federal Bureau of Prisons, 2017). 15 After individuals were picked up by the criminal justice system, new harsh mandatory sentencing laws and three-strikes laws forced judges to sentence even non-violent offenders to prison. Pratt (2009) noted that 77% of prison admissions come from four categories of non- violent offenses: drug possession, burglary, theft and fraud, and drug delivery. Related to mandatory sentences is a policy that proliferated in the mid 1990s: truth-in-sentencing laws. These laws prevented offenders from being released before 85% of their sentence was served, keeping more prisoners in prison longer (Shepherd, 2002). Another policy implicated in the prison boom is parole. Alexander (2012) claimed 35% of prison admissions are due to parole violations, one-third of which for new convictions and the rest from parole condition violations, such as failing drug tests. There are signs that the US is beginning to recover from the prison boom. Correctional populations have steadily shrunk from their peak around 2008; as of 2015, the overall correctional supervision rate of about 1 in 37 adults was at its lowest since 1994, and the incarceration rate was at its lowest since 2004 (Kaeble & Glaze, 2016). Racial disparities, however, remain stark; for example, of all federal prisoners in 2017, nearly 37% are Black even though only about 13% of the U.S. general population is Black (Federal Bureau of Prisons, 2017; U.S. Census Bureau, 2016). Collateral consequences. The punishment of convicted people in the U.S. does not stop at incarceration. Beyond the punishments doled out by courts (e.g., incarceration, fines, probation, sex offender registration), society further censures offenders with what are commonly called collateral consequences (Love, Roberts, & Klingele, 2013). Federal and state laws disqualify people with criminal history from receiving public services, like welfare and public housing, and from exercising civil rights, like the right to vote or hold public office. In addition, 16 employers have a legal right to screen applicants based on criminal history, and professional licensing bodies can prohibit wide categories of offenders from practicing in certain fields. As a result, the criminal record becomes a significant source of stigma that has lasting negative effects on a person’s ability to transition back into society, especially in finding employment (Pager, 2003). People with criminal histories also experience collateral consequences in higher education. A recent review of federal, state, and institutional policies found that admissions policies, financial aid policies, campus housing policies, student employment policies, student athletics policies, and others were barriers to justice-involved college students (Custer, 2018a). For example, most U.S. universities screen for criminal history in the admissions process (Custer, 2016), and a web of state and institutional policies prohibit some students (mostly those convicted of sexual offenses) from living in campus housing (Custer, 2018b). The ways in which justice-involved students are excluded from federal financial aid are described in detail below. Because financial aid is so critical for assisting low-income students to access higher education, blocking aid eligibility is a serious collateral consequence of a criminal conviction. Summary. Beginning in the 1960s, the tough-on-crime political movement caused a paradigm shift in public policy that created a more punitive criminal justice system. The prison population exploded, and only in the past decade have incarceration rates begun to fall (Kaeble & Glaze, 2016). It is within the same period–when tough on crime policies emerged and thrived– that this study of financial aid policies is situated. Alongside the stark changes in crime control policy, federal and state legislators began tinkering with student financial aid policy, creating massive grants and loans programs. But as shown next, lawmakers also aggressively eroded 17 federal financial aid opportunities for incarcerated students and students with criminal histories. This federal context sets up the study of state financial aid policies that follows. Student Financial Aid Government-funded financial aid is among the most important programs for supporting student access and completion of U.S. higher education. This section provides an overview of financial aid policy in the United States, including the history of federal and state programs. After setting the broader historical context, specific attention is then paid to the history of financial aid programs for incarcerated students and students with criminal records. As shown below, justice-involved students once had equal access to federal financial aid, but policymakers gradually took it away, leaving few financial aid opportunities for justice-involved students today. History of federal financial aid. The history of financial aid in the U.S. is as old as higher education itself (Fuller, 2014). Wealthy colonial philanthropists commonly donated money to universities to fund the education of students who could not afford it, a practice that continues today in the form of endowed scholarships (Fuller, 2014). Much later, beginning in the late 1930s, standardized tests became the norm for determining eligibility for admission and scholarships, which represented a shift from the longstanding tradition of need-based funding to merit-based funding (Fuller, 2014). Federal or state financial aid policies as we know them today did not yet exist, but that changed after the World Wars. Born from conflict over veterans’ pensions, the Servicemen’s Readjustment Act of 1944, later known as the G.I. Bill, “was a transformative watershed moment in the history of higher education” (Fuller, 2014, p. 48). The bill set precedent for federal funding issued directly to students rather than to institutions. Its housing loan program set precedent for future federal 18 student loan programs, and generally, the 2.6 million veterans who used their educational benefits between 1944-1956 forced higher education institutions to change how they admitted, funded, and provided services to students (Fuller, 2014). After the adoption of the G.I. Bill, the Perkins Loan predecessor was created under the National Defense Education Act of 1958, and private scholarship programs began to proliferate, such as the College Board’s College Scholarship Service (Fuller, 2014). The next major federal policy was President Johnson’s Higher Education Act of 1965, which created guaranteed student loan programs tied to new institutional quality standards and the first general need-based grant program, the Educational Opportunity Grants program. The 1972 amendments to the Higher Education Act revised the grant program, forming the Supplemental Educational Opportunity Grant program and the Basic Education Opportunity Grant Program, which would be renamed the Pell Grant Program in 1980 after Senator Claiborne Pell (Aschenbrener, 2016). Pell Grants were reserved for students from families with lower annual incomes, were not required to be paid back, and were administered directly to the student (Fuller, 2014). Though funding levels and philosophies on financial aid have ebbed and flowed with subsequent administrations and political contexts, such as the major growth of loan programs at the expense of grant funding after the 1970s (Hearn & Holdsworth, 2004), the general structure of the federal loan and grant programs established in the 1960s and 1970s remains largely unchanged today (Glater, 2018). State need-based programs. Alongside the developments in federal financial aid policy, states began adopting their own programs. At mid-Twentieth Century, states studied their higher education systems, their financial aid programs, and the needs of their residents; what resulted was an expansion of state-funded scholarship programs, like the 1955 California Competitive Scholarship and the 1957 Illinois State Scholarship (see Deutsch, Douglass, & Strayer, 1948; 19 Illinois Higher Education Commission, 1957). The introduction to Pennsylvania’s 1966 State Grant eloquently summarizes why states were investing so heavily in financial aid at the time: Although the enrollments of the postsecondary institutions of higher learning of this Commonwealth and throughout the nation continue to increase at a rapid pace, and although larger numbers of the Commonwealth's children graduate from both the public and nonpublic secondary schools each year, there continues to be a tragic underdevelopment of the Commonwealth's human talent because of the inability of many needy students to finance a postsecondary educational program. The Commonwealth of Pennsylvania can achieve its full economic and social potential only if every individual has the opportunity to contribute to the full extent of his capabilities and only when the financial barriers to his economic, social and educational goals are removed. It is therefore the policy of the Legislature and the purpose of this act to establish a broad- scale State scholarship program designed to guarantee that the most able students from all sectors of the Commonwealth, the most needy students and students with the capability to successfully complete postsecondary educational programs, and deserving postsecondary students are given the opportunity to continue their program of self-improvement in an institution of higher learning of their choice.2 The federal government further fueled the adoption of state grant programs with funding for the State Student Incentive Grant Program created by the 1972 reauthorization of the Higher Education Act (Fuller, 2014). Fast forward to today, nearly every state offers at least one grant program for students with financial need (National Association of State Student Grant and Aid Programs, 2018). Need-based grants, including federal Pell Grants, have been shown to improve 2 Pa. Act of Jan. 25, 1966 (P.L. 1546, No. 541) § 1. 20 the persistence of economically disadvantaged college students (Alon, 2011), improve persistence, course completion, and grades while closing the achievement gap between low- income and higher-income students (Goldrick-Rab, Kelchen, Harris, & Jenson, 2016), and reduce borrowing and subsequent debt (Marx & Turner, 2015). State merit-based programs. Until the 1990s, most state programs were awarded based on financial need (Orfield, 2002). A more recent state policy innovation has been the adoption and spread of merit-based grant programs. Although financial aid based on academic merit is not new (i.e., endowed scholarships), what is now referred to as a “broad-based” merit-aid program is a unique policy invention. These programs typically award scholarships to students based on a minimum high school GPA or standardized test score. They are generally understood to have three purposes: to incentivize high school students to achieve, to increase college enrollment in the state, and to retain students in their home states after college (Doyle, 2006; Heller, 2002). Although the influential Georgia HOPE Scholarship of 1993 is often heralded as the first of the broad-based state merit aid programs (Doyle, 2006; Heller, 2002; Ingle & Ratliff, 2015), Arkansas’ Academic Challenge Scholarship of 1991 deserves some recognition as an early leader (Cohen-Vogel, Ingle, Levine, & Spence, 2008; Dynarksi, 2004). Nonetheless, the lottery- funded Georgia HOPE Scholarship was a revolutionary program that is credited for spurring the spread of broad-based merit-aid grant programs (Ingle & Petroff, 2013; Ingle & Ratliff, 2015). At last count, about 14 states offered broad-based merit aid programs (Ingle & Ratliff, 2015), though funding for Michigan’s Merit Aid Program expired completely as of September 30, 2017 (MI Student Aid, 2017). Merit-aid programs, particularly Georgia HOPE and Florida’s Bright Futures Scholarship, have been frequently evaluated. They and other merit-aid programs have been 21 shown to improve high school performance and college preparation (Harkreader, Hughes, Tozzi, & Vanlandingham, 2008) and increase enrollment (Doyle, 2006; Dynarski, 2004; Zhang, Hu, & Sensenig, 2013). Results on the effects on college completion, however, are mixed (Bettinger, Gurantz, Kawano, & Sacerdote, 2016; Castleman, 2014; Doyle, 2006; Sjoquist, & Winters, 2015; Zhang, 2011). Merit-based aid programs have also earned much criticism for their investment in disproportionately middle- and upper-class, white, and high-achieving students (Orfield, 2002). In the case of Florida Bright Futures, researchers found that money is redistributed from lower- income families to higher-income families. Lower-income people are more likely to play the lottery and less likely to be awarded the top scholarship award, so higher-income families gained an estimated net program benefit of $2,200 while lower-income families incurred an estimated net loss of $700 (Stranahan & Borg, 2004). Researchers have also found that other merit-based aid programs widened the enrollment gap between White and Black and low- and upper-income youth (Dynarski, 2002), and drove up tuition prices statewide, adding more financial burden on students from low-income families (Long, 2002). Despite their popularity, merit aid programs may not be achieving all their intended goals. Summary. This section demonstrates that student financial aid is a key strategy of American higher education policy. In the 75 years since the GI Bill, lawmakers have continued to fund college students, and students in the U.S. have greatly benefitted from it. Thus, the denial of financial aid to any group of students poses significant barriers to college access. As shown next, policymakers have deliberately and continually vied to take away federal financial aid from justice-involved college students. 22 Financial Aid for Justice-Involved Students Financial aid policies necessarily prescribe limits on eligibility to target specific populations of students, but in an on-going controversy, justice-involved college students have been excluded from some financial aid programs. This section provides an overview of policies and available policy research regarding financial aid for justice-involved college students. Figure 1 displays a timeline of the major federal policy events discussed in this section. GI Bill 1944 Higher Education Act 1965 Pell Grants 1972 Some Drug Offenders Regain Pell 2005 Second Chance Pell 2014 1940 1950 1960 1970 1980 1990 2000 2010 2020 Anti-Drug Abuse Act 1988 Drug Offenders Lose Pell 1995 Confined Sex Offenders Lose Pell 2008 Prisoners Lose Pell 1994 Figure 1: Federal financial aid policy events for justice-involved students. Events that represent gains in federal financial aid for justice-involved students are displayed on top, and events that represent losses in aid are on the bottom. Federal aid to prisoners. As early as the 1830s, postsecondary educators were inside correctional facilities to teach incarcerated students, and formalized prison higher education programs developed in the mid-20th Century (Gehring, 1997). Incarcerated students from this period benefitted from federal funding from the Veterans Administration, the Office of Economic Opportunity, the Law Enforcement Assistance Administration, the G.I. Bill of 1944, 23 the Elementary and Secondary Education Act of 1965, the Manpower Development and Training Act of 1962, and the Vocational Rehabilitation Act of 1920 and subsequent amendments (Laird, 1971). But the 1972 Pell Grants created the biggest opportunity for incarcerated students, who qualified for the grants based on family income just like non-incarcerated students (Page, 2004). Pell Grants would become the primary source of funding for a growing number of higher education programs in U.S. prisons (Welsh, 2002). In the 1993-1994 academic year, about 27,000 incarcerated students received Pell Grants, which is just under 1% of the total 3.3 million students who received Pell Grants that year (Zook, 1994). Further, of the $5.3 billion spent on Pell Grants that year, incarcerated students received about $35 million, which is about seven- tenths of one percent (Zook, 1994). At such small proportions, the awards of Pell Grants to prisoners would have had no effects on the ability of other qualified students to get their awards (Zook, 1994). However negligible the expenditure, any amount spent on prisoners proved to be controversial. Soon after the 1972 amendments, public opinion turned against the allocation of Pell Grants to incarcerated students, and a “conservative, anti-correctional education trend in the U.S. Congress” took hold (Gehring, 1997, p. 46). Throughout the late 1970s to the 1990s, conservative legislators complained about prisoners receiving Pell Grants instead of other students (Gehring, 1997; Page, 2004; Taylor, 2008; Ubah, 2004), like Senator Jesse Helms (R- NC) who, on July 30, 1991, said on the floor of Congress: “the American taxpayers are being forced to pay taxes to provide free college tuitions for prisoners at a time when so many law abiding, tax-paying citizens are struggling to find enough money to send their children to college” (Taylor, 1999, p. 111). Congress would shortly thereafter succeed in eliminating federal grants for all incarcerated college students (Page, 2004; Ubah, 2004). 24 Pell Grants for prisoners was hotly debated amidst other contentious crime and punishment topics of the time (for detailed analysis, see Page, 2004). Page (2004) analyzed the political rhetoric of lawmakers and media sources on the “Pell Grants for prisoners” debate between 1991 and 1994. He found five central arguments against Pell Grants: 1) that Pell funds were being diverted from the intended target population, 2) prisoners were incapable of being rehabilitated, 3) the government already funded enough prison rehabilitation programs, 4) crime would increase because people would want to go to prison to get a free education, and 5) it was not fair to crime victims. Alternatively, he found six central arguments in favor of Pell Grants: 1) prisoners were indeed needy and therefore fit the Pell criteria, 2) overall Pell spending on prisoners was small, 3) prisoners were not taking away grants from non-prisoners, because Pell is an entitlement program 4) higher education in prison reduces recidivism, 5) education programs helped to maintain order in prisons, and 6) educating prisoners meant safer streets upon their release (Page, 2004). Despite the vigorous debates, prisoners lost their aid during this period. Though some Congressmen intended to fully remove federal aid from prisoners, other Congressmen resisted, including Senator Pell, resulting in initial legislative concessions (Page, 2004). The 1992 amendments to the Higher Education Act brought the first round of rollbacks in eligibility. Regarding Pell Grants, the award amount for incarcerated students was limited to the cost of tuition, fees, and books and supplies (thereby excluding living costs; DeLoughry, 1992) and people sentenced to death or life without parole lost eligibility altogether (Higher Education Amendments, 1992). In addition, Pell Grants were only to be issued to incarcerated students in a state “if such grants are used to supplement and not supplant the level of postsecondary education assistance provided by such State to incarcerated individuals in fiscal year 1988” (Higher Education Amendments, 1992, p. 481). This was to prevent prison officials from 25 misusing Pell funding for non-postsecondary education expenditures, a common complaint among lawmakers (Page, 2004). Finally, incarcerated students lost eligibility for federal student loans (Higher Education Amendments, 1992). Dissatisfied with these concessions, lawmakers would again change the Pell rules just two years later, constituting the biggest blow to incarcerated students to date. Instead of using Higher Education Act reauthorizations as the vehicle, amendments to the Crime Bill became the host for the next round of proposals on banning aid to prisoners. First, Senator Kay Bailey Hutchison (R-TX) introduced and passed Senate Amendment 1158 in November 1993, and Representative Bart Gordon (D-TN) introduced and passed House Amendment 521 in April 1994, both eliminating aid for all prisoners (Page, 2004). Signed by President Bill Clinton on September 13, 1994, the Violent Crime Control and Law Enforcement Act was a sweeping crime bill that, among many other things, amended the Higher Education Act, in saying: “No basic grant shall be awarded under this subpart to any individual who is incarcerated in any Federal or State penal institution” (Violent Crime Control and Law Enforcement Act, 1994, p. 1828). As a result, once thriving prison higher education programs were drastically cut, resulting in a 44% drop in enrollments the following year (Tewksbury & Taylor, 1996). Said differently, total higher education enrollments in prison as a percentage of the prison population fell “from 7.3 percent in 1994-95, to 4 percent in 1995-96, and 3.8 percent in 1997-98” (Tewksbury, Erickson, & Taylor, 2000, p. 48). Several researchers surveyed the directors of prison higher education programs in each state after the elimination of Pell Grants, consistently finding that program directors reported financial cuts to their programs, decreases in student access and enrollment, reduction in availability of programs, and reductions in state governmental 26 commitments to prison education (Tewksbury, Erickson, & Taylor, 2000; Tewksbury & Taylor, 1996; Welsh, 2002). Over two decades later, prison higher education still has not recovered. Today, it is estimated that 202 institutions – just 4% of degree-granting higher education institutions in the U.S. – offer credit coursework to students in prisons, half of which are community colleges (Castro, Hunter, Hardison, & Johnson-Ojeda, 2018). Ten states have just one institution offering courses in at least one of their prisons, and three states and Washington, DC have none (Castro, Hunter, Hardison, & Johnson-Ojeda, 2018).3 The final erosion of Pell Grants came in 2008 with the elimination of eligibility for sex offenders in civil confinement. A person convicted of certain sex offenses, typically after serving a jail term, can be involuntarily committed to a secure, residential treatment facility if a psychiatric evaluation indicates the person has a mental or personality disorder that contributes to a high likelihood of future sexual offending (Jackson & Covell, 2013). People confined in these facilities were known to enroll in distance education courses and used Pell Grants to fund them, angering some lawmakers (Norton, 2008). Congressman Ric Keller (R-FL) introduced the “No Financial Aid for Sex Offenders Act” in 2003 and 2005 and later succeeded in eliminating Pell eligibility for confined sex offenders via the 2008 amendments to the Higher Education Act (Higher Education Opportunity Act, 2008). Much criticism has come of the elimination of Pell Grants to prisoners and civilly confined sex offenders, and there have been several attempts to restore the grants. First, an incarcerated student attempted to sue over his loss of financial aid, but a federal judge ruled that 3 These figures exclude many prison educational programs, including those that do not require a secondary education credential for admission, that do not award college credit, that are not considered “postsecondary” in nature (i.e., vocational education, skills training), and that are not administered by a federally-recognized higher education institution. For example, over 200 correctional institutions have hosted Inside-Out courses, but not all of them offer college credit. See http://www.insideoutcenter.org/corrections-partners.html 27 Congress had the authority to change the grant program’s rules (“Judge Upholds Ban,” 1995). Organizations like the NAACP in 2007 and the American Federation of Teachers in 2010 issued resolutions in support of restoring prisoners’ grants (Scher, 2017). The Restoring Education and Learning (REAL) Act, which reverses the Pell bans, was introduced in the House in 2015, the Senate in 2016, and again in the House in 2017 (SpearIT, 2016). No Republicans have signed on as cosponsors to the bills, indicating the REAL Act has little chance to pass in the Republican- controlled 115th Congress. Most recently, in August 2018, the organizers of a 19-day nation-wide strike of prison workers included in their demands the restoration of Pell Grants for prisoners (Pilkington, 2018). In 2014, good news came to advocates of prison higher education. The U.S. Department of Education issued new guidance to clarify Pell eligibility: those in federal and state prisons are not eligible for federal Pell Grants; however, students in local, municipal, or county jails are eligible for Pell Grants, as are all individuals in juvenile facilities (Mahaffie, 2014; U.S. Department of Education, 2014). July 2015 brought better news; the Department announced an experimental program to issue a limited number of Pell Grants to incarcerated students in approved prison programs (U.S Department of Education, 2016). Known as the Second Chance Pell program, it was a capstone to the Obama Administration’s agenda for criminal justice reform. In July 2016, 67 higher education institutions were selected to partner with correctional institutions in awarding Pell Grants to nearly 12,000 students (U.S. Department of Education, 2016). Predictably, Second Chance Pell was not without its critics. In 2015 and again in 2017, Representative Chris Collins (R-NY) introduced the sharply titled “Kids Before Cons Act” to prevent the Department from issuing Pell Grants to incarcerated students (Scher, 2017). After only two academic years of implementation, the fate of the Second Chance Pell program under 28 the Trump Administration is still unknown, but Congressional Republicans in early 2019 expressed openness to restoring Pell Grant access to incarcerated students (Douglas-Gabriel, 2019). Federal aid to drug offenders. Prisoners are not the only justice-involved students who lost federal financial aid. Two federal laws affect the ability of students convicted of drug offenses to get federal financial aid. Signed into law on November 18, 1988, the Anti-Drug Abuse Act established guidelines for what is now referred to as the Denial of Federal Benefits Program. The law states, at the discretion of federal or state courts, any person convicted of drug trafficking may be denied federal benefits for up to five years for a first offense, ten years for a second offense, and indefinitely for a third offense; those convicted of drug possession could lose benefits for one year for the first offense and five years for the second (Denial of Federal Benefits, 1988). Among the many federal benefits that can be denied are Pell Grants, federal loans, work study, other educational grants programs (e.g., for adult, vocational, and migrant education), and eligibility for the Upward Bound program (U.S. Department of Justice, n.d.). In other words, as part of a drug offender’s sentence, a judge can permanently strip a person of their ability to get federal financial aid. This law remains in effect today, and no available research documents the extent to which judges have used the law to deny federal financial aid to drug offenders. This law operates independently of the next law, though overlap is possible. Ten years later, Representative Mark Souder’s (R-IN) Drug-Free Student Loan Provision was passed in the 1998 reauthorization of the Higher Education Act (Crawford, 2005). The amendment suspended eligibility for federal loans, grants, and work study to anyone convicted of a drug crime. Depending on the type, date, and number of offenses, a student’s eligibility could be suspended for one year, two years, or indefinitely from the date of the offense. 29 Restoring eligibility required the student to participate in a rehabilitation program (Higher Education Amendments, 1998). According to Souder, the purpose of the amendment was to deter drug crimes, to help students get drug treatment, and to hold financial aid recipients accountable for their use of taxpayer funds (Crawford, 2005). One study refuted the first claim, finding that youth were not deterred from committing drug crimes to get financial aid for college (Lovenheim & Owens, 2014). The new eligibility rule proved to be a significant barrier for many students. The Government Accountability Office “estimated that between 17,000 and 20,000 applicants per year would have been denied Pell Grants, and between 29,000 and 41,000 would have been denied student loans” for their drug convictions between the 2001-02 and 2003-04 academic years (U.S. Government Accountability Office, 2005). In 2005, an amendment to the Deficit Reduction Act greatly scaled back the suspension of eligibility, suspending instead only students whose drug offenses occurred while the student was receiving some form of financial aid. This narrower group of affected students faced the same one-year, two-year, or indefinite suspension depending on the circumstances of the offense(s) and were also required to complete a rehabilitation program to regain eligibility (Deficit Reduction Act, 2005). Even so, data from the Department of Education showed that between the 2013-14 and 2016-17 academic years, 3,989 students were suspended from federal aid eligibility because of drug convictions or because they left the question on the application blank; an additional 1,026 students during that period had their aid eligibility partially suspended for the same reasons (Kreighbaum, 2018). Today, unless a student was permanently stripped of financial aid eligibility by a judge under the 1988 Denial of Federal Benefits program, a college student only loses eligibility for federal financial aid under the 2005 law if they are convicted of a drug crime while receiving federal financial aid. 30 Many opposed the suspension of aid eligibility for drug offenders. In 1999 and 2001, Representative Barney Frank (D-MA) introduced failed bills to repeal the provision (Gehring, 2001). Later, the American Civil Liberties Union represented the Students for Sensible Drug Policy in their lawsuit against the Department of Education; they claimed students’ Fifth Amendment rights were violated, arguing that students were being punished again (double jeopardy) for their crimes. The U.S. Court of Appeals for the 8th Circuit rejected the challenge in 2008 (Walsh, 2008). In February 2016, Senator Bob Casey (D-PA) introduced the Stopping Unfair Collateral Consequences from Ending Student Success (SUCCESS) Act, which would have repealed the drug conviction rule, though the bill not make it past introduction (SUCCESS Act, 2016). As recently as February 2017, the American Bar Association passed a resolution in support of restoring aid to drug offenders (American Bar Association, 2017). State aid programs. Whereas writers have long lamented the loss of federal aid for prisoners, there appears to be no policy reports, historical reports, or research literature about state financial aid programs for justice-involved students, thus creating the need for this dissertation. Only brief mentions are occasionally found in the literature. For example, when Tewksbury, Erickson, & Taylor (2000) surveyed prison education program administrators from each state, they asked how students would fund their education without Pell Grants. In 1995-96, 15.9% of respondents reported that the “state version of Pell” (in other words, state need-based grants) was available, but that fell to just 2.1% in 1997-98, possibly indicating that states cut funding to prisoners following the change to Pell Grants. Perhaps the only other example is that Cohen-Vogel, Ingle, Levine, and Spence (2008) noted that broad-based merit-aid programs often require students to be “drug-free.” The lack of attention to these eligibility rules in the state 31 financial aid research is a conspicuous weakness of the literature, and this dissertation fills the gap and raises awareness of this policy issue. Summary. This section of the literature review revealed the 45-year struggle to retain financial aid for justice-involved college students, especially incarcerated students and students previously convicted of drug offenses. The provision of Pell Grants to incarcerated students in 1972 started a still-raging game of tug-of-war among policymakers. The pattern seems clear. Tough-on-crime legislators chipped away at justice-involved students’ access to financial aid, sparking protest from opposing lawmakers, students, and advocacy groups. Sometimes rights to financial aid were gained back, and then new targets came into the sites of lawmakers seeking to punish students for their past crimes. The matter is unresolved, and justice-involved students are vulnerable to losing their remaining financial aid. Conclusion The goal of this literature review was to present the context for Phase I of this study of state financial aid policies for justice-involved college students. The past 50 years has seen a dramatic change in the politics of crime control. The spirit of rehabilitation, while not entirely eradicated, was eclipsed by a popular, punitive philosophy of policymaking and law enforcement that led to the imprisonment of a historic number of people. At the same time, it appears that providing federal financial support to justice-involved college students also became unpopular. Despite the potential benefits of higher education for this population, federal lawmakers gradually stripped these students’ eligibility for federal financial aid, the consequences of which are not fully known. The condition of state financial aid policies for justice-involved students, however, is yet to be described in the literature. Phase I of this dissertation seeks to identify these state policies and to explore their history. 32 Theoretical Framework Why – according to theory – might lawmakers deny financial aid eligibility to justice- involved students? According to Schneider, Ingram and deLeon (2014), social constructions dictate which populations receive the benefits or the burdens of a policy (see also original article by Schneider & Ingram, 1993). Depending on how positively or negatively society views a group of people (social constructions), a “target population” is either rewarded or punished by public policy. In their four-part typology of social construction and power (the advantaged, contenders, dependents, deviants), justice-involved people would be categorized as deviants because they “lack both political power and positive social constructions and tend to receive a disproportionate share of burdens and sanctions” (Schneider, Ingram, & deLeon, 2014, p. 112). They are viewed as less worthy, undeserving, immoral, and dangerous by the public and “have few if any advocacy groups willing to speak on their behalf” (Schneider, Ingram, & deLeon, 2014, p. 112). Indeed, people generally have punitive attitudes towards people convicted of crimes (Cullen, Pealer, Fisher, Applegate, & Santana, 2002), especially those convicted of sexual offenses (Rogers & Ferguson, 2011). These attitudes appear to become harsher over time, perhaps due to media coverage, political rhetoric, and “moral panic” (Rogers & Ferguson, 2011). Furthermore, the punitive – and sometimes misguided – attitudes that policymakers hold are known to be based on information they get from the media and their constituents, which are then reflected in the harsh, punitive crime policies that they create (Sample & Kadleck, 2008). Thus, a compelling theoretical explanation for why lawmakers systematically strip justice-involved people of public benefits (e.g., financial aid) may be that lawmakers benefit politically from targeting and punishing this powerless, unpopular group of people. 33 The theory is especially useful to understand why policymakers “support policy provisions that distribute benefits at odds with their apparent self-interest” (Schneider & Ingram, 1993, p. 346). Consider the proposition that frames the stated problem for this dissertation. Policymakers establish financial aid programs so that more state residents can afford to go to college. It is in the interest of the policymakers that such programs are successful and lead to a more educated populace. At the same time, they prevent some populations from receiving those benefits, which seemingly works against the broad policy goal. The theory of social construction of target populations offers the best explanation for why: no matter how much the “deviant” population might stand to benefit from the policy, it is far more politically advantageous to exclude them. The review of federal financial aid above provides some supporting evidence to this theory. Throughout the 1990s and early 2000s, some lawmakers expressed clear disdain for people who committed crimes and vehemently argued against justice-involved people receiving federal education benefits. People who are incarcerated, people who committed drug offenses, and people who committed sexual offenses were the targeted subpopulations of justice-involved people who received the most denial of benefits during this time period. To give aid to these populations, or worse, to advocate on their behalf, would be decidedly “soft on crime” and therefore politically untenable. Targeting justice-involved students for the loss of federal aid opportunities was politically necessary to save face with the electorate, according to the theory. The theory of social construction of target populations presents a convincing argument for the need to understand how certain justice-involved populations are socially constructed and targeted by policymakers when it comes to financial aid. While the literature review identified the target populations within the federal financial aid context, there is no available literature to 34 define which groups of justice-involved people have been targeted by state policymakers within the financial aid context. Guided by the theory of social construction of target populations, the Phase I study seeks to identify which categories of justice-involved students were denied the benefits of state financial aid. Methods Because no previous literature has documented the existing state financial aid grant programs that deny aid to justice-involved students, Phase I of this dissertation required a national review and descriptive analysis of state policies. The purpose of Phase I was to unveil this policy phenomenon by providing a detailed description of these policies. An overarching research question and several sub-questions drove this review: What is the state financial aid landscape for justice-involved students across the U.S.? a. Which states have adopted rules for justice-involved students in their financial aid grant programs? b. When did they adopt those rules? c. Which categories of justice-involved students are targeted in state financial aid grant programs? d. What are the observable patterns in these policies? Sample Because there are many kinds of financial aid programs, it was necessary to narrow the scope of this policy review. In the development of their typology of state merit-aid grant programs, Delaney and Ness (2009) identified seven broad forms of financial aid, including grants, loans, loan assumption or forgiveness, conditional grants or loans, tuition waivers, work study, and others. For inclusion in this study, a program must be a grant (i.e., does not have to be 35 paid back by the student) for undergraduate students (i.e., not professional or graduate students), and it must be currently funded by the state (i.e., not discontinued, and not supported by federal or private funding). Table 1 provides examples of common program types that do not meet these criteria. I opted to study only undergraduate grant programs because they are, by far, the most common form of state financial aid and offer the most aid to students compared to the other types of aid. I excluded programs that are defunct because this study is meant to have current relevance for justice-involved students seeking financial aid now and in the near future. Table 1: Excluded State Financial Aid Programs Program Types State Program Examples Federally-Funded State Grant Programs College Savings Plans Mississippi GEAR UP, Michigan GEAR UP, California Chafee Grant, Louisiana Chafee Educational Training Voucher Program, Maine John R. Justice Grant Program Illinois Bright Start, Louisiana Achieving a Better Life Experience 529A Program Loans Texas College Access Loan Program Forgivable Loan Programs Minority Teachers of Illinois Scholarship Program, Next Generation Hoosier Educators Scholarship, Kansas Nursing Service Scholarship Work Study/ Paid Internship Programs EARN (Employment Aid Readiness Network) Indiana, Kansas Career Work Study Graduate and Professional Student Programs Tuition Waivers Programs for Students at Individual Institutions Kansas Osteopathic Medical Service Scholarship, Coal County Kentucky Scholarship Program for Pharmacy Students Utah Tuition Waivers for Wards of the State, Maine Tuition Waiver Program for Public Servants Grant Program for Participants in Southern Illinois University Carbondale Achieve Program, University of North Georgia ROTC Grant Dual Credit Programs Kentucky Dual Credit Scholarship Program Using the National Association of State Student Grant and Aid Programs’ (NASSGAP, 2018) program database as a starting point, I identified all currently-funded grant programs from 36 each of the 50 states and Washington DC. I also reviewed state government websites to find other qualified programs that were not on NASSGAP’s inventory. Together, the 357 identified programs constituted the population of currently-funded grant programs, but additional narrowing was needed to form the study sample. The element of true interest was the presence of eligibility rules for justice-involved students. For inclusion in the study, each state grant program had to contain a written rule that in some way denied eligibility to justice-involved students. Data Collection Answering the research questions of Phase I required a thorough review of state policies that met the sampling criteria. With 357 qualified grant programs identified, I then examined the content of the policies. State statutes, administrative codes, state government websites, and program manuals were reviewed. I analyzed the content of each grant program in search of explicitly written eligibility rules for justice-involved students. In a spreadsheet, I documented the state, name of program, and if applicable, the language of the eligibility rule pertaining to justice-involved students. It bears repeating that the content of state policies was analyzed for explicitly written eligibility rules. The rules were counted exactly as they appeared in laws, regulations, or programs manuals, without interpretation for actual effect. An example best explains why this point is important to note. Consider that Program A has a written rule that renders anyone with a felony conviction ineligible for aid, but it does not stipulate that incarcerated students are ineligible. For this study, Program A would only be recorded as having a felony conviction eligibility rule. However, in effect, most incarcerated students would also likely be ineligible for Program A, because if a person is incarcerated, they were probably convicted of a felony. This revelation has important practical implications for understanding the scope of how many students 37 might be affected by these eligibility rules; for every program that denies eligibility to those with convictions, one can safely assume that most incarcerated students are, by default, also rendered ineligible. In addition, some states may indirectly deny eligibility to justice-involved students as a function of other eligibility rules. For example, in North Carolina, there are no explicit state statutes or program rules regarding the eligibility of incarcerated students, but program rules for several of North Carolina’s major grant programs indicate that students must be eligible for federal financial aid in order to be eligible for North Carolina aid.4 This would mean incarcerated students or students with drug convictions in certain circumstances would not be eligible for the state programs. Officials from Nebraska (personal communication, July 19, 2018) and Virginia (personal communication, July 24, 2018) indicated this would be the case for programs in those states as well, though the program rules and state statutes themselves are not clear on this point. An official from Washington DC commented that a combination of eligibility requirements likely prevents all incarcerated students from receiving awards; the official also reported that no incarcerated student has ever received the DC Tuition Assistance Grant or the DC Mayor’s Scholars Undergraduate Scholarship, though some students with prior criminal histories have been known to receive the awards (personal communication, July 20, 2018). Finally, while some states do not have explicit rules regarding incarceration, some states may not have any eligible higher education programs operating in prisons in the state; thus, no incarcerated students would be eligible to get state financial aid, as may be the case in Idaho (personal communication, July 20, 2018). These situations confounded efforts to identify precisely the states and programs that 4 E.g., Education Lottery Scholarship, Community College Grant Program, UNC Need-based Grant. 38 deny aid to justice-involved students. To reiterate, only programs with explicitly-stated rules that target justice-involved students were counted and discussed. Once the relevant programs were identified, the next step was to identify when the policies were first adopted. I located the currently published version of the grant program’s statute from state government websites. This version sometimes provided annotations about when the policy was first adopted. Other times, I searched online archives of state legislation. I used the Hein Online State Session Laws Library and Historical State Statutes databases to find the original authorizing statute.5 Similarly, I then identified when eligibility rules for justice- involved students first appeared in the policy. In most cases, they were original to the policy, but in some cases, the eligibility rules were added after the original policy was adopted. Annotations pointed me to when laws were amended, or sometimes, I had to search through the legal databases to find versions of the laws over time to find when the eligibility rule was added. The dates of the authorizing statute and when the eligibility rule first appeared in the grant program were added to the spreadsheet. Finally, any other available information about these policies was collected to offer richer descriptions. Examples of sources include news articles, research literature, governmental and non-governmental reports, and information from websites. I also contacted officials from the state agencies that administer financial aid, like the Illinois Student Assistance Commission and Oregon’s Office of Student Access and Completion. I requested documents pertaining to grant programs’ histories, such as historical reports, evaluation reports, informational pamphlets, and program manuals. I contacted officials in every state by email or phone to request information or 5 The HeinOnline State Session Laws Library is an online database containing over 12.7 million pages of state session laws for all 50 states and more with coverage going back to each state’s inception. The HeinOnline State Statutes: A Historical Archive contains over 2.6 millions pages of superseded state statutes for all states, going back to 1717. 39 to ask for clarity on a program’s laws or rules. These communications for Phase I did not occur under the guise of research (i.e., where consent was confirmed); rather, agency representatives responded to technical questions about the policies or provided documents in their official capacaties, as required under public records laws. The collection and analysis of archival documents play a more central role in the methods of Phase II, and a more thorough discussion of these sources is offered in sections below. Analysis Quantitative and qualitative description was the appropriate analytic method for Phase I, because prior research had not yet offered basic information about these policies. “Descriptive analysis characterizes the world or a phenomenon— identifying patterns in the data to answer questions about who, what, where, when, and to what extent” (Loeb et al., 2017, p. v). This method was suitable to answering my research questions that ask: how many and which states adopted these policies, when were they adopted, and who was excluded? In a report from the U.S. Department of Education, Loeb et al. (2017) outlined a procedure for collecting data and conducting descriptive analysis that I followed to answer the research questions. Step one was to identify the phenomenon, which is that some state financial aid policies deny eligibility to justice-involved college students. This was achieved through background research in preparation for this project and through the literature review. Step two was to consider which feature of the phenomenon is most salient for study. The salient features of these policies include the presence of eligibility rules for justice-involved students, the subpopulations of justice-involved students who are affected by those rules, and when the policies were adopted. These data were collected following the procedures outlined above. 40 Step three was to identify the measures that best represent the salient features, including quantitative and qualitative data. Data were drawn directly from the policies and were documented in a spreadsheet, as described above. The spreadsheet contained the state, name of program, type of justice-involved student affected, the year of policy adoption, and if applicable, the year of amendment. These data (i.e., the sum or proportions of the spreadsheet columns) constitute the quantitative descriptive data. When available, qualitative data from historical reports were also gathered. Combining data from the spreadsheet and any other information that I collected, I wrote a descriptive profile for each state of the events related to the creation of policies with rules for justice-involved students. Step four was to determine whether there were observable patterns in the data (Loeb et al., 2017). Quantitative descriptive analysis includes generating simple statistics, such as frequencies (counts) of state policies, proportions of grant policies within each state that do and do not have these rules, and the proportion of total states that contain at least one policy with these rules. In plain numbers, these statistics reveal how common these policies are across the country. In addition, simple statistics also reveal the degree of variation in policies, such as the proportion of policies that target incarcerated students versus students with drug offenses. Similarly, descriptive statistics can reveal basic patterns in the data, such as regional patterns or patterns in the date of adoption. Finally, step five was to communicate these patterns to further describe the phenomenon (Loeb et al., 2017). In the results section, I display a variety of tables and map infographics that are useful for illustrating the findings. Following this step-wise process for descriptive research, I succeeded in collecting sufficient data, identifying the salient policy features, analyzing the data for patterns, and presenting analysis of those patterns. 41 Limitations This dissertation is not a comprehensive history of state financial aid grant programs. The state profiles only describe the key events related to programs with certain eligibility rules. There is much more history to each grant program than these brief accounts, and additional research is needed to document their histories. Indeed, when asked, most state officials did not have reports on the history of their own programs, which is a gap in the study of state financial aid. The sampling approach of Phase I focused on state grant programs because they are the most common and most important form of state financial aid, but states offer many other types of financial aid (see Table 1). Following a similar methodology, researchers may find it important to identify and describe how these other forms of financial aid affect justice-involved students. Similarly, the sampling approach only considered currently-funded programs. Some programs that contained the eligibility rules for justice-involved students have been discontinued and were therefore excluded (e.g., Michigan Merit Award). In future case studies of state policies, including these discontinued policies would present a complete historical account. Finally, the purpose of this study was to describe the policy phenomenon, which is only the foundation for a larger policy evaluation agenda. Much more can be learned about the effects of denying aid to justice-involved students by studying the educational outcomes of justice- involved students who entered higher education despite not having access to state grant funding. An agenda for future impact evaluations of these policies is presented in the implications section of the Phase I study, below. Results Across 50 states and Washington DC, 357 grant programs met the sampling criteria. Of those, 131 (37%) programs across 26 states (51%) contained a rule that affects justice-involved 42 students. The table in Appendix A displays all 131 programs with their eligibility rules and dates of adoption. In this section, results are presented in several formats, including tables, map infographics, and written descriptions. Tables and Infographics The results from Phase I are, by design, descriptive. The tables and infographics in this section present visual representations of the national review of state grant policies and their histories. First, five categories of rules that affect justice-involved students were identified within state financial aid grant programs. Said differently, policymakers targeted five different populations of justice-involved students for denial of education benefits. Table 2 displays a description for each category with the total number of unique policies identified that contain the rule. In most of the 26 states that have one or more of these policies, there is considerable variation in the presence of those eligibility rules across a state’s programs. For example, in Michigan, of the six total programs, three (50%) deny aid to incarcerated students, and one (17%) denies aid to students with certain criminal convictions. The other three have no rules affecting justice-involved students. Alternatively, some states have no variation in their rules, like New York, New Jersey, and Pennsylvania, where incarcerated students are ineligible for all state grant aid (100%). Table 3 displays the proportions of policies within a state that have one of the five eligibility rules. Only the 26 states that have such rules are included. 43 Table 2: Five Categories of Eligibility Rules Eligibility Rule Description # Unique Policies with Rule # States with Rule Incarceration Students who are confined in a penal 83 16 Criminal Convictions institution, jail, or prison. Rarely, policies stipulate whether juvenile facilities are included within the definition of penal institution. Typically, students with any felony conviction. Sometimes certain categories of felonies are delineated (other than drugs). Rarely, certain lesser crimes are included (i.e., misdemeanors).+ Drug Convictions Typically, students with any felony drug conviction. Occasionally, students with any drug conviction, including misdemeanors.+* Federal Drug Conviction Rules Drug-Free Pledge Most commonly reflects the rule for federal financial aid that stipulates anyone who was convicted of a drug crime while receiving federal aid is not eligible for federal aid, pending a reinstatement process. Here, students who are ineligible for federal aid under the drug conviction rule are also ineligible for state aid. Occasionally, rules require students to comply with other federal drug laws (i.e., Anti-Drug Abuse Act). As a condition of receiving state grant aid, students “pledge” or sign agreements to refrain from illegal drug use, sometimes including illegal alcohol use and other crimes. Students who violate the agreement, typically by being convicted of a crime, lose eligibility for the grant. Often, students in middle or high school sign the pledges and agree to stay crime free until they graduate. Other times, students sign the agreement as they apply or accept state grant aid ahead of enrolling in college. 35 14 21 17 4 2 14 4 Notes. +May include misdemeanor drunk driving offenses, in some states. *These drug conviction rules are not double counted in the “criminal convictions” category. 44 Table 3: Within State Proportions of Eligibility Rules State Criminal Convictions Incarcerated Drug Convictions Federal Drug- Free Rules Drug-Free Pledge 17% 0% 0% 0% 0% 0% 7% 0% 0% 100% 0% 0% 0% 0% 0% 0% 9% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 13% 24% 0% 0% 7% 14% 67% 0% 17% 10% 25% 0% 0% 100% 9% 0% 0% 73% 0% 0% 29% 17% 0% 71% 17% 71% 0% 0% 91% 64% 14% 43% 0% 0% 50% 0% 0% 100% 100% 20% 18% 17% 100% 0% 80% 0% 0% 0% 33% 71% AR CA DE FL GA IL IN KY LA MD MI MO MS NJ NY OH OK OR PA SC TN TX UT WA WV WY 0% 0% 0% 0% 91% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 10% 40% 0% 0% 0% 0% 0% 0% 0% 73% 0% 33% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 87% 0% 0% 0% 0% 0% 45 Map infographics demonstrate the geographical spread of these policies. Figure 2: States prohibiting aid to incarcerated students in at least one program. Students who are incarcerated are explicitly barred from receiving aid from at least one state grant program in 16 states across the US.6 Visually, this eligibility rule is represented in all regions of the US, though except for Wyoming, the central states of the American West do not contain these policies. There is a concentration of these policies in a cluster of connected states in Appalachia and the Upper Midwest, reaching west to Oklahoma, south to Georgia, North to Michigan, and East to New York. California and Oregon are represented on the West Coast. 6 It would be misleading, however, to suggest that only the states in Figure 2 bar incarcerated students from aid. As explained in the data collection section, programs barring students with certain criminal convictions would, by default, also bar individuals from aid who are incarcerated. Thus, Figure 5, which represents all the rules for justice- involved students, may be a better representation of where incarcerated students could technically be denied aid, depending on the circumstance. 46 Figure 3: States prohibiting aid to students with criminal convictions in at least one program. Fourteen states deny eligibility to students with criminal convictions (usually felony convictions) in at least one of its grant programs. The distribution of these policies is more scattered than the programs with incarceration rules. Delaware is the lone Northeastern state, and Washington is the lone West Coast state, while Florida and South Carolina are separated in the Southeast. In the middle of the country, there are several small clusters of bordering states. 47 Figure 4: States prohibiting aid to students with drug involvement in at least one program. When the three categories of rules relating to drug offenses are combined in one map, a visual pattern becomes clearer. States in the South tend to deny eligibility based on drug involvement, though not uniformly. No states in the West deny aid to drug offenders, and Indiana is the Great Lakes state to do so. 48 Figure 5: All states that deny grant eligibility to at least one type of justice-involved student. Finally, to visualize all the states where at least one type of justice-involved student is barred from at least one state grant program, Figure 5 represents all 26 states identified in this study. There is a block of states from Texas to Florida, up through the Midwest of the country to New York that is nearly solid. Only the tip of the Northeast and a swath of the Midwest are completely devoid of grant programs that explicitly deny aid to justice-involved students. State Profiles The tables and infographics have thus far described the policies identified in this study in terms of their patterns and relationships with one another, but each policy was unique and warranted individual description. For each of the 26 states identified as having at least one policy 49 where justice-involved students are affected, a brief state profile is provided.7 These profiles offer chronological descriptions of the policies within a state by explaining when the policies were adopted and amended, which justice-involved students are affected, and the mechanisms by which justice-involved students were affected (i.e., state statutes, agency rules, court decisions, etc.). The state profiles do not offer complete backgrounds or contexts, nor do they capture a state’s entire history of financial aid policymaking. Generally, the descriptions were derived from available legislative documents (i.e., statutes, bills), rules and regulations (i.e., state administrative codes, state bulletins or registers), financial aid agency rules (i.e., program manuals, grant program applications), secondary sources (i.e., reports on program history), or email or phone communications with state financial aid agency program administrators. In some cases, fuller descriptions of policies and their contexts are reserved for deeper exploration in Phase II, when additional data collection, including research interviews, occurred. Arkansas. Though Arkansas currently administers six grant programs,8 only one affects justice-involved students. The Arkansas Academic Challenge Scholarship of 1991 is arguably the first in what would be a wave of new merit-aid programs in the country, coming two years before the heralded Georgia HOPE program (Cohen-Vogel, Ingle, Levine, & Spence, 2008; Dynarksi, 2004). The program offered full-tuition scholarships to Arkansas residents who had financial need, academic merit (i.e, 2.5 GPA and 19 ACT score), and who completed a college- prep high school curriculum, among other requirements. An original stipulation of eligibility 7 The heavy reliance on state legislative documents in this section makes conforming to the Publication Manual of the American Psychological Association (APA) difficult. Instead, Bluebook-style legal footnotes are used for state statutes, bills, administrative codes, registers, and the like. APA citations are used for research articles, websites, and other documents. 8 Throughout the State Profiles section, I mention the total number of grant programs in a state compared to the number of programs that affect justice-involved students. As a reminder, this number of total programs represents those programs that meet my definition of a state grant program (see Chapter 2 Methods). Most states offer additional financial aid programs that do not meet this definition. 50 affecting justice-involved students was that “applicants must certify that they are drug-free and must pledge in writing on the application form to refrain from the use or abuse of illegal substances in order to maintain eligibility for this program.”9 Senate Bill 299, sponsored by Senator Jerry Bookout and others, was passed as Act 352 on March 5, 1991 and later signed by Governor Bill Clinton.10 Today, students certify their drug-free pledge when they sign to accept their award. In 2009, the state legislature created the Arkansas Scholarship Lottery to provide funding for higher education scholarships, following other southern states like Texas, Georgia, and Tennessee (Pittman, 2017). Legislators also tinkered with the basic eligibility requirements of the Academic Challenge Scholarship by adding: “The applicant is not incarcerated at the time of the application for or during the time the applicant receives a scholarship under this subchapter.”11 Senate Bill 26, sponsored by Terry Smith and others, and House Bill 1002, sponsored by Robbie Wills, were passed on March 25, 2009 as Acts 605 and 606, respectively. Today, the state financial aid agency does not collect data on students who lose eligibility for the program because of substance use or incarceration, which are rare occasions (personal communication, July 11, 2018). California. California currently administers seven grant programs, five of which affect justice-involved students. The four Cal Grant programs,12 as they exist today, were created in 2000,13 but these programs have historical roots in the Competitive Scholarship of 1955,14 the 9 Act of Mar. 5, 1991, No. 352, 1991 Ark. Acts 867. 10 Id. 11 Ark. Code Ann. 6-85-206 (8). 12 Cal. Edu. Code § 69430 – 69460. 13 Ortiz-Pacheco-Poochigian-Vasconcellos Cal Grant Act, Cal. Stats. 2000, Ch. 403. 14 Cal. Stats. 1955, Ch. 1846. 51 College Opportunity Grant of 1968,15 the Occupational Education and Training Grant of 1973,16 the California Educational Opportunity Grant of 197517 and other earlier scholarship programs (Century Foundation, 2018). Now, the Cal Grant High School Entitlement Award is a grant for recent high school graduates; the Cal Grant Transfer Entitlement Award is for students who transfer from a community college to a four-year university; the Cal Grant Competitive Award is for students who do not qualify for either of the entitlement awards; and the Cal Grant C is for students pursuing non-transferring associate degrees and vocational certificates. The 2000 reconfiguration of these programs was described as “the greatest expansion of access to higher education in California since the federal government implemented the G.I. Bill” (California Student Aid Commission, 2016, p. 3). Senate Bill 1644 was a bipartisan effort sponsored by, among others, Senators Deborah Ortiz (D), Chuck Poochigian (R), and John Vasconcellos (D) and Assemblyman Rod Pacheco (R), for which the 2000 act is named.18 However, prior to the 2000 amendments, nothing in the various statutes or in the Cal Grant program manuals suggested that incarcerated students would be ineligible, but such a rule was adopted in the 2000 law.19 While many other Californians gained access through these expanded grant programs, incarcerated students lost access. In 2013, the Middle Class Scholarship was created to offer financial aid to students with family income up to $150,000.20 Eligibility for this scholarship was tied to the general eligibility requirements of the Cal Grant programs, thus barring incarcerated students. 15 Cal. Stats. 1968, Ch. 1410. 16 Cal. Stats. 1973, Ch. 981, § 7. 17 The first to be called “Cal Grant”; Cal. Stats. 11975, Ch. 1270. 18 Ortiz-Pacheco-Poochigian-Vasconcellos Cal Grant Act, Cal. Stats. 2000, Ch. 403. 19 Cal. Edu. Code § 69433.9. 20 Cal. Edu. Code § 70022. 52 Delaware. Delaware currently administers eight grant programs, but only one affects justice-involved students. Created in 2005, the Student Excellence Equals Degree (SEED) program pays full tuition to state residents pursuing an associate of arts degree at Delaware Technical and Community College or the University of Delaware who have at least a 2.5 high school GPA.21 However, “the student shall not have been convicted of any felony and the student's parent, legal guardian, or relative caregiver shall certify such fact.”22 The bill was sponsored by Senator Harris B. McDowell (D), Representative Nancy Wagner (R), and Representative Stephanie Ulbrich (R) and was signed on September 6, 2005 by Governor Ruth Ann Minner. Additional information about the SEED program appears in the Phase II study. Florida. Florida currently administers 17 grants and scholarships, but only the four programs comprising the Bright Futures Scholarship program affect justice-involved students. Bright Futures is a lottery-funded, merit-based grant program created in 1997.23 The original program delineated three awards: Florida Academic Scholars have the highest academic eligibility requirements and receive the highest awards; Florida Merit Scholars have slightly lower academic eligibility requirements and receive slightly less funding; and the Florida Gold Seal Vocational Scholars program funds students in career education or certificate programs. In 2016, the legislature created a fourth award: Florida Gold Seal CAPE (Career and Professional Education) Scholars for students who earned credit hours towards specific CAPE industry certification programs. Original to the 1997 programs is a rule stating that in order to be eligible, a student must: “Not have been found guilty of, or entered a plea of nolo contendere to, a felony charge, unless the student has been granted clemency by the Governor and Cabinet sitting as the 21 75 Del. Laws, c. 222. 22 Id at §3404A. 23 Ch. 97-77, at 422, Laws of Fla. (1997). 53 Executive Office of Clemency.”24 Despite a fairly robust research literature on the Florida Bright Futures program, none of the published studies offer analysis of the impact of this eligibility rule or any historical explanation for how or why it was included (see Castleman, 2014; Harkreader, Hughes, Tozzi, & Vanlandingham, 2008; McKinney, 2009; Stranahan & Borg, 2004; Zhang, 2011; Zhang, Hu, & Sensenig, 2013). Senate Bill 858 was sponsored by the Senator Donald Sullivan (R) and was signed by Governor Lawton Chiles on May 23, 1997. Georgia. The state of Georgia currently funds eleven grant programs administered by the Georgia Student Finance Commission. Georgia is unique in that it is one of the only states to have passed a law that targets the financial aid eligibility of all students with drug convictions, the Drug-Free Postsecondary Act of 1990. In addition, all incarcerated students are ineligible for state aid, though this occurred over time through a series of regulatory actions. The Drug-Free Postsecondary Act of 1990. The Drug-Free Postsecondary Education Act was sponsored in the Georgia House of Representatives as House Bill 1231 by Thomas Bryant Buck III (D-95th), William J. Lee (D-72nd), Larry Walker (D-115th), and Bob Lane (D-111th), and Calvin Smyre (D-92nd), and others (Georgia, 1990). First introduced on January 8, 1990, it was approved by the Committee on the University of Georgia System on January 31 and was approved by the House with a vote of 148 to 2 on April 16, 1990 (Georgia, 1990). The new law laid out consequences for students who were convicted of drug crimes. According to the authors of the law: The General Assembly finds that the manufacture, distribution, sale, possession, or use of marijuana, controlled substances, or dangerous drugs in an unlawful manner is a serious threat to the public health, safety, and welfare and to the academic achievement of 24 Fla. Stat. § 1009.531(1)(e) (2018). 54 students enrolled in the public and nonpublic colleges, universities, and postsecondary technical institutes of this state. It is declared to be a primary purpose and goal of this state, of all of its agencies and instrumentalities, and of all of its public officials and employees to take all reasonable steps possible to eradicate the unlawful manufacture, distribution, sale, possession, and use of marijuana, controlled substances, and dangerous drugs. With this purpose in mind, the General Assembly declares that the public and nonpublic colleges, universities, and postsecondary technical institutes in this state must be absolutely free of any person who would knowingly manufacture, distribute, sell, possess, or use marijuana, a controlled substance, or a dangerous drug in an unlawful manner.25 Under the Act, students at public institutions were to be immediately suspended upon conviction for a drug crime, but students at private students were only to lose eligibility for state financial aid for one school term following the date of the conviction. Because the state of Georgia does not publish transcripts of floor debates or committee reports (University of Georgia, 2018), it is not clear what motivated the creation of this law. However, the journals of the Georgia House offer context that hint at why the Drug-Free Postsecondary Education Act might have been created. On the same date that Bill 1231 was introduced, at least 16 other anti-drug bills were also introduced in the House, including bills related to drug testing, sanctions for drug convictions, drug-free workplaces, and a bill to punish college student organizations for drug violations (Georgia, 1990). Furthermore, the Governor’s Commission on Drug Awareness and Prevention released a report touting the 35 anti-drug bills and resolutions passed in the 1990 session (Governor’s Commission, 1990). This suggests that the Georgia legislature, overall, 25 Drug-Free Postsecondary Education Act, Ga. Code Ann. §20-1-21 (2018). 55 actively pursued an anti-drug legislative agenda, and the Drug-Free Postsecondary Education Act was just one of many anti-drug initiatives that year. Still, additional research is needed to uncover the motivations to target the financial aid of college students convicted of drug offenses. The Act has never been substantially amended by the Georgia General Assembly and is still in force today. The Drug-Free Postsecondary Education Act stipulated that the Board of Regents of the University of Georgia, the State Board of what is now the Technical College System of Georgia, and individual private institutions were to promulgate rules on how to implement this law. Although the law took effect on July 1, 1990, it is not clear how the law was enforced in the early years. After the creation of the HOPE scholarship programs in 1993, the Georgia Student Finance Commission assumed broader responsibilities for administering Georgia’s financial aid programs and began implementing rules. Despite the distinction in the law that only students at private colleges would lose financial aid for drug convictions, the Commission would come to enforce the rule for all students, regardless of institution type. There is no available explanation for why the Commission interpreted the law in this way. In the first regulations published by the Commission in September 1993 for the HOPE scholarship programs, a new rule was promulgated about the law: A student is ineligible if, in accordance with the Drug-Free Postsecondary Education Act of 1990, he/she has been convicted for committing certain felony offenses involving marijuana, controlled substances, or dangerous drugs. A student is ineligible to receive a HOPE grant from the date of conviction to the completion of the next academic term. (Georgia Student Finance Commission, 1993, p. 20) 56 Similar eligibility statements were added to the program regulations for all the grant programs administered by the Commission and are still in force today.26 According to an official at the Georgia Student Finance Commission, the Commission has never studied the impact of the Drug-Free Postsecondary Education Act (Personal communication, 2018). Incarcerated Students. Currently, incarcerated students are ineligible for all of Georgia’s grant programs that are administered by the Commission. How this came to be was much less coordinated than how students convicted of drug offenses lost eligibility. Using the best available records, this is an account of how and when incarcerated students lost eligibility for grant aid. Tuition Equalization Program. On April 14, 1971, the Georgia Assembly approved Act 803 (Senate Bill 141) to enact a grant program for students attending private institutions, which was approved by general election in November 1970. Later called the Tuition Equalization Grant Program (TEG or GTEG), the grant offers aid to students attending private institutions. One eligibility rule original to the law that remains today states: “The term ‘eligible student’, as used herein, shall not apply to any person who knowingly promotes or engages in any activity which is determined by the approved institution governing body to be detrimental to the institution,” 27 though it is not clear how this is interpreted and enforced. Still, nothing in this statute, then or now, makes any reference to incarcerated students, but it appears that incarcerated students were able to get this grant for decades before being blocked. When the Commission began issuing regulations in the 1990s for the Tuition Equalization Grant, it made one reference to incarcerated students. TEG recipients must be enrolled full-time taking 12 credits per semester, but “in the GTEG program, 10 academic hours 26 Including the Tuition Equalization Grant, HOPE scholarships and grants, Zell Miller scholarship and grant, Georgia HERO, and Georgia Public Safety Memorial Grant 27 This is likely one example of the many state laws enacted in the late 1960s and early 1970s in response to incidents of campus unrest, mirroring the enactment of a similar rule in federal financial aid program rules in 1968. 57 for students who are inmates incarcerated by the Georgia Department of Corrections is considered a full-time load” (Georgia Student Finance Commission, 1995, p. 3). This definition remained in the TEG regulations through the 2012-2013 academic year. However, in the 2011- 2012 regulations, a section 604.8 “Incarceration” was added, stating: “Through Summer term 2011, incarcerated students who are otherwise eligible may receive TEG funding” (Georgia Student Finance Commission, 2011, p. 14). In the regulations of the following year, section 604.8 stated: “A student is ineligible for TEG payment while Incarcerated. Upon release from Incarceration, such student may begin receiving TEG payments, if he or she meets all other TEG eligibility requirements,” as it still does today (Georgia Student Finance Commission, 2012, p. 14). There were no statutory changes compelling this change in eligibility for incarcerated students, and it is not clear what prompted it. HOPE programs. The Georgia HOPE program is arguably the most celebrated state financial aid program in the country, credited for the rapid spread of broad, merit-based aid programs across the Southeast states and beyond (Dynarski, 2000). HOPE (Helping Outstanding Pupils Educationally) was created on November 2, 1992 when voters in Georgia approved a new lottery intended to fund education programs. Although no statutes were passed to define the rules of the scholarship program, the Georgia Student Finance Commission – up until then primarily responsible for providing and servicing student loans – acted quickly to issue HOPE’s first regulations in September 1993 (Georgia Student Finance Commission, 1993). That fall semester, the first HOPE scholarships were issued. While most people think of the HOPE program as a merit-aid scholarship, there are several programs within HOPE. Originally, there was a merit- based scholarship for degree-seeking students attending public institutions (HOPE scholarship), a scholarship for certificate-seeking students attending public institutions (HOPE grant), a 58 scholarship for students attending private institutions, and a scholarship for students who obtained a GED and are pursuing additional postsecondary education (HOPE GED). These programs were reconfigured over time.28 Like the TEG program, incarcerated students initially had access to HOPE funding but lost it over time through regulatory and statutory changes. First, special provisions were carved out for incarcerated students applying for the HOPE GED voucher in the first regulations published in September 1993: Incarcerated students who receive GED vouchers present a special case. They often have the opportunity to continue their education beyond the GED while incarcerated, and therefore, should be evaluated for HOPE eligibility. However, the "costs of attendance" for an incarcerated student are much more restrictive than for their non-incarcerated peers. For example, they have no costs for such items as room and board or transportation. Therefore, GSFC will hold the GED voucher for incarcerated students and will pay to the teaching institution any direct instructional costs - such as books - that the student incurs while pursuing a higher education. The balance of the voucher will be held by GSFC for the student when he or she is free from incarceration and enters a college or technical school in Georgia. (Georgia Student Finance Commission, 1993, p. 10) For those seeking the HOPE grant at private colleges, it was stipulated: “Incarcerated students must be taking 12 credit hours per quarter to qualify for the HOPE grant, the same as their nonincarcerated counterparts; rather than the 10 hours required of incarcerated students in the existing TEG program” (Georgia Student Finance Commission, 1993, p. 11). The 1993 28 The HOPE grant for students attending private colleges was phased out between 1996-1999, and instead, students attending private colleges became eligible for the merit-based HOPE scholarship in 1996. The HOPE career grant was created in 2013. 59 regulations made no other mention of incarcerated students, so it is unclear if incarcerated students were eligible for the other HOPE programs. Just two years later, in the 1995 regulations, incarcerated students would lose eligibility for all HOPE programs.29 This change was described as a “regulatory improvement” as opposed to a “legislative improvement” indicating it was decided by the Commission, not the Georgia General Assembly (Georgia Student Finance Commission, 1995). This policy change is explored in depth in Phase II. Until 1998, the regulations issued by the Georgia Student Finance Commission were the only official rules for the HOPE programs. In the Georgia state statutes, the only references to HOPE were in appropriations bills and other brief mentions. But in 1998, the Georgia General Assembly passed Act 799 (House Bill 1556), which for the first time codified program definitions and eligibility rules for HOPE programs, including the rules about drug convictions and incarceration. As it still reads today: A student is ineligible for any scholarship or grant described in this part if the student: (5) Has been convicted of a felony offense involving marijuana, a controlled substance, or a dangerous drug as set out in Code Section 20-1-23 or 20-1-24 of the "Drug-free Postsecondary Education Act of 1990," provided that such ineligibility extends from the date of conviction to the completion of the next academic term; (6) Is incarcerated.30 House Bill 1556 was sponsored by Representatives Mike Polak (D-67th), Calvin Smyre (D-136th), Larry Walker (D-141st), Bob Irvin (R-45th), Kathy Ashe (R-46th), and others 29 A caveat is that students who earn a GED while incarcerated are still eligible for the HOPE GED voucher upon release. Under normal circumstances, the voucher is only available to students who earned their GED from a Georgia public or private institution. 30 Ga. Code Ann. § 20-3-519.1 (2018). 60 (Georgia, 1998).31 On March 6, 1998 it passed the House with a unanimous vote of 156 ayes (Georgia, 1998). The history of how and why the HOPE program was codified in 1998 is also explored in Phase II. When the HOPE Career Grant was created in 2013, these 1998 eligibility rules applied to it, which was reflected in the Commission’s HOPE Grant regulations that year. Public Safety Memorial Grant. The Law Enforcement Personnel Dependents Grants were created by the Georgia Assembly in 1980 for children of deceased of disabled public safety officers, firemen, or prison guards. In the Georgia Code, the program is called Grants for Children of Law Enforcement Officers, Firefighters, and Prison Guards.32 When the Georgia Student Finance Commission first issued regulations for the program in 1995, the program was called the Law Enforcement Personnel Dependents Grants, but sometime later, the commission began calling it the Public Safety Memorial Grant Program, as it is known today. In the 2015- 2016 program regulations, the Commission added section 1204.8: “A student is ineligible to receive GPSM Grant funding while incarcerated” (Georgia Student Finance Commission, 2015b, p. 7). It is not clear what instigated this rule change or if incarcerated students were ever eligible for the award before 2015. Georgia HERO. The Georgia HERO (Helping Educate Reservists and their Offspring) Scholarship was created by the Georgia Assembly in 2005 for members of the Georgia National Guard and their children to attend college. Nothing in the original statute or in the statute as it appears today mentions eligibility for incarcerated students,33 but because the Georgia Student Finance Commission administers the program, the commission first issued regulations for the 31 Notice that Representatives Smyre and Walker were also sponsors of the Drug-Free Postsecondary Education Act of 1990. 32 Ga. Code Ann. § 20-3-450 et seq. (2018). 33 Ga. Code Ann. § 20-3-485 et seq. (2018). 61 2005-2006 academic year. Like the Public Safety Memorial Program, it was not until the 2015- 2016 regulations that the commission inserted section 804.8: “A student is ineligible to receive HERO funds while incarcerated” (Georgia Student Finance Commission, 2015a, p. 7). It is not clear what instigated this rule change or if incarcerated students were ever eligible for the award before 2015. Zell Miller programs. Georgia created two new scholarship programs in recent years, including the Zell Miller Scholarship in 2011 and the Zell Miller grant in 2014. The Georgia Student Finance Commission issued new rules for both programs that included the incarceration rule from the start. Illinois. Illinois currently administers 11 grant programs, seven of which affect justice- involved students. The Illinois Monetary Award Program (MAP) grant was created in 1967 for financially needy students and includes a stipulation that applicants be of “good moral character.”34 Interpreting this rule proved to be problematic, which is explained in detail in Phase II. In 1989, the Illinois legislature created a caveat to the definitions of higher education institutions, which effectively barred most incarcerated students from receiving state financial aid.35 The definition stated, as it still does: “For otherwise eligible educational organizations which provide academic programs for incarcerated students, the terms ‘institution of higher learning’, ‘qualified institutions’, and ‘institution’ shall specifically exclude academic programs for incarcerated students.”36 In addition to the MAP grant, the provision applies to the Grant Program for Dependents of Police and Fire Officers of 1971,37 the Grant Program for 34 Act of Aug. 3, 1967, 1967 Ill. Laws 2644, 2646. 35 P.A. 86-1002, 1989 Ill. Laws 6809, 6820. 36 110 Ill. Comp. Stat. 947/10 (2018). 37 P.A. 77-704, 1971 Ill. Laws 1451. 62 Dependents of Correctional Officers of 1973,38 the Grant Program for a Person Raised by a Grandparent of 2006,39 the MAP Plus award of 2006,40 the Community College Transfer Grant of 2010,41 and the Grant Program for Medical Assistants in Training of 2015.42 However, in 1996, an exception was carved out so that incarcerated students could still get the Illinois National Guard Grant and Illinois Veteran Grant programs, though it is not clear why.43 Indiana. Indiana currently administers 14 grant programs, three of which affect justice- involved students. Around 2005, two of Indiana’s oldest need-based grants were effectively combined under the name Frank O’Bannon Grants to memorialize a former Indiana governor, though the two grants are still described separately in state laws.44 The Higher Education Award, created in 1965, is for students at public institutions, 45 while the Freedom of Choice Award, created in 1973, is for students at private institutions.46 In 1987, the chapter on the Higher Education Award was amended by Senate Bill 236 to say: The [Commission for Higher Education] may deny assistance under this chapter to a higher education award applicant or recipient who is: (1) convicted of a felony; (2) sentenced to a term of imprisonment for that felony; and (3) confined for that felony at a penal facility” (emphasis added).47 Despite granting permission to the commission to deny aid to incarcerated students, the state continued to award grants to incarcerated students, even after the rules for federal financial aid 38 P.A. 78-291, 1973 Ill. Laws 992. 39 P.A. 94-968, 2006 Ill. Laws 3508. 40 P.A. 94-1056, 2006 Ill. Laws 4954. 41 P.A. 96-1299, 2010 Ill. Laws 5277. 42 P.A. 99-0359, 2015 Ill. Laws 5372. 43 20 Ill. Reg. 9170, effective July 1, 1996. 44 Ind. Code 21-12-1-10. 45 Ind. Scholarships Act, P.L. 157-1965. 46 Ind. P.L. 229-1973. 47 Ind. P.L. 235-1987. 63 were changed in 1994 (personal communication, May 30, 2018). That changed with the passage of House Bill 1001, the biennial appropriations bill.48 In it, “may deny” was changed to “may not provide assistance to” in the Higher Education Award law, thus ending eligibility for incarcerated students.49 Though the law technically amended just the Higher Education Awards eligibility rules, because of the merger into the Frank O’Bannon Grant program, the Commission on Higher Education interprets this rule to apply to the Freedom of Choice Award (personal communication, May 30, 2018). Additional explanations for this change in the law are discussed in Phase II. The third affected program in Indiana is the 1990 Twenty-First Century Scholars program, created in 1990, which is an early-promise program that rewards graduating seniors with college scholarships.50 Students enroll in the program in seventh or eighth grade and pledge to graduate. Among the pledge requirements, students certify that they “will not use illegal drugs or alcohol or commit a crime or delinquent act” (Indiana CHE, 2017, p. 5). Students who do commit such an act would be ineligible for the scholarship. Despite one of the stated program goals being to “decrease drug and alcohol abuse by encouraging higher educational pursuits,” none of the published evaluations of the program have addressed the impact of this eligibility rule (see St John, Musoba, & Simmons, 2003; St John, Musoba, Simmons, & Chung, 2002; St. John, E. P., Musoba, G. D., Simmons, A., Chung, C. G., Schmit, J., & Peng, C. Y. J., 2004). Kentucky. Kentucky currently administers six grant programs, three of which affect justice involved students. In 1996, the state legislature passed a law to bar incarcerated students 48 Ind. P.L. 229-2011. 49 Id. 50 Ind. P.L. 56-1990; IC 21-12-6. 64 from receiving state aid. House Bill 285, sponsored by Representative John Stacy (D) and others, states: No loan guarantee shall be approved, nor any loan, grant, scholarship, recognition award, work-study reimbursement, or other form of financial aid shall be disbursed, under programs authorized by KRS 164.740 to 164.785, to any student who is an inmate or who is incarcerated in a penal institution unless funds have first been provided to all other eligible students who have applied in accordance with the administrative regulations of the authority.51 It was signed into law on April 4, 1996 by Governor Paul E. Patton. The programs affected by this law include the Go Higher Program (created in 2007 for students over the age of 24),52 the College Access Program (CAP) grant (created in 1994 for financially needy students),53 and the Kentucky Tuition Grant (created in 1972 for students a private institutions).54 Although these programs clearly fall within the state statute sections affected by the 1996 law (KRS 164.740 to 164.785), the regulations for CAP and Kentucky Tuition Grant do not mention incarceration as an eligibility rule.55 Curiously, when the Go Higher Grant program was created in 2007, the original rules stated that an eligible student shall “not have been convicted of a felony.”56 A year later, that rule was replaced with: shall “not be currently incarcerated,”57 and in October of 2008, it was amended again to state: “not be incarcerated at either the time of application or receipt of an 51 Ken. Rev. Stat. 164.767. 52 11 Ken. Admin. Reg 5:200. 53 K.R.S. 164.7535. 54 K.R.S. 164.780; 164.785. 55 11 KAR 5:033; 11 KAR 5:034. 56 33 Ky.R. 2226; eff. 4-6-2007. 57 34 Ky.R. 1761; eff.4-4-2008. 65 award,”58 as it remains today. There are no notes in the Administrative Register of Kentucky – where these amendments are recorded – that explain why the eligibility rule was changed three times. Presently, the Go Higher Grant application form still asks: “Have you ever been convicted of a felony?” and “If yes … are you currently incarcerated?” According to the rules, however, only a student who is incarcerated would be ineligible, and despite the superfluous felony question on the application, an official in the Kentucky Higher Education Assistance Agency confirmed that an affirmative response to the felony question, alone, would not render a student ineligible (personal communication, March 7, 2018). In 1998, the state legislature created the Kentucky Educational Excellence Scholarship (KEES), a merit-based scholarship for Kentucky residents who earn at least a 2.5 GPA. In the original statutory definitions, an eligible student is defined as any person who, among other qualifications, “is not a convicted felon.”59 Senate Bill 21 was sponsored by Senators Tim Shaughnessy, Lindy Casebier, David Karem, and Kim Nelson and was signed by Governor Paul Patton on April 4, 1998. More information on the 1996 incarceration bill and the 1998 KEES program are provided in Phase II. Louisiana. Louisiana currently administers six grant programs, four of which affect justice-involved students. The Taylor Opportunity Program for Students (TOPS) is a merit-based program with roots in the 1989 Louisiana College Tuition Program – an initiative of oil tycoon Patrick F. Taylor.60 Originally, the program had financial need and merit eligibility requirements, but the reconfiguration of 1997 removed the financial need components (Patrick F. Taylor Foundation, 2018). Now, it is comprised of four awards, three of which are original to the 1997 58 35 Ky.R. 949; eff. 1-5-2009. 59 K.R.S. 164.7874(8). 60 1989 La. Acts No. 789. 66 law61 plus the TOPS Tech award added the following year.62 The TOPS Honors Award goes to students with a minimum 3.0 GPA and 27 ACT score; TOPS Performance goes to students with a minimum 3.0 GPA and a 23 ACT score; TOPS Opportunity goes to students with a minimum 2.5 GPA and a 20 ACT; and TOPS Tech goes to students with a minimum 2.5 GPA and a 17 ACT who are enrolled in vocational or technical degree or certificate programs.63 Original to the 1989 law was a rule that a student “has no criminal conviction.”64 Senate Bill 280 was sponsored by Senator John Hainkel and others and was signed by Governor Buddy Roemer on July 10, 1989. In the 1997 law, however, the rule was amended to state: “a student shall not have any criminal conviction, except for misdemeanor traffic violations.”65 House Bill 2154 was sponsored by Representative Charles McDonald and others and was signed by Governor Mike Foster on July 11, 1997. Maryland. Maryland currently administers 11 grant programs, all of which are affected by a 1991 law requiring financial aid recipients to be drug-free.66 Amended only slightly in 2010,67 the statute now reads: “As a condition of receiving student financial assistance awarded by the Office [of Student Financial Assistance], each recipient shall sign a statement pledging to remain drug free. The Commission shall determine the contents of the drug free statement.”68 Indeed, students sign the pledge when they accept their financial aid packages, which states: As a condition of receiving a Maryland State scholarship or grant, I pledge to remain drug free for the full term of the award. Unlawful use of drugs and alcohol may endanger 61 1997 La. Acts No. 1375. 62 1998 La. Acts No. 165. 63 La. Admin. Code 28:701 et seq. (2018). 64 1989 La. Acts No. 789 § 1. 65 La. R.S. 17:5028 (2018). 66 Higher Education Scholarship Reform Act of 1991, Md. Acts 1991, ch. 462, § 2. 67 Md. Acts 2010, ch. 217. 68 Md. Ed. Code Ann. § 18-111 (2017). 67 my enrollment in a Maryland college as well as my Maryland financial aid award (personal communication, April 13, 2018). Michigan. Michigan currently administers six grant programs, four of which affect justice-involved students. The Michigan Competitive Scholarship of 1964,69 a need and merit- based award, and the Michigan Tuition Grant of 1966,70 a need-based award for students at private institutions, are both still funded programs available to Michigan students. Until 1981, both programs required students to be of “good moral character,” though it is not clear how this was defined or enforced in practice. However, in a series of 1980 bills sponsored by Senator Jerome Hart (D) and others, state lawmakers replaced the good moral character rules with a rule prohibiting incarcerated students from getting state aid. The bills were signed by Governor William Milliken on January 22, 1981. The incarceration rules remain in force today in the Michigan Competitive Scholarship71 and in the Michigan Tuition Grant program.72 A deeper historical analysis of the events that led to the incarceration amendments in these programs is provided in Phase II. Another active program in Michigan is the Tuition Incentive Program (TIP), which is a two-phase, last-dollar, need-based program for students who qualified for Medicaid coverage. Phase I is for students enrolled in certificate or associate’s degree programs, and Phase II is for students enrolled in a bachelor’s degree program. The program was created by the legislature in 1987 in an appropriations bill under the name “school completion program, 73 and the program was renamed TIP around 1991.74 69 Mich. 1964 P.A. 208. 70 Mich. 1966 P.A. 313. 71 Mich. Compiled Laws 390.974 (2017). 72 MCL 390.993 (2017). 73 Mich. 1987 P.A. 124 § 9. 74 See Mich. 1991 P.A. 110 § 406. 68 In 2010, the eligibility requirements were amended in the annual higher education appropriations bill to prohibit incarcerated students from receiving the grant,75 as still is the case today.76 According to bill analysis documents, the TIP program was expected to need an additional $6.2 million “due to continued increase in number of Medicaid-eligible students graduating from high school and enrolling in associate's degree programs” (Jen, 2010, p. 2). As such, Governor Jennifer Granholm proposed eliminating the Phase II awards and capped the awards under Phase I at a reduced rate (Jeffries, 2010). These changes, however, were not accepted by the House or Senate, and the only remaining substantive amendment was the ban on eligibility for incarcerated students. This suggests, though it is not explicitly stated in the analyses, that eliminating aid for incarcerated students was a cost-saving strategy. Additional research is needed to confirm this. Senate Bill 1157 was sponsored by Senator Tony Stamas and was signed by the governor on October 11, 2010.77 Administrative rules issued by the Michigan Department of Treasury’s Office of Scholarship and Grants reflected the legislative change in their 2010-2011 TIP Fact Sheet – published in October 2010 – which includes “must not be incarcerated” in the eligibility rules (Michigan Department of Treasury, 2010). The following year, in 2011, the TIP program was codified into law for the first time. 78 The fourth active Michigan program that affects justice-involved students is the oldest in the study sample: the Children of Veterans Tuition Grant.79 Created in 1935, administration of the program has transferred back and forth several times between state departments of education and veteran’s services. In 2005, the 1935 statute was repealed and replaced with a grant program 75 Mich. 2010 HB 1157 § 310. 76 MCL 388.1856 (2017). 77 Mich. 2010 P.A. 203 § 310. 78 MCL 388.1856 (2017). Previously, the funding and rules for the program were authorized in appropriations bills but were never codified in state statutes. 79 Mich. 1935 P.A. 245. 69 to be administered by Michigan’s higher education authority and to be funded by the general fund.80 Included in the statute, for the first time, were program eligibility rules, including that an applicant “has not been convicted of a felony involving an assault, physical injury, or death.”81 This language likely came from the text of the now-defunct Michigan Merit Award Scholarship of 1999, which had the same rule.82 Mississippi. Mississippi administers four grant programs, just one of which affects justice-involved students. What is now called the Higher Education Legislative Plan (HELP) for Needy Students Grant is a need- and merit-based program created in 1997.83 An original rule stipulates that applicants have “no criminal record, except for misdemeanor traffic violations.”84 Senate Bill 2231 was sponsored by Senator Tommy Gollott and others and was signed by Governor Kirk Fordice on March 18, 1997. However, according to a state agency official, there is no mechanism in place to screen applicants for criminal records, and the official was not aware of any situations where the agency had to make an eligibility decision for a student with a known criminal record (personal communication, March 1, 2018). Missouri. Missouri administers ten grant programs, five of which affect justice-involved students. In 1972, the state legislature created a need-based grant program for full-time college students.85 The section on eligibility (§173.215) included a rule stating: Has never been convicted in any court of an offense which involved the use of force, disruption or seizure of property under the control of any institution of higher education 80 Mich. 2005 P.A. 248. 81 MCL 390.1344 (2017). 82 Mich. 1999 P.A. 94. 83 1997 L. of Miss, ch. 381. 84 Miss. Code Ann. § 37-106-75 (2017). 85 L. of Mo., 1972 S.B. 613. 70 to prevent officials or students in such institutions from engaging in their duties or pursuing their studies.86 Programs created later made §173.215 the base eligibility rules, but it was repealed in 2007.87 Today, only the 1994 National Guard Member Educational Assistance Program88 maintains the campus disruption rule in its program regulations,89 a likely vestige of the old statutory langauge. The same 2007 law that repealed the old eligibility rules also created a new need-based program, the Access Missouri Financial Assistance Program.90 The new section on eligibility (§173.1104) included a rule that reflected the federal rule on certain drug convictions: If an applicant is found guilty of or pleads guilty to any criminal offense during the period of time in which the applicant is receiving financial assistance, such applicant shall not be eligible for renewal of such assistance, provided such offense would disqualify the applicant from receiving federal student aid under Title IV of the Higher Education Act of 1965, as amended. 91 However, according to a state official, this rule was the result of negotiations (personal communication, April 11, 2018). Indeed, in the introduced bill from January 2007, a proposed rule stated that an eligible student: “has not been found guilty of or pled guilty to any criminal offense.” In response: Higher education institutions and the Department of Higher Education warned legislators that the implementation and tracking of that requirement would be extremely onerous and could exclude some worthy students. Since the legislature was unwilling to completely 86 Mo. Rev. Stat. § 173.215.1 (Repealed L. 2007 S.B. 389 § A). 87 L. of Mo., 2007 S.B. 389 § A. 88 Mo. Rev. Stat. § 173.239. 89 Mo. Code Regs. tit. 11, § 10-3.015. 90 L. of Mo., 2007 S.B. 389 § A. 91 Mo. Rev. Stat. § 173.1104. 71 strike this requirement, the compromise of piggy-backing on the federal requirement was reached. (personal communication, April 11, 2018) In February 2007, the Senate Committee on Education filed its substitute bill, which replaced the first proposed rule on criminal convictions with the negotiated rule. In 2009, with §173.215 repealed, the program regulations for the 1986 Higher Education Academic Scholarship Program92 – a merit-based award now sometimes referred to as the Bright Flight Program – were updated to make the Access Missouri program’s new eligibility rules (§173.1104) the base requirements for it.93 Thus, students with certain drug convictions became ineligible for Bright Flight. In 2010, the statute was updated to reflect the adjustments made in the program regulations.94 In 1993, the A+ Schools Program was created, which funded grants to schools to encourage high student performance standards, including taking rigorous course work, high graduation rates, and high college-going rates.95 In addition, the program established a grant program for students graduating from A+-designated high schools to attend public community colleges or technical/vocational programs. From the original 1995 scholarship program regulations, a high school student must have “maintained a record of good citizenship and avoidance of the unlawful use of drugs.”96 By statute, the Department of Elementary and Secondary Education administered the A+ Scholarship Program until 2010, when Governor Jeremiah Nixon transferred it to the Department of Higher Education.97 In its first program regulations a year later, the Department of Higher Education left in the original eligibility rule 92 L. of Mo., 1986 S.C.S.H.B. 1356; Mo. Rev. Stat. § 173.250 93 Mo. Code Regs. tit. 6, § 10-2.080 (2009). 94 L. of Mo., 2010 S.B. 733. 95 L. of Mo., 1993 S.B. 380 § 14. 96 Mo. Code Regs, tit., 5, § 60-120.060 (1995). 97 Mo. Exec. Order 10-16 (Jan. 29, 2010). 72 regarding good citizenship and illegal drugs and added the federal financial aid rule about drug convictions, consistent with the Bright Flight and Access Missouri programs.98 The final affected Missouri program is the Advanced Placement Incentive Grant.99 Created in 2011, this program awards a one-time supplemental grant to students who earn high scores on STEM AP tests and who are awarded either the A+ Scholarship or the Access Missouri grant. As such, students must be eligible for those programs, which include the rules on drug convictions. New Jersey. New Jersey administers 10 grant programs, all of which are unavailable to incarcerated students. In 1999, the New Jersey legislature consolidated its financial aid program statutes and its three student financial aid agencies under a new chapter, creating the Higher Education Student Assistance Authority of today.100 It also created a new statute on student eligibility that applies to all the state’s programs, including a new rule: “A person who is incarcerated shall not be eligible for student financial aid under this chapter.”101 New Jersey may be the first state poised to remove eligibility rules for justice-involved students (O’Dea, 2018). In February 2018, Democratic Senators Sandra Cunningham and Teresa Ruiz introduced a bill to eliminate the word “not” from the rule, rendering incarcerated students eligible for state financial aid once again.102 The bill passed the Senate on June 25, 2018, and as of March 2019, the bill was still being considered by the General Assembly. New York. New York administers 17 grant programs, all of which are unavailable to incarcerated students. In 1995, the New York State Assembly passed a large state finance law 98 Mo. Code Regs, tit., 6, § 10-2.190 (2011). 99 L. of Mo., 2011 H.B. 223. 100 L. of N.J. 1999, c. 49, §1. 101 N.J. Stat. Ann. § 18A:71B-2. 102 N.J. State Leg., S-2055, Reg. Sess. 2018-2019 (2018). 73 that amended the New York Education Law by adding: “No student who is incarcerated in any federal, state or other penal institution shall be eligible for any general or academic performance award made pursuant to this article.”103 In 2011, the legislature added a somewhat redundant provision that states: “Tuition assistance program awards that are made available to students pursuant to paragraph b of this subdivision shall not be awarded if an applicant: … is incarcerated in any federal, state or other penal institution.”104 Both provisions appear in the Education Law section on general eligibility requirements.105 Ohio. Ohio administers five grant programs, all of which affect justice-involved students. The Ohio Instructional Grant was a need-based grant created in 1969.106 In 1981, a new eligibility rule was added: “An instructional grant shall be paid to an eligible student through the institution in which the student is enrolled, except that no instructional grant shall be paid to any person serving a term of imprisonment unless that person is eligible for parole within five years of making application for the grant” (amended language italicized).107 In 1997, the “unless” clause was deleted, leaving all incarcerated students ineligible for the grant.108 In 2005, the Ohio General Assembly phased out the Ohio Instructional Grant, creating in its place a nearly identical Ohio College Opportunity Grant.109 The new Ohio College Opportunity Grant maintains the eligibility rule that prevents all incarcerated students from receiving the award.110 In 2003, the Ohio legislature passed a law that created an eligibility statute that applies to all active state-funded scholarship programs.111 The law renders anyone convicted of the 103 1995 L. of NY, ch. 83 § 357. 104 2011 L. of NY, ch. 58, pt. Z, § 1. 105 NY Educ. Law § 661 (2018). 106 Am. Sub. H.B. No. 531, 1969-1970 Ohio Laws 90, 116-118. 107 Am. Sub. H.B. No. 694, 1981-1982 Ohio Laws 3709-3712. 108 Am. Sub. H.B. No. 215, 1997-1998 Ohio Laws 1244-1251. 109 Am. Sub. H.B. No. 66, 2005-2006 Ohio Laws, 3600-3615. 110 Ohio Rev. Code § 3333.122 (2018). 111 Am. Sub. H.B. No. 95, 2003-2004 Ohio Laws 1144-1145. 74 following crimes to be ineligible for state scholarships for two calendar years: riot (misdemeanor), aggravated riot (felony), failure to disperse (misdemeanor), and misconduct at emergency (misdemeanor). The convictions provisions apply to Ohio’s five currently-funded programs, including the Ohio College Opportunity Grant, Ohio War Orphans Scholarship, Choose Ohio First Scholarship, Ohio National Guard Scholarship, and Ohio Safety Officers Memorial Fund.112 According to an official from the Ohio Board of Regents, each institution verifies scholarship eligibility (personal communication, March 4, 2019), so it is not clear how the conviction eligibility rules are enforced. Oklahoma. Oklahoma administers eleven grant programs, two of which affect justice- involved students. The 1971 Oklahoma Higher Education Tuition Aid Act created a need-based grant program.113 An original rule stated that eligible students must demonstrate: “high moral character, good citizenship and dedication to American ideals.” Though this rule remains in the statute,114 it is not clear how this rule is interpreted, and an official from the Oklahoma State Regents for Higher Education was not able to offer an explanation (personal communication, March 23, 2018). Perhaps this is why, in 1995, a rule was added to the statute to clarify that an eligible student: “is not incarcerated in a state, federal or private correctional facility.” House Bill 1075 was sponsored by Representative Laura Boyd (D) and Senator Don Williams (D), though this new incarceration amendment was, apparently, subject to some negotiation.115 The original bill introduced on February 6 makes no mention of the incarceration rule, but the House engrossed bill states: “Persons incarcerated in state correctional facilities and persons who have been convicted of a felony offense are ineligibile [sic] to receive Oklahoma tuition aid grants.” It 112 Ohio Rev. Code § 3333.38 (2018). 113 1971 Okla. Sess. Laws, ch. 149. 114 70 Okla. Stat. 626.6 (2018). 115 1995 Okla. Sess. Laws, ch. 247. 75 is not clear from legislative records who inserted this language, but the engrossed bill was approved by the House on March 7 before it was read in the Senate the same day. The bill then bounced back and forth between the House and Senate for two months, though it is not clear from available records what issues were being debated. However, in the final bill, legislators agreed to remove the language about students with felony convictions, leaving only incarcerated students without aid. The final bill was signed by Governor Frank Keating on May 25, 1995. More information on this policy is provided in Phase II. Oklahoma’s Promise was created in 1992 as the Oklahoma Higher Learning Access Act.116 It is an early commitment program whereby financially-needy 8th, 9th, or 10th grade students pledge to attend school regularly, do their homework, and participate in program activities.117 Those who complete the program earn scholarships for college. According to the original statutes, students also pledge to “refrain from substance abuse” and “refrain from commission of crimes or delinquent acts.”118 High school program administrators are responsible for verifying the extent to which students complied with these and other requirements of the program, and according to a state official, some students are disqualified because of their conduct under these rules (personal communication, March 23, 2018). To clarify what is meant by these two behavioral requirements, the Oklahoma State Regents for Higher Education proposed new administrative rules in December 2017 that offer definitions. The yet uncodified administrative rules state: “Refrain from substance abuse, meaning the student was not adjudicated delinquent as a juvenile nor convicted of a crime as an adult by a court of law for an offense involving a controlled dangerous substance, as defined by 63 O.S. § 2-101(8)” and 116 1992 Okla. Sess. Laws, ch. 353. 117 Okla. Admin. Code § 610: 25-23-4 (2018). 118 Id. 76 “Refrain from commission of crimes or delinquent acts, meaning the student was not adjudicated delinquent as a juvenile nor convicted of a crime as an adult by a court of law” (italics indicate amended language).119 The new substance abuse rule thus means that only crimes will result in disqualification, as opposed to school discipline for drinking or drug use. Though incarcerated students would be disqualified from the program under the pledge requirements for having a criminal conviction, a rule explicitly barring incarcerated students from program eligibility was added to the program rules in 2003,120 stating: “any person incarcerated in a state, federal, or private correctional facility shall not be eligible to receive program benefits.”121 An official listed as the contact person on the amendments had no memory of the rationale for why the incarcerated rule was added (personal communication, April 23, 2018). Another official, however, remembered that around 2002, program administrators became aware that an incarcerated student was receiving an Oklahoma’s Promise award (personal communication, July 10, 2018). In an apparent loophole, the student was incarcerated for a crime committed after high school, after the student would have been determined to be eligible for the award. Thus, “the amendment to the OKPromise administrative rules was intended to make the OKPromise policies consistent with the statutory requirements of the state’s other primary need- based program and consistent with the objective of the OKPromise program to require students to refrain from criminal or delinquent acts” (personal communication, July 10, 2018). Oregon. Oregon administers six grant programs, just one of which affects justice- involved students. According to the Oregon Student Assistance Commission’s website, a student is not eligible for the Barber and Hairdresser Grant program if “you are in prison.”122 Although 119 Draft of proposed administrative rules obtained from Okla. St. Regents for Higher Educ. staff on March 23, 2018. 120 20 Ok. Reg. 2572, eff 7-11-03. 121 Okla. Admin. Code § 610: 25-23-5(f) (2018). 122 See https://oregonstudentaid.gov/barbers-grant.aspx 77 there are not state statutory or regulatory rules that state this,123 an official from the commission explained that there are no barber/hairdresser training programs operating in Oregon prisons that have agreements with the commission to accept this award (personal communication, October 10, 2017). Thus, practically speaking, no incarcerated student would be able to access this grant program, and the rule listed on the website is likely a vestige of old program rules (personal communication, July 23, 2018). Pennsylvania. Pennsylvania administers six grant programs, all of which affect justice- involved students. Pennsylvania’s path to its current interpretation of eligibility rules for incarcerated students emerged after nearly 25 years of legislation, administrative rule-making, and lawsuits. As such, a more detailed examination is offered in Phase II. In 1966, what is now called the State Grant Program was created by the state legislature. An original rule states that students must meet “qualifications of ‘financial need,’ character and academic promise, as well as academic achievement” (emphasis added).124 In 1969, the legislature added a new eligibility section to the law: (a) The agency may deny all forms of financial assistance to any student: (l) Who is convicted by any court of record of a criminal offense which was committed after the effective date of this act which, under the laws of the United States or Pennsylvania, would constitute a misdemeanor involving moral turpitude or a felony; or (2) Who has been expelled, dismissed or denied enrollment by an approved institution of higher learning for refusal to obey, after the effective date of this act, a lawful regulation or order of any institution of higher education, which refusal, in the opinion of the 123 Or. Admin. Regs. 575-035-0030; Or. Rev. Stat. § 348.290 (2017). 124 Pa. Act of Jan. 25, 1966 (P.L. 1546, No. 541) § 4. 78 institution, contributed to a disruption of the activities, administration or classes of such institution; or (3) Who has been convicted in any court of record, of any offense committed in the course of disturbing, interfering with or preventing, or in an attempt to disturb, interfere with or prevent the orderly conduct of the activities, administration or classes of an institution of higher education.125 These new rules, however, would be the subject of three lawsuits and regular tinkering by the Pennsylvania Higher Education Assistance Agency (PHEAA) in the 1970s and 1980s. In short, PHEAA in 1988 decided to deny grant aid only to incarcerated students as a legally-defensible compromise. Ever since, the agency has applied this rule to all its other programs for consistency, including the Blind or Deaf Higher Education Beneficiary Grant Program,126 the Children of POW/MIA Soldiers program,127 the PATH (Partnerships for Access to Higher Education) Program,128 the Pennsylvania Targeted Industry Program (PA-TIP),129 and the Ready to Succeed Scholarship (RTSS) Program.130 Like in Georgia, the rules were added more recently to the program manuals and applications over time to achieve consistency, though officials at PHEAA claim it was always the case that incarcerated students could not get aid (personal communication, June 25, 2018). The incarceration rule was first added to the 2015-2016 PA-TIP application and the 2016-2017 PATH manual. The rule appears in the 2018 Blind or Deaf grant program manual, but it is not clear from available records when it first appeared there. Though 125 Pa. Act of Dec. 18, 1969 (P.L. 171, No. 169) § 2. 126 Pa. Act of March 10, 1949 (P.L. 30, No. 14) § 1381. 127 Pa. Act of October 11, 1972 (P.L. 1397, No. 213). 128 Pa. Act of Dec. 20, 1989 (P.L. 689, No. 80). 129 Pa. Act of June 30, 2012 (P.L. 684, No. 82) § 32. 130 Pa. Act of Jul. 10, 2014 (P.L. 1053, No. 126) § 15. 79 the rule is not listed on the POW/MIA grant application, it is most likely that incarcerated students cannot receive the award. South Carolina. South Carolina administers 11 grant programs, eight of which affect justice-involved students. The South Carolina Legislature created these programs over a nearly 40-year period, all the while tinkering with the eligibility rules for justice-involved students. In 1970, the Higher Education Tuition Grant program was created for students at private institutions. An original rule states that applicants must be “of good moral character.”131 It is not clear what this originally meant, and according to an official from the state agency that administers this program, the good moral character rule is not interpreted and enforced today (personal communication, April 12, 2018). In 1988, the Palmetto Fellows Scholarship Program was created as a merit-aid program.132 There were no rules about character or criminal history in the original law. In 1996, the South Carolina Need-based Grant program was created with a rule that stated students must be “of good moral character and has never been convicted of a felony.”133 It was created within a general appropriations bill (H.B. 4600) that was sponsored by the House Ways and Means Committee. In 1998, the merit-based Legislative Incentives for Future Excellence (LIFE) Scholarship was created, also by the House Ways and Means Committee (H.B. 4535). An original eligibility rule states: Students must not have been adjudicated delinquent or been convicted or pled guilty or nolo contendere to any felonies or any alcohol or drug-related offenses under the laws of 131 S.C. 1970 Act No. 1191. 132 S.C. 1988 Act No. 629, ch. 104. 133 S.C. 1996 Act No. 458, pt. II, sec. 20(A). 80 this or any other state or under the laws of the United States in order to be eligible for a LIFE Scholarship.134 This rule, however, was amended several times. A year later, the rule was amended to reinstate eligibility for students convicted of misdemeanor alcohol or drug offenses after one year, again in a general appropriations bill (H.B. 3696): Students must not have been adjudicated delinquent or been convicted or pled guilty or nolo contendere to any felonies or any alcohol or drug related offenses under the laws of this or any other state or under the laws of the United States in order to be eligible for a LIFE Scholarship, except that a student who has been adjudicated delinquent or has been convicted or pled guilty or nolo contendere to an alcohol or drug related misdemeanor offense is ineligible only for one calendar year after the adjudication, conviction, or plea occurred.135 State legislators tinkered with the language of this rule again in 2000; although it did not substantially change the rule, the bill inserted this second iteration of the rule into the Need- based Grant, Palmetto Fellows, and the Higher Education Tuition Grant for the first time.136 House Bill 4650 was sponsored by Representative Glenn Hamilton and others and was signed by Governor Jim Hodges on May 19, 2000. In a final adjustment in 2007, the legislature reinstated aid for students with a first misdemeanor alcohol or drug offense; students with two or more alcohol or drug-related misdemeanors lost eligibility for one year.137 This 2007 law updated all four programs with the new language. Senate Bill 213, which also amended many other laws related to penalties for 134 S.C. 1998 Act No. 418 § 59-149-90. 135 S.C. 1999 Act No. 100, pt. II, sec. 73(B). 136 S.C. 2000 Act No. 289 § 2. 137 S.C. 2007 Act No. 103 § 21. 81 alcohol/drug crimes, was sponsored by Senator Joel Lourie and others and was signed by Governor Mark Sanford on June 15, 2007. In 2001, the state legislature created a lottery system and a new, lottery-funded HOPE Scholarship.138 Eligibility for HOPE was tied to the same requirements for LIFE, including the criminal history provisions.139 In 2007, “enhancement” awards were created for students who earned LIFE or Palmetto Fellows scholarships.140 Under these awards, students pursuing STEM degrees can qualify for an additional stipend beginning in their second year. Eligibility for the enhancement is based on the initial requirements, including the criminal history provisions. Finally, the Dayco Products Inc., Scholarship was created in 1988 when the company donated funds to the state of South Carolina’s scholarship fund (S.C. Commission on Higher Education, n.d.). The funds were matched by the state, and now the Commission on Higher Education administers the scholarship program from the interest on the fund to employees or dependents of employees of Dayco Products. Though this program does not exist in statute, CHE issues program guidelines that generally match the basic eligibility requirements of the other state grant programs. A rule in the current guidelines resembles the 1999 criminal history rule, which states a student with a felony is ineligible and a student with an alcohol or drug-related misdemeanor will lose aid for one year. This suggests this rule was added in 1999, before the 2000 or 2007 language changes were made (S.C. Commission on Higher Education, n.d.). Tennessee. Tennessee currently administers 13 grant programs, all of which affect justice-involved students. In 1976, the Tennessee Student Assistance Award was created as a 138 S.C. Education Lottery Act, S.C. 2001 Act No. 59 139 S.C. Code Ann. § 59-150-370(C) (2018). 140 S.C. 2007 Act No. 115 § 7. 82 need-based award after a 1971 financial aid program was struck down by the courts.141 In the original program rules for the new program promulgated by the Tennessee Student Assistance Corporation, students were considered eligible when: He is a person of good moral character. An application shall be considered a person of good moral character if he has been accepted for admission or is enrolled in a Tennessee college or university and has not been convicted of a felony during the twelve-month period preceding the date of his application for a tuition grant.142 However, this rule did not last long. In 1980, the corporation deleted the entire program rules and replaced it with new ones, which made no mention of a student’s character or criminal history.143 In wasn’t until the early 1990s that the current configuration of rules for justice-involved students were adopted. In 1990, a rule referencing the federal Anti-Drug Abuse Act of 1988144 was added requiring higher education institutions to certify of award recipients: “that the student has on file with the institution a statement for the periods of instruction beginning on or after July 1, 1989 certifying that the student is in compliance with the Anti-Drug Abuse Act.”145 In short, this means that no eligible student could have been convicted of any drug crime. Then, in 1992, the Tennessee legislature passed Senate Bill 1815, amending the statutory eligibility requirements to say that the corporation could only make awards students who “are not incarcerated.”146 The bill was sponsored by a Senator Montgomery, Representative Richard Venerable, and Representative Ken Givens and was signed by Governor Ned McWherter on April 28, 1992. Iterations of the drug-free and incarceration rules would be added to the program 141 Ten. Acts 1976, ch. 415. 142 2 Ten. Admin. Reg. (Feb. 1976), Repeal and new rule filed July 6, 1976; effective Aug. 5, 1976, pp. 22-27. 143 6 Ten. Admin. Reg. (Nov. 1980), amend. filed Oct. 31, 1980; effective Jan. 28, 1981, pp. 522-528. 144 Anti-Drug Abuse Act of 1988, Pub. L. No. 100-690, 102 stat. 4181. 145 16 Ten. Admin. Reg. (Feb. 1990), amen. filed Feb. 9, 1990; effective May 29, 1990, pp. 18-20. 146 Ten. Acts 1992, ch. 775 § 1; Ten. Ann. Code § 49-4-301 (2017). 83 regulations for all of Tennessee’s awards, including those that existed before 1990147 and those created after.148 Created in 2017, the Tennessee Reconnect program is one of the programs affected by these rules for incarcerated students and students with drug convictions. 149 It extends the existing Tennessee Promise program by making college free for adult, independent students who are pursuing a two-year degree or certificate. In its first year, the program gained much attention for the nearly 31,000 students who applied for the grant, far exceeding projections (Gonzales, 2018). In January 2019, Tennessee General Assembly Representative Barbara Ward Cooper (D) introduced a bill that would allow incarcerated students to receive the Tennessee Reconnect (only) making it just the second state to consider a reversal on the ban of state aid to incarcerated students, following New Jersey.150 The bill was under review in the House and Senate education committees as of March 2019. Texas. Texas currently administers six grant programs, two of which affect justice- involved students. Created in 1999, the Toward EXcellence, Access & Success (TEXAS) Grant Program is a need-based grant program.151 Original to the law is an eligibility rule regarding students convicted of drug crimes: (b) A person is not eligible to receive a TEXAS grant if the person has been convicted of a felony or an offense under Chapter 481, Health and Safety Code (Texas Controlled Substances Act), or under the law of another jurisdiction involving a controlled substance 147 Christa McAuliffe Scholarship. 148 Helping Heroes Grant, Tennessee HOPE Access Grant, Wilder-Naifeh Technical Skills Grant, Tennessee HOPE Scholarship, General Assembly Merit Scholarship, Aspire Award, Tennessee HOPE Scholarship Nontraditional, Tennessee HOPE Foster Child Tuition Grant, Tennessee Reconnect Grant, Tennessee STEP UP Scholarship, Tennessee Promise Scholarship. 149 Ten. Acts 2017, ch. 448. 150 Ten. HB 30, 2019. 151 19 Tex. Admin. Code § 22.230 (2018) (Tex. High. Educ. Coord. Bd., Discontinuation of Eligibility or Non- Eligibility). 84 as defined by Chapter 481, Health and Safety Code, unless the person has met the other applicable eligibility requirements under this subchapter and has: (1) received a certificate of discharge by the Texas Department of Criminal Justice or a correctional facility or completed a period of probation ordered by a court, and at least two years have elapsed from the date of the receipt or completion; or (2) been pardoned, had the record of the offense expunged from the person's record, or otherwise has been released from the resulting ineligibility to receive a grant under this subchapter.152 Earlier versions of the Senate bill would have also made students convicted of “crimes of moral turpitude” ineligible, but this provision was not included in the final bill.153 The bill was authored by Representative Henry Cuellar (D) and was signed by Governor George W. Bush on June 19, 1999. Two years later, the TEXAS Grant II program was created for students at public two-year colleges.154 The program contains the same eligibility rule regarding students convicted of drug crimes as the TEXAS grant. Utah. Utah currently administers seven grant programs, two of which affect justice- involved students. When the Regents’ Scholarship Program155 – a merit-based award – was created in 2008, it included a stipulation that eligible applicants “may not have a criminal record, with the exception of misdemeanor traffic citations.”156 The program was sponsored by Senator Lyle Hillyard (R) and was signed by Governor Jon Huntsman, Jr., on March 17, 2008. 152 Act of June 19, 1999, Tex. 76th Leg. R.S., ch. 1590 § 56.304. 153 Conference Committee Report, May 199, retrieved from http://www.lrl.state.tx.us/LASDOCS/76R/HB713/HB713_76R.pdf 154 Act of June 11, 2001, Tex. 77th Leg. R.S., ch. 624. Renamed Texas Educational Opportunity Grant Program (TEOG) in 2005, see Act of June 18, 2005, Tex. 79th Leg. R.S., ch. 1181 § 39. 155 Utah Code Ann. § 53B-8-2 (2018). 156 Ch. 271 § 3, L. Utah 2008. 85 Two years later, the 1999 New Century Scholarship157 – a merit-based award for students who completed college coursework during high school and who enroll in college after graduating – was amended to include the same eligibility rule regarding criminal records.158 The bill was sponsored by Senator John Valentine (R) and was signed by Governor Gary Herbert on March 29, 2010. One high-ranking official involved in financial aid policy in Utah could not remember why the criminal history rule was included in the 2008 Regents’ Scholarship but suspected the 2010 New Century amendment was to make the two programs consistent on eligibility rules (personal communication, July 5, 2018). Another official recollected that legislators researched other state scholarship programs and found many had similar requirements. Additionally, the rules for justice-involved students for federal financial aid programs may have also been a motivating factor (personal communication, July 6, 2018). More information is provided in Phase II. Washington. Washington administers six grant programs, one of which affects justice- involved students. Created in 2007, the College Bound Scholarship is an early commitment program where needy 7th and 8th graders pledge to graduate from high school and to not be convicted of a felony.159 After graduation, successful students are awarded grants for college. Senate Bill 5098 was sponsored by Senators Phil Rockefeller (D), Karen Keiser (D), and others and was signed by Governor Christine Gregroire on May 9, 2007.160 157 Utah Code Ann. § 53B-8-1-105 (2018). 158 Ch. 270 § 1, L. Utah 2010. 159 Rev. Code Wash. § 28B.118.010 (2018). 160 Wash. Laws of 2007, ch. 405. 86 In 2014, the Washington legislature passed a law to create a workgroup to evaluate the College Bound Program.161 Resulting from this workgroup was a 2015 law that required the state agency to: “Develop and implement a student application, selection, and notification process for scholarships, which includes working with other state agencies, law enforcement, or the court system to verify that eligible students do not have felony convictions” (amendments italicized).162 Where eligibility related to felony convictions was initially verified through self-reporting, this new legislation created a process of checking all applicants against public records searches and Washington State Patrol records (personal communication, March 26, 2018). An evaluation of this process in 2016 found that significant resources were spent checking 14,000 applicants to find only one student who had been convicted of a felony, which called into question the efficacy of the new process (Washington Student Achievement Council, 2017). According to an official in the Washington Student Achievement Council in March 2018, only two students in the program’s history were denied the scholarship because of their felony convictions, though they had other disqualifying characteristics as well (personal communication, March 26, 2018). West Virginia. West Virginia administers three grant programs, one of which affects justice-involved students. In 1999, the legislature created the Higher Education Adult Part-time Student (HEAPS) grant program for financially needy undergraduate students attending college part-time.163 An original rule states that an eligible student “is not incarcerated in a correctional facility.”164 House Bill 2482 was sponsored by Delegate Jerry Mezzatesta (D) and others and was signed by Governor Cecil Underwood on April 7, 1999. 161 Wash. Laws of 2014, ch. 215. 162 Wash. Laws of 2015, ch. 244 § 4(3). 163 W. Va. Acts, 1999 R.S., ch. 116. 164 W. Va. Code § 18C-5-7(c)(5) (2017). 87 Wyoming. Wyoming administers seven grant programs, five of which affect justice- involved students. In 2005, the legislature created an endowed fund for higher education scholarships,165 and in 2006, the legislature created the Hathaway Scholarship Program to draw from that fund.166 The Hathaway Scholarship Program is comprised of four merit-based awards and one need-based award. The Honors, Performance, Opportunity, and Provisional awards each have varying high school curriculum, GPA, and ACT score requirements, with the Honors award having the highest level of requirements and the highest award amount. The need-based scholarship is a supplement for merit-award students with unmet financial need.167 For all five Hathaway awards, there are two eligibility rules affecting justice-involved students, both original to the authorizing statute. A student is not eligible if he/she: “is incarcerated” or “has been convicted of a felony in this state or another jurisdiction and has not been granted an exception by the department of education...”168 The first version of Senate File 85 introduced on February 15, 2006 included the incarceration rule, but the felony convictions rule amendment was later sponsored by the House Committee on Education and was approved by the whole House on March 2, 2006. There are no available legislative records that explain why this rule was added. The Joint Education Interim Committee (comprised of Senator Hank Coe (R), Representative Jeff Wasserburger (R), and others) sponsored the bill, and it was signed by Governor Dave Freudenthal on March 10, 2006. More information on the Hathaway Scholarship is provided in Phase II. 165 2005 Sess. Laws of Wyo., ch. 190. 166 2006 Sess. Laws of Wyo., ch. 36. 167 Wyo. Stat. § 21-16-1306 (2018). 168 Wyo. Stat. § 21-16-1303(e) (2018). 88 Policy Adoption Events The timing of the policy adoption events deserves a closer look. Reading the written state profiles alone somewhat obscures the trendline of when state policymakers were implementing bans on aid to justice-involved students. From the full list of 131 programs, Table 4 presents the primary policy adoption events that signify when a new program was created with rules for justice-involved students for the first time or when new rules were created that applied to all programs for the first time. Though the event dates range from 1969 to 2011, examining the event date column of Table 4 illuminates a trend: most of the sweeping policy changes affecting justice-involved students occurred in the 11-year period between 1989 and 2000. Notably, incarcerated students lost aid for some or all programs during this period in Illinois, Tennessee, Georgia, New York, Oklahoma, Kentucky, New Jersey, West Virginia, and California. Table 4 also shows the fervor of anti-drug policies that sprung onto the policy scene beginning in 1990. As likely reflections of federal War-on-Drug policies,169 eight states implemented drug-free rules into some or all financial aid programs during the 1990s. The policy events timeline also reveals geographical patterns. The earliest actions, between 1969 and 1990, occurred mostly in a cluster of bordering Great Lakes states, including Pennsylvania, Ohio, Indiana, Michigan, and Illinois. In the 1990s, mostly Southern states adopted policies, and after 1990, states in the West adopted policies. 169 See Anti-Drug Abuse Act of 1988, Pub. L. No. 100-690, 102 stat. 4181; Drug-Free Schools and Communities Act of 1989, 20 U.S.C. § 1011i. 89 Table 4: Primary State Policy Adoption Events affecting Justice-Involved Students Policy Pennsylvania State Grant Program amended Competitive Scholarship & Tuition Grant amended Ohio College Opportunity Grant amended Oregon Barber and Hairdresser Grant created Pennsylvania program rules amended Incarcerated students in Illinois lose aid TOPS Awards created Georgia Drug-Free Postsecondary Education Act 21st Century Scholarship created Drug-Free rules created Academic Challenge Scholarship created Maryland Drug-Free Pledge law Incarcerated students in Tennessee lose aid Oklahoma's Promise created A+ Scholarship Program created HOPE Programs amended Incarcerated students in New York lose aid Oklahoma Tuition Aid Grant Program amended Incarcerated students in Kentucky lose aid South Carolina Need-Based Grant Program created Florida Bright Futures Programs created Higher Education Legislative Plan Grant created Kentucky Educational Excellence Scholarship created Incarcerated students in New Jersey lose aid Excluded Population State Criminal Convictions PA MI Incarcerated OH Incarcerated Incarcerated OR PA Incarcerated (only) Incarcerated IL Criminal Convictions LA Drug Convictions Convic. + DF Pledge Federal DF Rules GA IN TN AR Drug-Free Pledge MD Drug-Free Pledge Incarcerated TN Convic. + DF Pledge OK Drugs + Fed. Drugs MO GA Incarcerated Incarcerated NY OK Incarcerated Incarcerated KY Criminal Convictions SC Criminal Convictions FL Criminal Convictions MS Criminal Convictions KY Incarcerated NJ Toward Excellence Access and Success Grant created Higher Education Adult Part-Time Student Grant created Cal Grant Programs created Ohio criminal convictions rule created TX Drug Convictions WV Incarcerated Incarcerated CA Criminal Convictions OH Delaware Student Excellence Equals Degree Act created Children of Veterans Tuition Grant amended Hathaway Scholarship Program Awards created Criminal Convictions DE Criminal Convictions MI Incar. + Convictions. WY College Bound Scholarship created Access Missouri Financial Assistance Program created Regent's Scholarship created Criminal Convictions WA Federal Drug-Free MO Criminal Convictions UT Academic Challenge Scholarship amended Tuition Incentive Program amended Tuition Equalization Program amended Frank O'Bannon Grants amended Incarcerated Incarcerated Incarcerated Incarcerated AR MI GA IN Year 1969 1981 1981 1985 1988 1989 1989 1990 1990 1990 1991 1991 1992 1992 1993 1995 1995 1995 1996 1996 1997 1997 1998 1999 1999 1999 2000 2003 2005 2005 2006 2007 2007 2008 2009 2010 2011 2011 Notes: DF Pledge = drug-free pledge; Fed. Drugs = federal financial aid drug convictions rule; Convic. = criminal conviction rule. 90 Discussion Today, justice-involved students in 26 states are ineligible for at least 131 grant programs. Since the 1960s, state policymakers have targeted justice-involved students for denial of state financial aid. Early and ambiguous “good moral character” rules evolved into explicit bans on aid for students in prison or students convicted of crimes. Consistent with the wave of penal populism at the federal level during the 1980s and 1990s (Page, 2004), this period saw the most states slashing the financial aid eligibility of justice-involved students. In the decade between 1989 and 2000, at least nine states cut funding to incarcerated students, and thirteen cut aid to other justice-involved populations. Additional research is needed to determine if the punitive attitudes of the time directly influenced these policy changes (a pursuit of Phase II), but the policy actions clearly fit within the contemporary policy landscape. Though most of the rules were adopted in the 1980s and 1990s, states as recently as 2010 and 2011 were amending state laws and administrative rules to ban incarcerated students from aid. Wide-reaching actions to ban justice-involved students from all state aid programs have slowed, the last being New Jersey in 1999, suggesting that such actions are no longer in vogue. In fact, New Jersey legislators in the 2018-19 legislative session may be the first to buck the trend by removing such rules. Across the country, few obvious geographic policy patterns are observable from the descriptive data, with justice-involved students affected in every corner of the country, perhaps with a block of central Western states being the general exception. Southern states tended to target students convicted of drug offenses, but states across the country targeted students with criminal convictions (generally) and incarcerated students. Southern states also appear to be more punitive in that more types of justice-involved students are ineligible for more programs, 91 compared to other regions of the country. For example, all incarcerated students plus students with other types of convictions are ineligible for most state aid programs in Georgia, Tennessee, and Kentucky. For students in prison or students with criminal convictions in their past, these findings paint a grim picture for the condition of rehabilitation programming in America. At the same time jails and prisons were filling up, state policymakers eliminated funding opportunities for higher learning, self-improvement, and vocational training through higher education. Even as recent federal and state criminal justice policy reforms have begun to slow the rate of mass incarceration, and as advocates continue to push for the restoration of federal Pell Grants for prisoners, there appears to be little appetite for reinstating state aid to justice-involved students, perhaps except for in New Jersey and Tennessee. Implications Results of Phase I have implications for future research, student services professionals, state policymaking, theory, and for the Phase II study. An Evaluation Agenda Arguably, the most important questions that this descriptive analysis cannot answer relate to the impact of these policies. None of the state financial aid agency officials that I contacted reported ever conducting analyses on the impact of the eligibility rules for justice-involved students.170 In fact, many states reported that they did not track cases when a student applied but was denied because of criminal history, and no one can know how many justice-involved students might have applied for aid had there not been disqualifying eligibility rules. The 170 Only Washington state officials engaged in an implementation evaluation to understand if their existing process to identify students with convictions was accurate. After running 14,000 student applicants through a new background check process, they found just one student who had a felony conviction, concluding that this process was expensive and yielded little benefit to the program (Washington Student Achievement Council, 2017). 92 following is an evaluation agenda for understanding the impact and cost/benefit of denying state financial aid to justice-involved students. Regarding financial impact, it may be possible to estimate the number of individuals in a state who would qualify for state aid if not for the incarceration eligibility rule (see Oakford et al., 2019). For example, it may be possible to collect personal information (i.e., academic, financial, educational aspirations) from incarcerated persons in a given state. Of those who would otherwise qualify for a grant, and of those who might aspire to attend college if possible, one could calculate the additional state funding needed to cover grant aid for these incarcerated students. In some states, given population size and availability of higher education in prisons, perhaps the overall added expenditure would be low and therefore feasible for states. In other states, like California, the large prison population and strong higher education programs would likely mean that many more individuals would qualify for the Cal Grant programs, requiring a substantial increase in funding for the programs. Any advocate who lobbies for reinstating aid eligibility to incarcerated students must first estimate the costs. Another important impact question is one of social impact. Had more justice-involved students been eligible for state grants, how would their lives be different now? Would they have been more likely to enroll in or complete college? Would they be more likely to get a stable job? Would they be more likely to desist from subsequent crime? Would they be more likely to contribute more to the economy? Most would agree these are desirable outcomes for justice- involved people, but only additional research could estimate the extent to which changing the eligibility rules in state financial aid grant programs might yield such results. If New Jersey succeeds in dropping its ban on state financial aid for incarcerated students, it may present an 93 opportunity to explore questions of social impact by comparing student outcomes from before and after the policy change. Finally, there is the question of behavioral impact. Several of the “promise” or “early commitment” programs explicitly or implicitly suggest that state-funded scholarships are intended to incentivize good behavior, including abstinence from illegal alcohol, drugs, and other crimes. Such programs include: Indiana’s 21st Century Scholarship, Delaware’s SEED, Washington’s College Bound Scholarship, Wyoming’s Hathaway Scholarship, Utah’s New Century Scholarship, and others. Whether these incentives were effective in curbing juvenile criminal behaviors has yet to be evaluated. Whatever the effects of these policies are determined to have on justice-involved students – be it financial, social or behavioral – the last evaluation task is to compare those benefits with the costs. In states that do subsidize justice-involved students’ college, do the benefits outweigh the costs of the aid? In states that do not, would such aid be a cost-effective investment in a state’s citizenry? These questions are complex and require sophisticated methodology, but their answers constitute critical information that state policymakers have never had since they began denying aid to justice-involved students some fifty years ago. Seeing and Supporting Justice-Involved Students At the 2018 meeting of the Association for the Study of Higher Education, Lee and House (2018) presented an impact analysis of Tennessee’s primary need-based grant program, the Tennessee Student Assistance Award. Lee shared background on the program, but in his explanation of the eligibility requirements, he neglected to disclose that incarcerated students and students who are ineligible for federal aid under the drug-conviction rule are ineligible for the award. Similarly, hundreds of articles have been written about state financial aid programs, 94 almost none of which mention eligibility requirements for justice-involved students. These oversights are emblematic of the historical tendency to ignore justice-involved students in higher education. For the first time, this study identified all the programs where justice-involved students are excluded from eligibility. Importantly, it also identified who was being denied aid. This is the first step in seeing justice-involved students. Anyone who studies state financial aid must acknowledge who can and cannot participate in the programs, rather than treating justice- involved students like a footnote or ignoring them altogether. In addition to researchers, state financial aid agency officials need to acknowledge when justice-involved students are denied aid. I found many examples of state financial aid agencies that did not list complete eligibility requirements on official websites, though I did not a conduct a systematic analysis of this. For example, the Tennessee Higher Education Commission’s website for the Tennessee Higher Education Award does not state that incarcerated students are ineligible for the award in a list of eligibility requirements, and nowhere on New Jersey’s Higher Education Student Assistance Authority websites or brochures is any mention of the total ban on aid to incarcerated students. The absence of such information is imprecise and likely leads to justice-involved students being misinformed about their eligibility. State financial aid agencies must rectify this by providing complete explanations on their websites and marketing materials of any conditions that might affect justice-involved students, including available procedures for appealing denials or for seeking special considerations. Given how the rules affecting justice-involved students are so rarely mentioned in research and on state websites, it stands to reason that financial aid professionals at individual institutions may also lack awareness of these rules. Future research might assess financial aid 95 officers’ knowledge of rules for justice-involved students in federal and state aid programs. Regardless, campus professionals could likely benefit from having more knowledge about how federal and state financial aid rules affect justice-involved students. In that way, the results of this study can build awareness of how justice-involved students may be excluded from state financial aid. Once justice-involved students are seen by more institutions and individuals, the next step is to support them. In states where there are no restrictions on aid eligibility, justice- involved students should be highly encouraged to seek state financial aid. Considering that the average incarcerated person in the U.S. has completed less education, has lower literacy and numeracy skills, and less paid work experience than the average person (Rampey et al., 2016), state grant programs could be critical resources for accessing higher education, especially after incarceration. In states where justice-involved students are denied aid, additional financial support may be needed. Financial aid professionals should be knowledgeable about other scholarships or funding opportunities for students when they are ineligible for federal or state aid. For example, Corrections to College California is an initiative that provides $500 scholarships specifically to formerly-incarcerated college students to cover the costs of attending conferences or events (Corrections to College California, 2018). Going even further, some institutions with particularly large justice-involved populations may benefit from creating privately-funded scholarships specifically for justice-involved students.171 171 To my knowledge, there are no available estimates of how many justice-involved students are enrolled in higher education. It stands to reason that some colleges have higher populations of such students; for example, urban colleges that do not have admissions restrictions on applicant’s with criminal histories. 96 An Advocacy Agenda For advocates of justice-involved students in higher education, Table 4 presents an advocacy agenda for re-enfranchisement. Each policy action whereby justice-involved students became disenfranchised from aid eligibility can be reversed. Except for the pending bill in the New Jersey legislature that would reverse the state’s ban on aid to incarcerated students (O’Dea, 2018) and the bill in Tennessee that would reinstate the new Reconnect grant for incarcerated students, I am not aware of any other substantial effort to reverse these bans in other states. While much attention continues to be paid to Pell Grants for incarcerated students, advocates have much to gain by focusing efforts on state policy. State-level criminal justice reform is experiencing a wave of bipartisan support and momentum right now (American Civil Liberties Union, 2012; Porter, 2018), which could present an opportunity to restore scholarship eligibility to justice-involved students. In addition, the findings in Phase I can also support advocacy down to the individual student level. Many organizations provide advisors, counselors, mentors, lawyers, and the like to support to justice-involved people, and these helpers probably commonly get questions about going to college. Justice-involved students and their advocates now have access to these data showing where justice-involved students can and cannot get state financial aid. Having this information is important when advising students where to attend college and how to pay for it. Inclusive Policymaking In addition to reinstating aid eligibility in existing programs, states also have the option to create scholarship programs specifically for justice-involved students. I next highlight two such programs that I discovered during the policy review. These programs may be good models for states that are considering how to expand higher education access to justice-involved students. 97 Wisconsin. Since at least the mid-1970s, Wisconsin has provided scholarships to “uniquely needy students,” later called the Talent Incentive Program (TIP) Grant.172 As of 1978, one of the 13 listed qualifying characteristics of a uniquely needy student was “the student has a correctional history.”173 In 1990, this provision was redefined as “the student is currently or was formerly incarcerated in a correctional institution,” as it remains today.174 This may be the only state grant program in the country that explicitly recognizes the need to support justice-involved students. Unfortunately, an official in Wisconsin’s Higher Educational Aids Board (HEAB) reported that they do not track how students qualify for TIP, meaning it is not known how many incarcerated or formerly-incarcerated students have received TIP (Personal communication, June 11, 2018). Another signal of Wisconsin’s unique support of justice-involved students is a statement in the HEAB’s Policies and Procedures Manual: “There is nothing in current state statutes or current state administrative code that would prohibit students who [have a drug-related conviction or are incarcerated] from receiving state aid” (Wisconsin Higher Educational Aids Board, 2018, n.p.). Though more transparency is needed to understand the extent to which justice-involved students are accessing Wisconsin’s programs, the TIP grant and the Wisconsin program manual are nonetheless models of how to explicitly recognize justice-involved students. Exonerees. The term “justice-involved student” may include college students who were convicted but later exonerated of their crimes. When people are wrongfully convicted of crimes, states increasingly offer compensation to the exonerees. As of 2012, one researcher identified compensation policies in 27 states, including eight states that provide educational benefits to 172 Wis. Admin. Code § HEA 5.04. 173 Wis. Admin. Reg., March, 1978, No. 267, eff. 4-1-78. 174 Wis. Admin. Reg., Feb., 1990, No. 410, eff. 3-1-90. 98 exonerated students (Norris, 2012). Today, at least 33 states have compensation policies for the wrongfully convicted (Innocence Project, 2017). Table 5 displays at least some of the known state financial aid programs that serve as compensation for wrongfully-convicted people. The policies in Table 5 provide models for states that seek to support justice-involved students who were exonerated. Additional research is needed to track how many exonerees have successfully accessed their higher education benefits. In Illinois, for example, the grant program is not funded, and a grant has never been awarded (Personal communication, February 16, 2018). 99 Table 5: State Financial Aid Programs for Exonerees State Statute Eligible Student Benefits at In-State Institutions MT Mont. Code. Ann. § 53-1-214 Exoneree who was incarcerated Tuition/fees waiver at public institutions up to five years MA Mass. Gen. Laws ch. 258D § 5 Exoneree incarcerated for at least 1 year 50% reduction of tuition/fees at public institutions Va. Code. Ann. § 8.01-195.11. Exoneree who was incarcerated Up to $10,000 for tuition/fees at public community college La. Stat. Ann. § 15:572.8 Exoneree who was incarcerated Tuition/fees at public institutions up to five years Conn. Gen. Stat. § 54-102uu Exoneree who was incarcerated Tuition/fees at public institutions 2008 Fla. Stat. Ann. § 961.06 Exoneree who was incarcerated Tuition/fees waiver at public institutions up to 120 credit hours Year 2003 2004 2004 2005 2008 2008 2009 2013 2014 2015 2017 2018 VA LA CT FL NC TX CO IL DC KS N.C. Gen. Stat. Ann. § 148-84. Tex. Civ. Prac. & Rem. Code Ann. § 103.054 Colo. Rev. Stat. § 13-65-103 Any exoneree Tuition/fees at public institutions up to five years Exoneree who was incarcerated Grant for tuition/fees at public institutions up to 120 credit hours Exoneree incarcerated for at least 3 years and their children Exoneree who was incarcerated Tuition/fees waiver at public institutions Reimbursement/payment of tuition/fees up to value of a four- year degree Grant for tuition/fees at public institutions up to 4 full-time years MN Minn. Stat. Ann. § 611.365 110 Ill. Comp. Stat. § 947/62 Any exoneree DC Code § 2- 423.02 Exoneree who was incarcerated Reimbursement for tuition/fees at public institutions Kan. State. Ann. § 74-32-195 Exoneree who was incarcerated Tuition/fees waiver at public institutions up to 130 credit hours 100 Theory of Social Construction of Target Populations The Phase I study was guided by the theory of social construction of target populations, though it was beyond the scope of the research questions to test the theory in whole (Schneider & Ingram, 1993; Schneider, Ingram, & deLeon, 2014). As I analyzed financial aid program eligibility rules, I considered which populations of justice-involved people had been targeted for disenfranchisement. Table 2 displays the five target populations that I identified from the 131 grant programs with eligibility requirements for justice-involved students. Incarcerated students were targeted most frequently, in 83 programs across 16 states, followed by students with criminal convictions targeted in 35 programs across 14 states. But Schneider and Ingram (1993) did not develop the concept of target populations; rather, they introduced an explanation for how and why those target populations are created by politicians (social constructions) and how they can be exploited politically (benefits and burdens to the target populations). Through the identification of target populations in this study, an interesting tension within the theory is revealed: when a person occupies multiple levels of power and deservingness at the same time. College students, as conceptualized by Schneider, Ingram, and deLeon (2014), are viewed as relatively low in power but quite high in deservingness of public benefits. Thus, for policymakers, it is typically politically popular to afford public benefits to college students through scholarships. Constituents likely view the politicians as helping a needy population. But what if students hold other social identities? When a college student is also justice-involved, a veteran, a single mother, undocumented, Native American, in debt, wealthy, elderly, or a business owner, for example, what happens to the student’s power and deservingness? Which identities are more important in the eyes of lawmakers? The policies identified in Phase I suggest that politicians had much to gain by separating the “deviants” from the otherwise deserving 101 college students. And, perhaps, in order to justify to their constituents that spending tax dollars on scholarships is beneficial, excluding justice-involved students was politically necessary to show that they still understood who does not deserve public benefits. These statements are propositions that are supported by the theory but that need further empirical investigation in this context. Other than criminal history, college students face a wide range of eligibility requirements when applying for financial aid, including academic merit, financial need and history, citizenship and residency status, age, membership in certain special populations, choice of academic major or program, and much more (Delaney & Ness, 2009). Generally, students who have good grades, financial need, are US citizens, or who are veterans are likely to be afforded more financial aid than those who are in default on loans, are undocumented, or are entering college long after leaving high school. These intersections of social identities raise interesting theoretical questions about how policymakers, according to the theory, would award benefits or burdens to financial aid applicants. In addition to the five categories of justice-involved target populations I identified, additional research on financial aid eligibility requirements is needed to identify other subpopulations of college students that are constructed by politicians for benefits or burdens. In other words, it is yet to be determined empirically which populations of students are socially constructed as most deserving compared to those that are constructed as least deserving according to politicians. Certainly, justice-involved students in 26 states seem to appear to be among the latter group. What the Phase I results do not offer, however, is direct evidence as to whether these five target populations were consciously constructed by policymakers in the ways the theory suggests. Did politicians exclude incarcerated students from getting state aid in 26 states because 102 incarcerated people are social deviants and because voters expect politicians to punish them, as the theory suggests? Or are there other explanations for why incarcerated students lost access to the grant programs? These types of questions are precisely the pursuit of the Phase II, to understand how and why state policymakers disenfranchised certain justice-involved populations from financial aid. Though the theory of social construction of target populations is not carried through explicitly to the next phase as the central framework, the conceptual model that is used incorporates similar theoretical concepts. Conclusion Phase I succeeded in describing the state financial aid landscape for justice-involved students. At least 131 state financial aid grant programs were identified that explicitly deny aid to one of five targeted populations of justice-involved students, including: incarcerated students, students with criminal convictions (generally), students with drug convictions (specifically), students who are ineligible for federal financial aid because of drug convictions, and students who violate pledges to be remain drug-free. The history of each program was described, including policy adoption dates, the evolution of eligibility rule language, and key policymakers. In addition, patterns in the chronology of policy adoption, program type, and geography were explored to provide a full description of the landscape. These findings have many implications for researchers, college student services providers, and policymakers, which were presented in turn. However interesting the results of Phase I may be by themselves, the purpose of conducting the Phase I study was to inform a qualitative dissertation on the adoption of state higher education policies. Now knowing that 131 programs are off-limits to justice-involved students, it raises the questions as to how and why state policymakers disenfranchised them. In 103 the next chapter, a selection of the 131 state grant programs that deny eligibility to justice- involved students are examined in depth to uncover their histories and to explore the factors that influenced their adoption. 104 CHAPTER 3: POLICY ADOPTION STUDY (PHASE II) The review of state policies in Phase I identified 26 states that adopted financial aid grant programs containing eligibility rules for justice involve students. With the descriptive analysis from Phase I as a foundation, further investigation into several of these states and their policies was needed to understand how and why these policies were adopted. Investigating how and why state legislators adopt policies is a tradition of political science scholarship, and policy adoption research within the higher education literature framed this portion of the dissertation. Specifically, Phase II employed a new conceptual model of state policy diffusion and innovation in higher education (Hearn, McLendon, & Linthicum, 2017), and the results of this study were among the first to examine the extent to the which the model explains policy adoption in a particular context. Literature Review Answering the how and why research questions of Phase II requires a theoretical understanding of the state policy adoption process. This section presents a review of policy innovation and diffusion theories, a review of the state financial aid policy adoption literature, and a conceptual model for state policy innovation and diffusion in higher education. The methodologies of this literature proved to be especially insightful, and a new conceptual model offered the best-available synthesis of the many factors theorized to influence the adoption of state higher education policies (Hearn, McLendon, & Linthicum, 2017). This body of literature framed the design of Phase II of this dissertation. State Policy Adoption The comparative study of state policies represents a vast, rich vein of research in the fields of public policy and political science (Hearn, McLendon, & Linthicum, 2017). 105 Understanding why one state decided to adopt a policy, or to “innovate,” is important in understanding the whole policymaking process (Berry & Berry, 2014). From over forty years of policy research, there are two dominant explanations for why states adopt new policies. First, internal determinants refer to the conditions within a state. States adopt policies because of the demographic, political, fiscal, and many other factors within the state. Second, diffusion refers to conditions outside of the state. States adopt new policies by mimicking another state’s policies. Although early policy innovation research focused on these explanations separately, most research since the 1990s considers these explanations together (Berry & Berry, 2014). In other words, states are now understood to adopt policies because of both internal determinants and diffusion factors. To begin, I provide an overview of the classical diffusion and internal determinants theories before reviewing policy adoption studies from the field of higher education. Diffusion. Broadly, “policy diffusion is defined as one government’s policy choices being influenced by the choices of other governments” (Shipan & Volden, 2012, p. 788). To scholars of American politics, diffusion is more specifically “the process by which a public policy or program…spreads among the members of a social system, most frequently understood to mean the governments of the 50 American states” (McLendon, Cohen-Vogel, & Wachen, 2015, p. 116). Berry & Berry (2014) identified five mechanisms for policy diffusion. First, states adopt policies from other states when they learn that the policy is effective or is otherwise attractive to adopt. Second, states sometimes simply imitate other states when they want to act more like another state. Third, normative pressure leads states to adopt policies that are becoming popular in other states. Fourth, competition can result in policy adoption when states want “to achieve an economic advantage over other jurisdictions or, equivalently, to prevent 106 other jurisdictions from securing an advantage over it” (Berry & Berry, 2014, p. 294). Finally, coercion occurs when one government incentivizes or forces another government to adopt a policy, which in the context of U.S. state policymaking typically occurs when the federal government or courts mandate changes to state laws. These mechanisms are implicated in several earlier models of policy diffusion. First, the national interaction model predicts that states will adopt policies following an S-shaped curve; that is, states are at first slow to adopt a new policy (bottom of the S), then many states adopt the policy within a short period of time (upward curve of the S), resulting in a leveling off in adopting due to the small number of potential adopters left (top of the S) (Berry & Berry, 2014). This model assumes that all 50 states are equally likely to adopt a particular policy over time, which is a weakness of the model considering the many differences between states. Second, the regional diffusion model predicts that a state has a higher probability of adopting a new policy when its bordering states (neighbor model) or other states in its region (fixed-region model) have already adopted the policy. Third, the leader-laggard model suggests some states or policymakers are leaders or pioneers that are known to take risks in adopting new policies, after which the laggard or follower states also adopt the policy (Berry & Berry, 2014). Internal determinants. Theorizing internal determinants as explanations for state policy adoption stands in contrast to that of diffusion. Instead of policies transferring across state borders, policymakers are understood to adopt policies based on the conditions within a state. There are two general hypotheses that drive this theory: motivations to innovate and availability of resources to innovate (Berry & Berry, 2014). There are many factors within a state that motivate policymakers’ decisions to adopt policies. For example, the severity of certain social problems can directly lead to policy adoption, or policymakers may be more likely to adopt 107 popular policies nearer elections in order to win. Still other factors relate to financial resources. Wealthier states are believed to be more innovative, as are states that invest in more professionalized legislatures (Berry & Berry, 2014). A combined model of policy innovation. Until the 1990s, policy scholars typically considered state policy adoption as either a function of diffusion or internal determinants, but in hindsight, early statistical approaches used to test diffusion and internal determinants theories were deeply flawed (for a review, see Berry & Berry, 2014). In their study of state lotteries, Berry and Berry (1990) revolutionized the study of state policy adoption with their novel use of event history analysis (EHA) to test simultaneously diffusion and internal determinants. EHA is a type of survival analysis that, in this case, predicts the probability that a state will adopt a policy in a given year, considering any number of other variables, including factors related to policymakers’ motivations, state funding conditions, and the number of neighbor states that adopted the policy (Berry & Berry, 2014). Many methodological improvements have been made on the use of EHA in policy studies, and more importantly according to Berry and Berry (2014), researchers have increasingly focused on the mechanisms for policy diffusion, rather than seeking to prove whether policies diffuse across borders at all. Policy Diffusion and Innovation in State Financial Aid Studies Amongst the higher education policy adoption literature, several studies have examined the adoption of state financial aid policies (Cohen‐Vogel & Ingle, 2007; Cohen-Vogel, Ingle, Levine, & Spence, 2008; Doyle, 2006; Ingle, Cohen‐Vogel, & Hughes, 2007; Ingle & Petroff, 2013). An important observation is that while most policy adoption research employs quantitative methods (Cohen-Vogel, Ingle, Levine, & Spence, 2008), most of the studies of state financial aid policy adoption employ qualitative methods. Perhaps more so than the actual 108 findings, the methods of these studies have useful implications for this dissertation. As such, I divide my review of this literature by method. Quantitative. Noting the lack of understanding of the reasons states were rapidly adopting merit-aid policies, Doyle (2006) was the first to study how characteristics of states affected their likelihood of adopting merit-aid programs. Using event history analysis (Cox proportional hazard model), data from 48 states were analyzed. The researcher posed ten hypotheses on the relationships between states’ likelihood to adopt merit-based programs and state factors related to diffusion, policy rationales, political characteristics, and demographic characteristics. Surprisingly, Doyle found that the number of neighboring states with merit-aid policies had no significant effects on likelihood to adopt the policy, thus refuting the theory that policies diffuse across states, at least in this case. Political factors, such as political party, also had no significant effects on policy adoption. States were more likely to adopt merit-aid policies when they had lower levels of high school students continuing directly to college, lower levels of educational attainment, lower levels of out-of-state college-going migration, and lower levels of education achievement. In addition, states with higher levels of median family income and with higher proportions of people aged 18-24 decrease a state’s likelihood to adopt the policies. The author concluded that state legislators adopt policy in response to conditions within the state, rather than as a function of political ideology or mimicking neighbor states (Doyle, 2006 Qualitative. In a series of articles, education policy scholars Lora Cohen-Vogel, W. Kyle Ingle, and their colleagues studied the policy adoption process of broad-based merit-aid policies in the Southeastern United States (Cohen‐Vogel & Ingle, 2007; Cohen-Vogel, Ingle, Levine, & Spence, 2008; Ingle, Cohen‐Vogel, & Hughes, 2007). Breaking with the quantitative tradition of policy adoption research, they used qualitative methods to understand why and how states 109 adopted the policies. The authors analyzed archival policy documents and interview data from 26 key policy actors, using semi-structured interview questions and both theory-driven and inductive coding. In each of the three articles reviewed next, the authors analyzed the rich data set in different ways to offer competing explanations for the adoption of the financial aid policies in the Southeast. The Southeastern United States is home to the biggest cluster of broad-based merit-aid policies, which suggests that regional diffusion is a logical explanation for their adoption. In their first article, Cohen-Vogel and Ingle (2007) sought to determine to what extent diffusion influences were present throughout the stages of policymaking process. They found strong evidence of diffusion factors in the agenda setting stage, whereby external policy entrepreneurs were intentional in spreading merit-aid policy ideas across stateliness, and in the proposal formulation stage, whereby legislators used other state policies (particularly Georgia HOPE) as models. They did not find diffusion influences in the first stage of problem discovery or the final stage of passing and enacting the legislation. This study not only confirmed that regional diffusion occurred, but it also identified when those influences were most salient in the policymaking process (Cohen-Vogel & Ingle, 2007). Although evidence for regional diffusion of policies was found in the Southeastern states (Cohen-Vogel & Ingle, 2007), several states did not adopt the policy (or adopted but failed to fund), called “hold-outs” (Ingle, Cohen‐Vogel, & Hughes, 2007). To understand why some states did not adopt policies and to further test policy diffusion theory, the researchers then compared data from three hold-out states to six adopter states in the Southeast. Regarding diffusion factors, adopters commonly reported that interstate competition played a role in adopting the policies, including competing to keep students in the state for college and for work after college. Policy 110 networks were also important diffusion influences, such as governors’ associations, education organizations, and the National Association of State Student Grants and Aid Programs (Ingle, Cohen‐Vogel, & Hughes, 2007). In contrast, two of the hold-out states made no mention of these influences. Regarding internal determinants, adopters commonly reported the strength of state economies and a general lack of political opposition to merit-aid policies at the time of policy adoption. But for hold-outs, other spending priorities (like building new roads and schools) competed with the adoption of new merit-aid programs, anti-gaming coalitions succeeded in blocking new lottery-funded programs, and voters rejected tax increases to fund merit-aid programs (Ingle, Cohen‐Vogel, & Hughes, 2007). The strength of this study lies in its innovative use of qualitative methods to explore directly why policy actors did and did not fully adopt policies. In their third qualitative analysis of merit-aid policies from Southeastern states, Cohen- Vogel, Ingle, Levine, and Spence (2008) considered the mechanisms for regional diffusion. They specifically investigated three theoretical explanations for why the financial aid policies diffused across borders in the southeast. First, the authors found evidence of interstate competition. States were primarily competing to keep students from leaving the state (for revenue and to prevent brain drain) but also adopted the popular merit-aid policies to conform to normative pressure. Second, some evidence of convenience was uncovered, suggesting that policymakers adopted the policies simply because other states had already tried it and believed it was a good idea to borrow. Third, policy communities, professional organizations and associations, and policy networks were found to be influential in the diffusion of merit-aid policies in the Southeast. Governors, lawmakers and scholarship program administrators learned about merit-aid policies within these policy communities, which influenced their eventual adoption. This study suggests 111 that a qualitative approach is appropriate for uncovering why states adopt policies, and it adds empirical support for competition, convenience, and policy communities as explanations for regional diffusion (Cohen-Vogel, Ingle, Levine, & Spence, 2008). Because the Southeast states received adequate attention from financial aid policy scholars, Ingle and Petroff (2014) studied merit-aid policies in Alaska, Massachusetts, Michigan, and Nevada. Since regional diffusion factors were not logical explanations for their adoption, the authors used qualitative methods to explore the role of internal state determinants and policy entrepreneurs as explanations for the adoption of merit-aid policies in these states. Like the studies reviewed above, the authors studied policy documents and interviewed 22 key policy actors to understand when and why merit-aid policies were adopted, using semi-structured interviews and theory-driven coding. Unlike the other studies, the authors framed this study as a case study, and the results are presented by state rather than by themes. Several political and economic factors were consistently found across the four states: a strong public advocate, a Republican governor advocate, a need to strengthen the state economy and diversify its workforce, and a strong or improving state budget (Ingle & Petroff, 2013). A final study is relevant here not because it examined the adoption process of state financial aid programs but because it examined how policy actors determined the academic eligibility criteria for them (Ness, 2008; 2010). Using three policy process frameworks (advocacy coalition, multiple streams, electoral connection), Ness (2008; 2010) conducted a comparative case study analysis of merit-aid policies in New Mexico, West Virginia, and Tennessee. He interviewed 56 policy actors, including governors, legislators, their staff, university system officials, state agency officials, and researchers, and triangulated his findings with archival documents. Finding that the multiple streams framework held the best explanatory 112 power for how policy actors developed the scholarship criteria, he revised the framework to highlight the role of policy entrepreneurs and to account for “organizational structures and policy trends both within the state and external to the state” (Ness, 2010, p. 52). Ness’ (2010) use of an analytical framework and the way he presented his case analysis are excellent models for this dissertation. For each state, he provided an overview of state context (including background on the executive and legislative branches), a narrative, chronological history of the events related to the policy being studied (from interviews and document collection), and a case analysis of those events against the dimensions of his analytical framework (comprised of key questions from the three theoretical frameworks). Then, he engaged in a comparative cross-case analysis to find similarities and differences in how well each of the three theoretical frameworks explained the policy decisions across the states (Ness, 2008). This approach provided a model for the strategy and organization of this dissertation. Conceptual Model of State Policy Innovation and Diffusion in Higher Education Drawing on decades of state policy adoption research from political science and higher education, Hearn, McLendon, and Linthicum (2017) developed a conceptual model of state policy innovation and diffusion in higher education (see Figure 6). As such, it incorporates both internal state factors (internal determinants) and interstate factors (diffusion factors) in explaining why a state might adopt a new higher education policy. These explanatory factors are organized into “four distinct sets of forces,” which are described next (Hearn, McLendon, & Linthicum, 2017, p. 321). 113 Adoption of State Postsecondary Policies Socioeconomic Context Demographic: population size, age-group distribution, growth rate, density, % of color, # SMSA’s over one million, etc. Educational: educational attainment levels, postsecondary enrollment rate, etc. Economic: gross state product [GSP], GSP per capita, GSP per-capita growth, median per- capita income, unemployment rate, etc. Organizational and Policy Context Organizational Ecology: number and the proportional distribution of institutions by control and level, the number and proportional distribution of students by institutional sector, student migration patterns, number and strength of research institutions in the state, agency research capacity, etc. Policy: postsecondary funding, tuition, and enrollment patterns. Politico-Institutional Context Political Ideology: state ideology indicators Legislative Professionalism: Squire Index Partisanship: legislative/gubernatorial party affiliations and strength, etc. Electoral Conditions: degree of parties’ electoral competitiveness, election timing, etc. Gubernatorial Strength and Tenure: constitutional authority, time in office, etc. Interest-group Climate: interest-group characteristics (density, strengths, concentration) and governance arrangements (consolidated governing board across institutional sectors, strong coordinating board, weak coordinating board, planning agency) Policy Diffusion Context Decision Efficiency: states may learn from other states’ actual experiences with policy choices, thus informing and ideally improving their policy choices Competitive Advantage: states may adopt policies to position them to better compete with other states Normative Pressure: states may adopt policies favored by regional or national-level peers, intermediary organizations and policy entrepreneurs. Coercive Pressure: states may be forced to adopt policies imposed by critical outside resource providers Figure 6: A conceptual model of state policy innovation and diffusion in higher education. 114 From “Conceptualizing State Policy Adoption and Diffusion,” by J.C. Hearn, M. K. McLendon and K. C. Linthicum, 2017, in Higher Education: Handbook of Theory and Research, 32, p. 320. Copyright 2017 by Springer International Publishing. Recreated and reprinted with permission. State socioeconomic context. The socioeconomic context of a state includes factors related to population demographics, education, and economics. Regarding demographics, the size and age distribution of a state’s population are influential in postsecondary policymaking. Regarding education, states with lower educational attainment rates and/or a need for more educated workers have been shown to be more likely to adopt broad-based merit-aid programs, but states with higher levels of education have been shown to be more likely to innovate, generally. Finally, state economic factors like wealth are also expected to influence policy innovation. Organizational and policy context. A state’s higher education “organizational ecology” is comprised of factors related to its higher education governance system, types of institutions (public/private, 2-year/4-year, profit/non-profit), and enrollment behaviors of students, among others. The policy context includes factors related to a state’s history of higher education policymaking, including funding levels, tuition policy, and enrollment patterns. Politico-institutional context. The politico-institutional context is a particularly rich grouping of factors, each with growing evidence bases. Six “political conditions and institutionalized arrangements that can shape governments’ behavior in policy innovation” comprise the next set of forces, including: political ideology, legislative professionalism, partisanship, electoral conditions, gubernatorial strength and tenure, and interest-group climate (Hearn, McLendon, & Linthicum, 2017, p. 325). Political culture has long been theorized to influence state policy making, but research has not found strong evidence of such, including in higher education studies. More recent 115 conceptualizations of political ideology as attitudes on a continuum of liberal-conservative have shown much stronger empirical links to state policy outcomes. Both citizens and policymakers in a state can be scored on the continuum, allowing for each state to be assigned an overall score of political ideology for study over time (Berry, Ringquist, Fording, & Hanson, 1998). Links between state higher education policymaking and political ideology have been found, but the direction of influence appears to be inconsistent across studies, such that: some studies have found liberal-leaning citizenries and states more likely to support robust public spending on public colleges and universities and more prone to enact the newer financing initiatives, while other studies have found conservative leaning states more likely to do so. (Hearn, McLendon, & Linthicum, 2017, p. 327) Distinct from political ideology is partisanship, which refers to the political party in control of state governments. Only recent research on higher education policymaking has found trends related to political party, including that Democrat-controlled states tend to fund higher education at higher levels and that Republican-controlled states tend to adopt more performance- accountability measures (Hearn, McLendon, & Linthicum, 2017). Legislative professionalism refers to the structure of a state’s legislative body. For example, in some states, legislators work full-time year-round, are well-paid, and have large staffs, whereas in other states, legislators work part-time for only a few months out of the year, receive small stipends, and have little staff support. These factors, and others, describe the extent to which a state legislature is professionalized, and more professionalized legislatures tend to be more innovative (Hearn, McLendon, & Linthicum, 2017). Electoral conditions are comprised of two factors: electoral competitiveness and timing of elections. Electoral competitiveness is the extent of competition between political parties in 116 elections, and timing of elections refers to the periods of time between a policy action and the next election. Policy adoption behaviors of policymakers appear to change depending on if they are facing an imminent election cycle (Hearn, McLendon, & Linthicum, 2017). Governors in U.S. states are influential political leaders, though research on their direct influence on state policy adoption is mixed. Still, the length of a governor’s tenure, how generally strong they are, and the extent of their formal powers are understood to influence state policy adoption (Hearn, McLendon, & Linthicum, 2017). Finally, interest-group climate is the last factor in the politico-institutional context. Lobbying and the density of lobbying groups in a state are elements of this factor, but Hearn, McLendon, Linthicum (2017) focus on how the structure of state higher education governance shapes the interests that influence state policy. The institutionalized operations of the three forms of state higher education governance (planning agency, consolidated governing board, and statewide coordinating board) have unique implications for how policy interests are channeled and transformed into higher education policy. Policy diffusion context. For the reasons outlined above in the section on policy diffusion theory, interstate factors comprise the final category of the conceptual model. Though evidence is mixed on the extent to which states influence other states’ higher education policy, policy diffusion as a theory is far too prominent in the study of state policy adoption to be dropped from the conceptual model. Hearn, McLendon, and Linthicum (2017) conceptualized four types of diffusion in their model, not unlike those described by Berry and Berry (2014). Decision efficiency diffusion suggests that states learn from other states and copy their policies. Competitive diffusion suggests states adopt policies to gain competitive advantage over other states. Normative pressure diffusion occurs when states adopt policies that are popular or seen as 117 best practices. Lastly, coercive diffusion occurs when a state is seemingly forced to adopt a policy because of pressures from influential groups (Hearn, McLendon, & Linthicum, 2017). Hearn, McLendon, and Linthicum’s (2017) conceptual model is grounded in a rigorous review of the state higher education policy adoption literature. The large number of explanatory factors identified within the four thrusts (socioeconomic, organizational and policy, politico- institutional, and policy diffusion contexts) suggests the policy adoption process is complex and context-dependent. Some combination of these factors, according to the model, is likely to explain why a state adopted a given higher education policy. Unlike a fully-developed the theory, however, the model does not suggest directionality of influence, which requires additional research to resolve. The model is well-suited for this study because it offers an organized, concise starting point for understanding the factors that might have influenced states’ adoption of financial aid policies that deny eligibility to justice-involved college students. Conclusion The goal of this literature review was to establish a theoretical grounding for the Phase II study. The study of state policy innovation and diffusion is a rich tradition in American political science, and its application in higher education policy studies is growing. State merit-aid policies are a favorite topic of higher education policy scholars, but a theme of much of this literature is that of mixed results. No single theory or factor is adequate to explain state policy adoption in higher education studies or in the broader literature. Rather, state policy adoption is a complex process affected by many internal and external influences. The conceptual model for policy adoption in higher education appropriately merges four broad categories of factors that all offer some empirical evidence of causal influence (Hearn, McLendon, & Linthicum, 2017). For this dissertation, the model gives clues as to why state legislators would adopt rules for justice- 118 involved college students in financial aid programs. In addition, the methodologies of the literature reviewed in this section proved to be particularly insightful for this dissertation, as several broke from the quantitative tradition to employ qualitative methods. These qualitative studies succeeded in uncovering the actual rationales for why and how policymakers moved to adopt them, which suggests similar methods can be successfully employed here. Methods Phase II of this dissertation explores the history of several states’ financial aid policies and the factors and conditions that led policymakers to adopt rules for justice-involved students. There is one, two-part research question: How and why did state policymakers in some states adopt eligibility rules for justice-involved students in financial aid grant programs? To break the question down, the how question investigates the mechanisms and processes used to adopt the policies. For example, did state agency officials conduct research before proposing the policy? How was the proposal created and considered? Who was involved? Did state legislators create the policy using the legislative process or did agency officials promulgate regulations? The why question investigates legislative intent and the rationale behind adopting the policies. For example, what was the purpose of denying aid to justice-involved students and why was it done so at the time? What were the reasons used to justify the decision to adopt the policies? What did legislators expect the intended outcome to be? As described above, the conceptual model presented a range of possible explanations to these questions. This study explored the extent to which the model applied to this specific state policy adoption context: state financial aid grant programs for justice-involved students. 119 Assumptions & Positions Before describing the methodology, it is important to share my assumptions and positions on this research task. Responsible qualitative researchers reflect on their strengths, shortcomings, motivations, and biases throughout the research project to demonstrate “honesty and authenticity with one’s self, one’s research, and one’s audience” (Tracy, 2010, p. 842). First, I assumed that state legislators were rational (Baumgartner, Jones, & Mortensen, 2014), meaning I believed they formulated reasons for deciding to include, or not include, rules for justice-involved students in their financial aid policies. The conceptual framework suggests many possible influences for why they were adopted, but I do not believe these rules could have been coded into law for no reason at all. I assumed, therefore, that people involved in the policy adoption process could explain their reasons for writing in those rules. Similarly, I assumed those explanations were trustworthy accounts of the policy adoption events. I also have a position on the question of whether justice-involved college students should be eligible for financial aid. My previous research makes clear that I, at times, openly advocate for improving opportunities for incarcerated students and students with prior criminal history to access higher education (see Custer 2013; 2016; 2018a; 2018b). Based on my work and the large body of literature on the positive effects of higher education in prison programs, I believe higher education is one of the best available rehabilitative tools for people involved in the criminal justice system. If society expects justice-involved people to stay crime free, it would make the most sense that they would be supported in their pursuit of higher education, including through student financial aid. In my opinion, policies that disenfranchise justice-involved people from higher education are discriminatory and should be eliminated. 120 However, I did not approach this study with contempt for the policy actors who advocated denying financial aid to justice-involved students, whatever their reasons may have been. As I assumed the roles of historian and social scientist for this study, I committed to respecting the subjects of this research, to avoid generalizing about them, and to be aware of my own biases in interpreting the findings (Rampolla, 2012). I understand that I am thinking and writing in 2017-2019 about events and decisions that took place in different times, in different places, and under different conditions than the present. I agreed to study, describe, and critique the viewpoints of people from the past without judging them. Methodology Using previous studies of state financial aid policy adoption as methodological models (Cohen‐Vogel & Ingle, 2007; Cohen-Vogel, Ingle, Levine, & Spence, 2008; Ingle, Cohen‐Vogel, & Hughes, 2007; Ness 2008; 2010), Phase II of this dissertation employed qualitative methods to uncover the rationales and historical contexts for adopting eligibility rules for justice-involved students in a selection of state policies. Because individual state policies are the objects of study, my methods were informed by qualitative case study methodology. “Qualitative case methodology is an approach to research that facilitates exploration of a phenomenon within its context using a variety of data sources” (Baxter & Jack, 2008, p. 544). The methodology is best suited for research questions that ask “how” and “why” and that require an in-depth investigation of the phenomenon to answer those questions (Yin, 2009), which precisely suited the Phase II study. This case study was explanatory in nature because it uncovered explanations for how and why the policies adoption in a selection of states (Baxter & Jack, 2008). Specifically, the conceptual model (Hearn, McLendon, & Linthicum, 2017) provided the theorized factors that influence state policy adoption, and the purpose of the case 121 investigation was to determine if these factors, or others, explained how and why state lawmakers adopted rules for justice-involved students in financial aid policies. These explanations were derived from the direct accounts of policymakers (i.e., through interviews or debate transcripts) and/or other historical legislative records. An important ontological assumption of this study is that the explanations drawn from these sources – especially from the policymakers themselves – represent a valid version of the truth that can be relied on for analysis. In a holistic multiple-case design (Yin, 2009), each policy becomes a case, and each policy has its own unique context that needs to be investigated as part of the study. Said differently, “in a multiple case study, we are examining several cases to understand the similarities and differences between the cases” (Baxter & Jack, 2008, p. 550). “Holistic” distinguishes the study from an embedded study; an embedded study has individual embedded units of analysis within each case, which does not apply here (Yin, 2009). Replication is a key concept of multiple-case studies. For each of the cases (state policies), I replicated the same process of collecting data and writing individual case profiles; then, I conducted a cross-case analysis to find themes, commonalities, and differences in relation to the conceptual model (Yin, 2009). The analytical procedures are described in more detail below. Data Collection A strength of case study methodology is its reliance on multiple sources of data, which improves study credibility (Baxter & Jack, 2008). The use of multiple sources is a natural form of triangulation, whereby the researcher seeks to confirm findings across the sources (Tellis, 1997). Of the six sources of evidence described by Yin (2009), three are relevant to this study: documentation, archival records and interviews. (The differences between “documentation” and 122 “archival records” are minimal, so the terms archival records, historical records, or historical documents will be used interchangeably throughout this study). First, archival records may constitute any pre-existing document containing relevant information, and these records can be further distinguished as being primary or secondary sources. “Primary sources are materials produced by people or groups directly involved in the event or topic under consideration, either as participants or as witnesses” (Rampolla, 2012, p. 6). Primary sources collected and reviewed included state statutes and regulations, legislative records (e.g., various versions of bills, bill histories, bill fiscal analyses, legislative journals), state government policy documents (e.g., agency reports, attorney general opinions, executive orders), court cases, transcripts of speeches and floor debates, letters and emails, and news reports. Secondary sources, which “synthesize, analyze, and interpret primary sources” (Rampolla, 2012, p. 7), were also collected. These included research articles and reports that were not written contemporaneously to the events they discussed. The availability of primary and secondary records varied depending on the date of the policy and state. For example, news reports for policy events that occurred in the past 20 years (when most news was published online) were more readily available compared to events that occurred farther in the past when news was published primarily in print. In addition, some states make more legislative records available online than others, like Illinois that publishes transcripts of all legislative floor debates. Many libraries, online databases, and state agencies were accessed to collect the records described above. I collected most records from state websites (e.g., state general assembly websites for bills, state agency websites for reports, etc.), but many were obtained through university and law libraries (i.e., via direct contact with librarians and interlibrary loan requests). For retrieving old state session laws, the Hein Online State Session Laws Library database 123 proved invaluable, from which I downloaded hundreds of pages of text. For state legislative journals, the online database LLMC-Digital was frequently used. To supplement the archival data, I conducted semi-structured interviews with key policy actors to get a deeper understanding of how and why state legislators adopted eligibility rules for justice-involved students in some state grant programs. After all, directly asking policymakers how and why they adopted certain policies is the most efficient way to answer the research questions. Key policy actors included anyone with first-hand knowledge of the genesis of the policy cases, including state legislators, legislative staff, staff from state agencies, and other individuals who were involved in the policy adoption process (e.g., lobbyists, policy experts). Between the legislative records, interview transcripts, and websites, hundreds of pages of text were collected and reviewed over a seven-month period. Sample Phase I identified 131 policies across 26 states that contained eligibility rules for justice- involved students; however, most of these did not make useful cases for analyzing policy adoption. Many of them contained the eligibility rules because of a policy decision that was applied uniformly across all financial aid programs at a different time from when it was adopted. For example, when Illinois legislators passed a law in 1989 to ban incarcerated students from all state aid, the law was retroactively applied to at least three existing programs and to at least four programs that would be created after 1989. These seven programs, therefore, did not make useful cases for analysis, but the 1989 incarceration law did. By examining the history and context for the 1989 law as a case, the research questions could be answered. Of all the identified policy adoption events (see Table 4), some were determined not to be feasible for investigation. One factor that affected sampling decisions was the current position of 124 the former state lawmakers who authored the laws. For example, consider the 2000 California law that created the Cal Grant Program and that made incarcerated students ineligible. Of the four state legislators whose namesakes commemorate the law (“Ortiz-Pacheco-Poochigian- Vasconcellos Cal Grant Act”), one is deceased, one is a state appellate judge, and one is a district attorney. Similarly, of the lead authors of the 1999 TEXAS Grant that made drug offenders ineligible, one is a sitting U.S. Congressman and the other was convicted of federal crimes during the time of this study. In these cases, and in others, I did not attempt to contact such individuals for interviews, assuming that a response was unlikely. California and Texas, therefore, were not pursued as cases. Another factor that affected sampling was the date of the policy event. At least six important policy adoption events took place prior to 1990, over thirty years ago (see Table 4). In these, the lawmakers are likely deceased, very elderly, or otherwise unable to be interviewed. Though I attempted to identify and locate participants from this period in states like Illinois, Pennsylvania, and Michigan, I ultimately did not make any contacts in those states pertaining to the pre-1990 policy events. Where possible, historical records for these older programs were collected to answer the research question in lieu of interviews. A third challenge was identifying the lead policy actors. In several cases, the policy in question was buried in lengthy omnibus education or budget bills, effectively obscuring who was primarily responsible for the policy. Such bills can have tens of sponsors, and the likelihood that the chief bill sponsor would remember one provision of it was unlikely. This limitation applied to policies in New York, New Jersey, Michigan, and Indiana. Despite attempts, I could not determine who the best policymakers were to try to contact in cases like these. 125 The last factor that affected sampling was the response rate. Given the limitations described above, individuals from 11 states were contacted for interviews, though individuals from only 7 states participated in interviews. The final sample of states for Phase II analysis includes Pennsylvania, Illinois, Michigan, Delaware, Georgia, Oklahoma, Indiana, Utah, Wyoming, and Kentucky. More information on how participants were identified and contacted for interviews is provided below. Identifying and Contacting Participants State policymakers, typically bill sponsors, were first identified from legislative records (e.g., bills, legislative journals). Internet search engines were then used to investigate an individual’s status (i.e., deceased, retired, employed). Many did not have a web presence, making it impossible to contact them for interviews. Through interviews, additional participants were identified. Participants were contacted primarily through email, though in a few cases, participants were contacted through social media (i.e., LinkedIn, Facebook) or by phone. Of those who did not respond after the first email, a follow-up email was sent two weeks later. Subsequent contacts were not made. A template of the interview invitation email is in Appendix B. In total, 43 people were contacted for interviews, and 13 agreed to be interviewed from 7 states, for an interview rate of 30%. Table 6 displays the number of identified individuals from each state who were contacted for interviews compared to those who were interviewed. Individuals from MI, MS, MO, and TN did not respond or declined to be interviewed, which meant those states were excluded from the Phase II sample.175 Table 7 displays profiles of the 13 individuals who agreed to be interviewed. 175 Historical records were used instead to create a case study for Michigan. 126 Table 6: State Policy Actors Contacted for Interviews State # Contacted # Interviewed Delaware Georgia Indiana Kentucky Michigan Mississippi Missouri Oklahoma Tennessee Utah Wyoming TOTAL 2 5 8 9 4 3 1 3 2 3 3 1 3 1 3 0 0 0 2 0 2 1 43 13 Table 7: Participant Profiles State Policy (Subject of Interview) Position at Time of Policy Event Delaware 2005 SEED Program Georgia Georgia Georgia Indiana 1993 HOPE Program 1998 HOPE Statute 1998 HOPE Statute State Legislator State Agency Official Lobbyist State Legislator 2011 Higher Education Award Amendments Governor’s Office Staff Kentucky 1996 Bill 285 Kentucky 1998 KEES Program Kentucky 1998 KEES Program Oklahoma 1995 OTAG Amendments Oklahoma 1995 OTAG Amendments Utah Utah 2008 Regents’ Scholarship & 2010 New Century Amendments 2008 Regents’ Scholarship & 2010 New Century Amendments State Legislator State Legislator Governor176 State Legislator State Legislator State Legislator State Agency Official Wyoming 2006 Hathaway Scholarship State Legislator 176 Because disguising the identity of a governor may be impossible, former Kentucky Governor Paul Patton gave me explicit, verbal permission to identify him in this report. 127 Instrument The questions on the interview protocols (see appendices C and D) were designed to elicit how well the conceptual model (Hearn, McLendon, & Linthicum, 2017) explained the policy adoption outcomes in the selected cases. The questions specifically asked about the policy adoption factors identified in the model. Some of the questions were drawn directly from, or are based on, the interview protocols provided in the appendices of Cohen-Vogel, Ingle, Levine, and Spence’s (2008) study and Ness’s (2008) dissertation. Because these studies were so successful in exploring the policy adoption process of state financial aid programs, their interview protocols were strong models for this dissertation. There were two interview protocols. Appendix C contains a protocol for policy actors who were directly involved in the policy adoption process, such as bill sponsors. Appendix D contains a protocol for individuals who were not directly involved in the policy adoption process but who are generally knowledgeable about the backgrounds of a policy. This was administered to individuals like state agency officials and legislators who were not active sponsors of bills. Interviews Interviewing current and former policymakers required sensitivities that may not be required for interviewing other types of people. Harvey (2011) offered advice on interviewing “elites,” such as CEOs or politicians, including length of interviews, how to conduct oneself with an elite participant, whether to record, handling participants who do not answer the questions, and other related issues. In practice, this advice proved useful, as the interviews did not resemble the ideal interview described in qualitative research methods courses or texts. None of the interviews were conducted in person; most were conducted over the phone, and a few were conducted through a web-conferencing program, with one being conducted over 128 video conference. Because none of the participants were in Michigan, phone interviews were the best option; they also provided necessary flexibility to the participants, some of whom were quite busy (Harvey, 2011). Some participants scheduled time with me, while others spoke with me during open windows of time while: driving, walking, at home from work due to illness, or otherwise between engagements. Harvey (2011) also predicted that my interviews with politicians would vary in length and may need to be shorter than the typical qualitative interview. “Asking for too much time might lead to respondents refusing to participate, but asking for too little time might lead to serious limitations in the quality and quantity of data provided by respondents” (Harvey, 2011, p. 436). Indeed, in my early rounds of initial interview invitations, I asked participants for 45-60 minutes of their time. Concerned that I was not getting many responses, I began asking instead for 30-45 minutes of time in subsequent invitation emails. In follow-up emails, I stated: “Even a 20-minute call will go a long way to assist me.” On average, the interviews lasted about 20-30 minutes, ranging from about 10 minutes to 1 hour. Many of the participants were rushed and could not allocate more than 20-30 minutes for the interview. In other cases, the participants’ memories were so limited that a lengthy interview was not possible, despite the willingness of the participants to speak with me. Another issue predicted by Harvey (2011) was deciding whether to record the interview, noting that some elites may prefer not to be recorded. Despite my best intentions, most of the interviews were not recorded. A few participants agreed to call into a web-conference program so that I could record and transcribe the interview. In these cases, I transcribed the entire interviews, first using an automatic transcription software program (with about 75% accuracy) and second by hand to correct errors. Most of the participants, however, preferred me simply to call their office or cell phone. Asking the participants to go through additional procedures so that 129 I could record the call seemed unreasonable. As such, I typed detailed notes during the interview instead of recording. Immediately after the interview, I reviewed my notes and typed additional thoughts to preserve the most important observations from the interview. A few of the participants also expressed discomfort with recording the interview, so simply making the decision not to record was a reasonable compromise to proceed with the interview. These realities of interviewing politicians may prove to be limitations of the study, though the participants’ responses, however short, still provided useful data for analysis. As Harvey (2011) reasoned, “some data from elite respondents was better than no data” (p. 435). Data Analysis The analytic techniques employed were aimed at explanation building (Yin, 2009), that is, explaining how and why policymakers made their decisions regarding aid eligibility for justice-involved students. Table 8 displays the analytical framework used to analyze the data against the conceptual model (Hearn, McLendon, & Linthicum, 2017), which I created following Ness’ (2008) analytical framework. For each of the four categories of forces proposed in the conceptual model, I wrote guiding questions to help me analyze the data. Procedurally, I analyzed each case separately. After data collection was complete for a given state (e.g., interviews, historical records, other background research), I engaged in a deep review of the data to understand the events that occurred. Once I felt I had a grasp of the story, I wrote a chronological account of the policy events in the form of a case study, or state profile (see Results; see Ingle & Petroff, 2013; Ness, 2008). I weaved interview testimony, policy language, and evidence from other sources to tell the story as to how and why policymakers created financial aid programs with eligibility rules for justice-involved students. With the events and context laid out, I then analyzed the case by asking myself the questions listed in the 130 analytical framework (Table 8). These questions pointed me directly to the presence (or absence) of the factors from the conceptual model, thus allowing me to offer theoretically-based explanations for how and why the policies were adopted. For each state, I replicated this process of collecting source materials, synthesizing the data into a case, and analyzing the case against the analytical framework. Following the individual case analyses, I conducted a cross-case analysis by comparing how a single factor influenced policy adoption in multiple cases, for example (Yin, 2009). In these cases, I noted similarities or differences in how the same factor affected policy adoption. After this round of analysis, I was left with two cases that could not be explained by the conceptual model. After isolating the foreign causal factors, I searched the policy literature for concepts and theories that offered better explanations of the policy adoption events. Using a combination of the case evidence and the research literature, I presented an argument for why two new factors should be added to the conceptual model. Lastly, I offered brief analysis of two ancillary findings that did not warrant adding to the model. 131 Table 8: Analytical Framework Forces on State Policy Adoption Analytical Questions Socioeconomic Context 1. To what extent do state demographic factors (or changes in those factors over time) influence the adoption of these eligibility rules? Organizational and Policy Context Politico- Institutional Context 2. Was information about a state’s prison population available to policy actors, and to what extent was that influential? 3. Was information about the educational attainment or postsecondary enrollment rates of justice-involved people available to policy actors, and to what extent was that influential? 4. Were there estimates of the sizes of justice-involved student populations in a state, and to what extent were they considered? 5. Was information about state crime rates available to policy actors, and to what extent was that influential? 6. At the time, what was the climate like for justice-involved people (i.e., public opinion, philosophies on corrections, etc.). 1. What role did state agencies (e.g., higher education departments, financial aid commissions) play in the policy adoption process? 2. Did the adoption of eligibility rules in financial aid policies follow the adoption of similar rules in other policies (i.e., housing, welfare)? 3. Did policy actors consider the economic consequences of denying justice-involved students state aid? 4. Did policy actors consider how such eligibility requirements might affect postsecondary enrollment? 5. Did policy actors have data on how many justice-involved students were enrolled in higher education, or how much scholarship funding was being spent on justice-involved students? 1. To what extent did political ideology or partisanship affect policy decisions? 2. What attitudes did policy actors hold towards justice-involved students, and to what extent did that affect policy decisions? 3. Did individual policy actors (or coalitions of policy actors) exert significant influence in adopting these eligibility rules? 4. Did policy makers have projections on the impact of such rules? 5. Did governors weigh in on this issue? 6. How might the election cycle have affected policy decisions? 7. What interest groups were involved in these policy decisions, and to what extent did they have influence over the policy decisions? 8. Who was for and who was against these eligibility rules? 9. Did higher education institutions within the state weigh on this issue of eligibility rules? 132 Table 8 (cont’d) Policy Diffusion Context Quality 1. To what extent did policy actors borrow these rules from federal-level policies? 2. To what extent did policy actors borrow these rules from other states’ policies? 3. To what extent did policy actors borrow from other policies within the state? 4. What motivated policy actors to borrow these rules? 5. How did policy actors learn of other states’ policies? 6. What pressures from outside the state influenced the adoption of these eligibility rules? Following best practices in qualitative methodology, several steps were taken to ensure high degrees of quality, credibility, rigor, transparency, and ethical practice in the study design, data collection, and analysis (Tracy, 2010). In the study design stage, the conceptual model formed the backbone of the study. It explicitly informed the research question, interview questions, and the analysis (see Table 8), which are characteristics of rigorous qualitative research (Tracy, 2010). The interview protocols were modelled off those used in other successful qualitative studies of policy adoption (Cohen-Vogel, Ingle, Levine, & Spence, 2008; Ness, 2008), closely tailored to the conceptual model used in this study. The selection of case study methodology was also a strategic design decision because the in-depth investigation of a single phenomenon using multiple sources of information (i.e., interviews, historical records) lends itself to higher credibility (Baxter & Jack, 2008), a natural form of data triangulation (Tracy, 2010). Great care was used in selecting cases, contacting and interviewing participants, and safeguarding their data. First, I had to identify cases from the Phase I study that were feasible to study based on the year of the policy adoption event and the likely availability of the key policy 133 actors to be interviewed. I did not reach out to policy actors who I thought would be unduly burdened by my request, and I only invited participants whose testimony had potential to offer valuable insights related to the research question. Willing participants consented to be interviewed after reading a consent agreement. Participants were promised confidentiality,177 and all recordings and notes generated from interviews exist only in electronic formats within password-protected devices and programs. Transcripts of interviews were stripped of identifying information. On the condition of implementing these safeguards, the Michigan State University Institutional Review Board approved the study as exempt. Transparency is a characteristic of good qualitative research (Tracy, 2010). Unlike in quantitative research, where readers can access a researcher’s data and statistical coding to replicate analytical steps, qualitative researchers must describe the steps of the process of data collection and analysis, including when things did not go according to plan. This has been called the “chain of evidence” (Gibbert & Ruigrok, 2010, p. 713). In the sections above, I described in detail my process for data collection and analysis. I was particularly transparent about the challenges I faced in interviewing policymakers and the limitations that may result. Throughout, I am honest about this study’s limitations, and I show readers how I draw my conclusions. Above, I was also transparent about my personal positions on the topic and how I intended to prevent my viewpoints from introducing undo bias into my analysis. This process of assessing personal strengths, shortcomings, motivations, and biases is called self-reflexivity (Tracy, 2010). Finally, I shared my full interview protocols and template recruitment email in the appendix, giving readers direct insight into how I communicated with participants. 177 Except for one participant, who consented to being identified. 134 Finally, credibility is a concept that invokes questions about how trustworthy the data are and how plausible the analysis is, called face validity (Tracy, 2010). As the author, I am responsible for persuading readers that I rigorously collected and analyzed data and that the conclusions I drew from my analysis are sufficiently trustworthy, such that they can be used as the basis for future research and policy development. In the presentation of the case studies and in the analysis, I used a strategy commonly known as thick description. I provided detailed descriptions of the case contexts plus verbatim text from interviews and official legislative records to demonstrate exactly the points that are relevant to the research questions. Much of the time, the texts that I cited are explicit and leave little room for alternate interpretations. In other cases, however, there are “hidden assumptions and meanings” in the texts that required reading between the lines and making reasonable interpretations based on the combination of available data (Tracy, 2010, p. 843). In these cases, I am intentional in offering the thickest possible description to support my claims, and I am careful not to overextend my interpretations beyond credibility. The combination of these concerted steps builds the case that this study meets acceptable standards of quality. Ultimately, it is up to readers to determine such for themselves and to offer substantive critiques of this study’s quality. Limitations Because the case studies of Phase II are drawn from the Phase I study sample, the Phase II study is limited for the same reasons as described in the Phase I limitations section. Only currently active, undergraduate grant programs with rules for justice-involved students are studied here. Programs that are no longer available to students are not explored, though they may have relevance in future research. For example, it may be possible that a now-defunct program 135 once had a rule for justice-involved students that was later removed (before the program went defunct). Though I am unaware of any such circumstances, it would be interesting to learn why legislators removed the rule. Another possible situation that is not explored is the “non-adopter” state, or a state that considered adding a rule for justice-involved students to a grant program but ultimately did not. Studying non-adopters can give unique insight into why state legislators make their choices (see Ingle, Cohen‐Vogel, & Hughes, 2007). In future research, case studies of these unique situations might be fruitful. One challenge of this study was the availability of sources. As described above, many factors affected my final sample of policy and interview data, including: ability to identify and contact policymakers, availability and willingness of policymaker to be interviewed, availability and accessibility of legislative records, and more. Similarly, the short length of the interviews and the inability to record the interviews possibly affected the depth of the data collected. Perhaps most importantly, the memories of the participants presented limitations to the data, a problem also experienced by Cohen-Vogel, Ingle, Levine, and Spence (2008). In some cases, participants were asked about policy decisions from over two decades ago, and many participants reported limited memories of the specific policy issues that I asked about. Even for those involved in more recent policy decisions, memories were still hampered by time. In my analysis, I was careful not to draw conclusions based on information that was not confidently corroborated. In some cases, as will be shown, I could not answer the research question as to how and why legislators in a particular state took certain policy actions, even after collecting records and interviewing one or more policy actors. As with all historical research, a challenge of this study was “evaluating the reliability and usefulness of [my] sources” (Rampolla, 2012, p. 10). Documents and interview sources can 136 be incorrect, incomplete, contradictory, or even intentionally misleading, particularly if there are political motivations to be misleading (Rampolla, 2012). To mitigate these problems, I compared sources for verification and identified biases in the sources in my interpretations. A final limitation involves this study’s use of the conceptual model. The conceptual model of state policy innovation and diffusion in higher education was derived from decades of research addressing many types of state higher education policies. When examining only one type of policy, the model creators acknowledged that the various factors affecting policy adoption can be variable or stable depending on the conditions (Hearn, McLendon, & Linthicum, 2017). In this case, the policy adoption events under study were particularly narrow, which may limit the applicability of the model. Results Findings from the multiple case study are presented in this section. The research question was: How and why did state policymakers in some states adopt eligibility rules for justice- involved students in financial aid grant programs? Results are presented as state profiles or case studies. Like the profiles from Phase I, they offer chronological, historical accounts of the policies under study, but differently, they go deeper into accounts of how and why the policies were adopted. In other words, presented here are the stories of how justice-involved students lost eligibility for grant aid. All the source material presented in the profiles (e.g., quotes from interviews or debate transcripts, policy language, background context, etc.) build explanations for the state policy adoption events. Analysis of the factors that led to the policy adoption events is reserved for the subsequent Discussion section The first five profiles (Pennsylvania, Michigan, Illinois, Georgia, and Indiana) offer the most compelling explanations. In these cases, the available data provided a more complete story 137 of the context and conditions that led policymakers to deny aid to justice-involved students. In the Discussion section that follows the case studies, analysis is focused on these five states. The results from the remaining five state case studies (Oklahoma Kentucky, Delaware, Wyoming, and Utah) were less complete and less compelling. As such, they are shorter and carry less weight in the discussion section. Pennsylvania Since about April 1988, incarcerated students in the Commonwealth of Pennsylvania have not been eligible for any state financial aid grants. This eligibility rule resulted from nearly twenty years of statutory changes, lawsuits, and rulemaking over the rights of justice-involved students to receive state grants. Because of the age of these events, I relied on historical records to investigate how and why incarcerated students in Pennsylvania lost eligibility for state financial aid. The modern history of financial aid in Pennsylvania began in 1963 with the creation of the Pennsylvania Higher Education Assistance Authority (PHEAA, 2014).178 Originally a student loan provider, PHEAA would soon come to administer the State Scholarship Program, created in 1965 and signed into law in early 1966.179 The State Scholarship was a “broad-scale” program, designed to guarantee that the most able students from all sectors of the Commonwealth, the most needy students and students with the capability to successfully complete postsecondary educational programs, and deserving postsecondary students are given the opportunity to continue their program of self-improvement in an institution of higher learning of their choice.180 178 Pa. Act of Aug. 7, 1963 (P. L. 549, No. 290). 179 Pa. Act of Jan. 25, 1966 (P.L. 1546, No. 541). 180 Id. § 1. 138 Pennsylvania high school graduates, under 21 years old, who immediately enrolled in college were eligible for the scholarship so long as they “satisfactorily meet the qualifications of ‘financial need,’ character and academic promise, as well as academic achievement” (emphasis added).181 Though “character” is not defined here, contemporaries likely intended that past misconduct, including crimes, would be disqualifying (Swisher, 2008).182 Perhaps the ambiguity of this clause caused difficulties, because in 1969, PHEAA and the state legislature would promulgate rules that better defined who would not be eligible for a State Scholarship. First, in February 1969, PHEAA “adopted [a] regulation to automatically deny loan and grant aid to any student who was convicted of a felony or misdemeanor involving moral turpitude. If charges were pending, aid was withheld until a verdict was reached” (Smith, 1990, p.1). It is not clear what prompted this rule adoption at the state agency level, but it preempted the rule’s adoption in state statute. In December of 1969, a series of amendments to the State Scholarship were signed into law, including: The agency may deny all forms of financial assistance to any student: (l) Who is convicted by any court of record of a criminal offense which was committed after the effective date of this act which, under the laws of the United States or Pennsylvania, would constitute a misdemeanor involving moral turpitude or a felony.183 The law also denied eligibility to students who were convicted of crimes related to campus disruptions and students who were dismissed or expelled from college. In addition, higher education institutions were required to report convicted or expelled students to PHEAA or face 181 Id. § 4. 182 “Good moral character” requirements have existed in American law since the 18th Century and can be found in citizenship laws, occupational licensing requirements, state financial aid programs, and much more. Much has been written about the “good moral character” requirements of the American bar for attorneys (see Lapp, 2012; Rhode, 1985; Swisher, 2008). 183 Pa. Act of Dec. 18, 1969 (P.L. 171, No. 169) § 2. 139 losing their status as an approved institution for scholarship recipients.184 The “moral turpitude” clause caused confusion, and the new reporting requirements drew ire from colleges. Haverford College v. Reeher (1971). In 1971, a class action lawsuit was filed in the United States District Court for the Eastern District of Pennsylvania against PHEAA and its executive director, Kenneth R. Reeher, over the 1969 reporting requirements and the eligibility rules related to criminal history.185 The plaintiffs included: Haverford College, representing institutions that executed an agreement with the state to abide by the new rules; Goddard College, representing institutions that refused to execute the agreement and whose students therefore lost eligibility for state financial aid; a group of students who lost eligibility for aid because their colleges refused to execute the agreement; a group of students at colleges that executed the reporting requirements and therefore had to disclose criminal history; and a group of students who refused to disclose criminal history under the new requirements and therefore lost eligibility for state aid. The plaintiffs sued on the following grounds. First, the plaintiffs argued that the new eligibility rules regarding criminal history (felonies or misdemeanors of moral turpitude),186 dismissal due to campus disruptions,187 and convictions related to campus disruptions188 were unconstitutionally vague and that such vagueness resulted in a violation of students’ First Amendment rights. The court agreed that the 184 Id. 185 Corporation of Haverford College v. Reeher, 329 F. Supp. 1196 (E.D. Pa. 1971). 186 “Who is convicted by any court of record of a criminal offense which was committed after the effective date of this act which, under the laws of the United States or Pennsylvania, would constitute a misdemeanor involving moral turpitude or a felony” 187 “Who has been expelled, dismissed or denied enrollment by an approved institution of higher learning for refusal to obey, after the effective date of this act, a lawful regulation or order of any institution of higher education, which refusal, in the opinion of the institution, contributed to a disruption of the activities, administration or classes of such institution; 188 “Who has been convicted in any court of record, of any offense committed in the course of disturbing, interfering with or preventing, or in an attempt to disturb, interfere with or prevent the orderly conduct of the activities, administration or classes of an institution of higher education.” 140 latter two rules related to campus disruptions were so vague that a student could not interpret which activities would result in discipline (i.e., dismissal or loss of aid). Without a strict standard of certainty, any undesirable behavior might be considered disruptive and therefore was punishable by university officials. The effect was that students would be deterred from committing constitutionally protected behaviors (i.e., campus protests) for fear of punishment by the university or PHEAA. The court also agreed that losing aid eligibility or being expelled from college were substantial punishments for which students were owed relief. For these reasons, the court struck these two eligibility rules as unconstitutionally vague and enjoined PHEAA from enforcing them. The court also considered the vagueness of the criminal history eligibility rule, particularly the phrase “misdemeanor involving moral turpitude.” At issue is whether a student could know what behaviors would constitute such a misdemeanor so that he/she could avoid the behavior and therefore avoid punishment. They concluded: “the term ‘moral turpitude’ tells us very little about what misdemeanors trigger the loss of financial aid eligibility.”189 As a result, the court struck down this regulation, adding: “If the state insists on legislating morality, we will insist at least that it spell out its moral code.”190 Second, plaintiffs challenged the three rules on the grounds of being overbroad, meaning the government’s attempt to control people’s behavior occurs in such a wide sweep that it infringes on people’s constitutional rights (i.e., freedom of speech). Here again, the court found that the two sections on campus disruption were unconstitutionally overbroad in that a student may be deterred from exercising rights of free speech for fear of punishment. However, the first rule on criminal history “punishes only students who have committed certain crimes. Since 189 Corporation of Haverford College v. Reeher, 329 F. Supp. 1196 (E.D. Pa. 1971). 190 Id. 141 criminal conduct is not protected by the Constitution, there is no overbreadth problem there.”191 According to the court, it was permissible for the state to deny financial aid to students convicted of felonies. Finally, plaintiffs alleged that requiring institutions to turn over records of students who had felony convictions to the state was an unconstitutional infringement on students’ search and seizure rights. The court found that the plaintiffs did not sufficiently present facts on this claim, so the court did not decide on this point. In summary, the two statutory provisions related to punishing students for conduct involving campus disruptions were struck down as unconstitutionally vague and overbroad, and the section relating to a “misdemeanor involving moral turpitude” was struck down as unconstitutionally vague. Colleges that lost their status as approved institutions were restored, students’ eligibility was restored, and PHEAA was enjoined from enforcing the unconstitutional sections as well as any related regulations.192 What remained intact, however, was the provision allowing PHEAA to deny aid to students convicted of felonies and the provision requiring colleges to report those students to PHEAA. Haverford College v. Reeher (1972). Only a few months after the district court issued its consent decree in Cooperation of Haverford College v. Reeher (1971), Haverford College again sued Reeher and PHEAA. This time, Haverford College claimed that PHEAA violated the consent decree.193 A student there, Henry D. Kelly, was under indictment for refusing to register for the draft, a felony charge. PHEAA withheld Kelly’s aid in accordance with its Regulation 191 Id. 192 Corporation of Haverford College v. Reeher, 333 F. Supp. 450 (E.D. Pa. 1971). 193 Corporation of Haverford College v. Reeher, 53 F.R.D. 374 (E.D. Pa. 1972). 142 105,194 which allowed for withholding aid from any student arrested or charged with a crime. Haverford College argued that PHEAA was enjoined from taking such action under the consent decree, but PHEAA argued that Regulation 105’s “satisfactory character” clause permitted them to take such action (Smith, 1990). In other words, PHEAA contended that someone charged with a felony for refusing to register for the draft had unsatisfactory character and was thereby disqualifiable for aid. The court sided with PHEAA, but for different reasons. Since the court upheld PHEAA’s authority to deny aid to a student convicted of a felony, they found it reasonable that PHEAA could withhold aid to a student charged with a felony, until the student’s felony status was finalized. However, “had Kelly’s crime been other than a felony, the court would likely have ruled against PHEAA” (Smith, 1990, p.1). Out of caution, beginning in the 1972-1973 academic year, PHEAA stopped this practice of withholding aid to students with pending criminal charges; instead, they only inquired about students with past convictions (Smith, 1990). With the first two lawsuits behind them, PHEAA was able to address the court’s orders by revising their agency regulations for state aid. In March 1972, PHEAA proposed to replace Regulation 105 with Regulation 107.195 Notably, PHEAA struck the unconstitutionally vague “misdemeanors of moral turpitude” and instead listed 22 categories of offenses (misdemeanors or felonies) that would result in the loss of aid eligibility, ranging from adultery, fornication and bastardy, corrupting the morals of children, to desecrating the flag of the U.S. or Pennsylvania, all clearly crimes of “moral turpitude.”196 Consistent with the court’s approval, all felony 194 I was unable to obtain a complete copy of Regulation 105, despite making requests to PHEAA and law libraries in Pennsylvania. Relevant excerpts of Regulation 105 are provided in footnotes of the court case. 195 2 Pa. B. 506 (March 25, 1972). 196 Id. This list did not last long, however. By 1978, the list was replaced with new regulations: “(1) A criminal offense which under the laws of the United States or any state constitutes a felony. (2) A criminal offense which under the laws of the Commonwealth constitutes murder of the first degree, felony of the first degree, felony of the second degree, felony of the third degree, misdemeanor of the first degree, misdemeanor of the second degree, or misdemeanor of the 143 convictions remained grounds for disqualification. In addition, PHEAA developed a new procedure for hearing appeals of decisions on student eligibility through its Administrative Review Committee. Carbonaro v. Reeher (1975). In Cooperation of Haverford College v. Reeher (1971), the district court upheld the state’s authority through the legislature and PHEAA to deny aid to students convicted of crimes, especially felonies. This would be the issue in PHEAA’s next lawsuit.197 Students who were convicted of felonies and who lost aid eligibility under PHEAA’s new satisfactory character regulations sued PHEAA on the grounds that discriminating against people with felony convictions violated their Fourteenth Amendment rights to equal protection under the law. The court analyzed the extent to which denying financial aid to people with felony convictions had a legitimate state purpose. The court quickly rejected the state’s first rationale: The state argues that, as a political reality, the legislature would not establish any assistance program at all unless it contained the provisions allowing the agency to deny felons aid. We reject this second basis, since political considerations do not constitute a legitimate state purpose. If they did, it would be close to impossible to challenge any statute as violative of equal protection since the state could almost always assert that the classification was necessary for passage of the state program.”198 In other words, the political motive of the time to deny felons aid was an insufficient rationale for doing so. Second, the state made a financial argument for denying felons aid with which the court agreed: “The state contends that the classification is reasonable because the state has only finite third degree. (3) A violation of section 13 of the Controlled Substance, Drug, Device and Cosmetic Act (35 P. S. § 780 - 113), except subsection (a)(31) thereof.” See 8 Pa. B. 2079 (July 22, 1978). 197 Carbonaro v. Reeher, 392 F. Supp. 753 (E.D. Pa. 1975). 198 Id. 144 resources and wishes to maximize the potential gain from the available funds by providing assistance only to those students with ‘satisfactory character.’”199 The court returned to the original language of the 1966 statute, which states that only “deserving” students should be given the aid and that students must “satisfactorily meet the qualifications of ‘financial need,’ character and academic promise.” Character, according to the court, is distinct from the other academic and financial requirements, and a felony conviction may be indicative of unsatisfactory character. The state, the court concluded, was justified in requiring felons to prove their satisfactory character to PHEAA, which the court noted many felons had done successfully in order to get aid. To reiterate, we hold that the statutory and administrative program whereby Pennsylvania places the burden on former felons to prove that they are of satisfactory character in order to receive state financial assistance for postsecondary education does not violate the equal protection clause of the Fourteenth Amendment. The felon classification bears a rational relationship to the legitimate state purpose of assuring that only responsible citizens receive state aid.200 Once again, PHEAA prevailed in court, but the agency still drew lessons from these cases. Seeking the most stable legal footing possible to avoid future litigation, PHEAA continued to make regulatory changes by loosening its requirements on justice-involved students. In the months after the 1975 lawsuit, PHEAA began approving all applicants with misdemeanors and first-degree felonies for aid “based on (1) the experience of [the Administrative Review Committee] approving such cases [on appeal] and (2) a question as to whether courts would uphold our denial of aid based on a misdemeanor conviction” (Smith, 1990). In 1978, PHEAA 199 Id. 200 Id. 145 eliminated the character test in the loan program so that any felon denied grant aid could still get a loan.201 In the 1980s, PHEAA further narrowed how it defined unsatisfactory character. In 1980, the Administrative Review Committee began automatically reviewing any felon applicants that initial staff could not agree on, and in 1985, staff were permitted to approve nearly all felons so long as they paid for their first term of college out pocket, which apparently was a demonstration of deservingness for future aid (Smith, 1990). By 1988, data showed that almost all students with felony convictions were eventually approved for aid after exhausting their appeal options. From then on, including still today, PHEAA assumes “that once an applicant is released from incarceration, he will be presumed to be of satisfactory character and eligible for aid. Therefore, only currently incarcerated applicants are now denied grant aid under the satisfactory character provision of the law” (Smith, 1990). Today, the provisions in state statutes that were struck down in Cooperation of Haverford v. Reeher (1971) remain on the books but are enjoined from being enforced. The legislature never repealed them.202 Instead, PHEAA’s regulations were amended many times between 1971 and the early 1990s, with the final regulation stating: If a State grant applicant is incarcerated, which shall be interpreted to mean confinement in a prison but to exclude residence in a ‘‘halfway house’’ under a so-called prerelease program, the applicant will not be eligible for State grant aid until the applicant has been released from incarceration.203 201 8 Pa. B. 3085 (November 11, 1985). 202 24 P.S. § 5158.2. 203 22 Pa. Code § 121.6(b). 146 In addition, as described in Phase I, most of PHEAA’s application forms and grant handbooks now indicate that incarcerated students, only, will be denied aid (see PHEAA, 2017). Denying aid to justice-involved students in Pennsylvania proved to be a complicated task. It is not clear why legislators in 1966 and 1969 created the vaguely-worded character requirements or what the requirements meant, but they likely did not foresee the ensuing lawsuits and perpetual regulatory changes that would occur over the next two decades. What once was a broad ban on all types of student financial aid (e.g., grants and loans) to all justice-involved students morphed into a ban on just grant aid to currently incarcerated students, only. Michigan Legislators in Michigan began creating financial aid programs in the mid-1960s, and like the 1966 Pennsylvania State Scholarship, they included rules about character; both the 1964 Michigan Competitive Scholarship and the 1966 Tuition Grant Program required students to be of “good moral character.” Though without the drama of the lawsuits and the back-and-forth rulemaking that happened in Pennsylvania, state legislators in Michigan came to the same conclusion that incarcerated students, only, should be prohibited from receiving state scholarships. In 1981, legislators passed a law that tinkered with a number of student eligibility rules across four state grant programs. This section relies primarily on published analyses of senate bills to understand how and why incarcerated students lost eligibility for some scholarship programs in Michigan. By 1978, the state of Michigan funded four financial aid grant programs for college students: the 1964 Michigan Competitive Scholarship,204 the 1966 Tuition Grant Program,205 the 204 Mich. 1964 P.A. 208. 205 Mich. 1966 P.A. 313. 147 1976 Legislative Merit Award program,206 and the 1978 Differential Grant program.207 Each having been created in different years and under separate acts of the Michigan legislature, a state audit in 1979 found inconsistencies in the basic eligibility requirements across the programs: A recent audit of the state’s higher education tuition grants and scholarships programs conducted by the Auditor General revealed inconsistencies among the programs, both in their administration and in the statutes governing them. The Tuition Grants Program, the Michigan Competitive Scholarship Program, the Differential Grants Program, and the Legislative Merit Awards Program are governed by different acts. While there is some agreement among the acts in terms of eligibility requirements and funding provisions they contain, it has been argued that certain discrepancies among the acts confuse or obscure the general policy of the higher education assistance program. (Michigan Senate Analysis Section, 1980a, p. 1). Discrepancies between the programs included: the number of required credit hours (full-time, part-time), the definition of state resident, whether a scholarship recipient could pursue a degree in theology or religion, academic requirements for maintaining a scholarship, and others. The audit, conducted by Michigan Auditor General Albert Lee, encompassed three state agencies – Student Financial Assistance Services, Michigan Higher Education Assistance Authority, and Michigan Higher Education Student Loan Authority – and contained 63 recommendations for improving Michigan’s loans and grants programs (Lee, 1979). At the recommendation of the State Auditor, Senator Jerome Hart (D) and other legislators sought to 206 Mich. 1976 P.A. 228. 207 Mich. 1978 P.A. 105. 148 iron out the inconsistencies in the grant programs by introducing program amendments in Senate bills 1275, 1276, 1277, and 1278 in the fall of 1980.208 The eligibility rule requiring “good moral character” in the Michigan Tuition Grant and the Michigan Competitive Scholarship also apparently troubled the bill sponsors. In the original versions of bills 1275 and 1276 introduced on September 23, 1980, the good moral character provisions were proposed to be struck from both programs. However, later in the policymaking process, a provision was added to replace the character rule. First, according to an October 8, 1980 entry in the senate journal, an amendment to Bill 1276 inserted a rule in the Competitive Scholarship that an eligible student “is not incarcerated in a corrections institution.” In the November 10 Senate Analysis, this change was believed to “sharpen the original legislative intent of the ‘good moral character’ provision, which the bill would delete” (Michigan Senate Analysis Section, 1980a, p. 3). Thus, this suggests that “good moral character” was intended to disqualify students with criminal histories, particularly incarcerated ones. It was also noted that “a similar amendment may be needed for Senate Bill 1275, which also would delete the ‘good moral character’ provision” (emphasis original, Michigan Senate Analysis Section, 1980a, p. 3). Indeed, this would happen soon after. On December 4, the same incarceration rule language was inserted into Bill 1275 for the Tuition Grant program. Once Bills 1275 and 1276 matched on this point, the updated December 22 Senate Analysis repeated the claim that the new incarceration rule “sharpens the original intent of the ‘good moral character’ provision” (Michigan Senate Analysis Section, 1980b, p.3). After clearing the House and Senate on December 10, 1980, the bills were presented to Governor William Milliken on January 8, 1981 208 Mich. 1980 SB 1275; Michigan Tuition Grants; Mich. 1980 SB 1276; Michigan Competitive Scholarship; Mich. 1980 SB 1277; Differential Grants Program; Mich. 1980 SB 1278; Legislative Merit Awards Program. 149 and approved on January 22, 1981. The incarceration rules remain in force today in the Michigan Competitive Scholarship209 and in the Michigan Tuition Grant program.210 Ironically, these amendments fell short of achieving consistency across the four programs, as the incarceration rule was not added to the Differential Grant or the Legislative Merit Award. Perhaps because they did not originally have “good moral character” rules, the legislators in 1980 did not think it necessary to add the new incarceration rules to these to programs. Regardless, neither program survives today. Though both still exist in state statutes today, the Differential Grant program was phased out in 1985,211 and the Legislative Merit Award was phased out around 1983. Illinois Like Pennsylvania and Michigan, the current ban on state financial aid to incarcerated students in Illinois was decided in the late 1980s. Likewise, this section relies on historical documents to answer the research questions. Two sources were especially valuable: a detailed report of the history of Illinois state scholarship programs from 1957 to 1982 (Eber, 1982) and transcripts of legislative floor debates from 1989. The two policy issues that affected justice- involved students in Illinois are described next in turn: an eligibility rule about “good moral character” and an explicit statutory ban on aid to incarcerated students. Good moral character. In 1955, Illinois Governor William Stratton created the Illinois Higher Education Commission (IHEC) to study the condition of higher education in Illinois. In 1957, the commission issued its report, which included a plan for a new state scholarship (IHEC, 1957). At least two sets of conditions drove the need for a new scholarship. First, college 209 Mich. Compiled Laws 390.974 (2017). 210 MCL 390.993 (2017). 211 Mich. 1985 P.A. 127 § 19. 150 enrollments were expected to increase sharply over the next two decades, pushing public colleges to capacity. Private colleges, however, still had some capacity, and a new scholarship program was expected to divert some of the flow of new students to the private colleges. Second, Illinois and the nation faced shortages of scientists, teachers, leaders, and other experts. Despite the influx in enrollment, there were still too many “Illinois high school graduates of superior capacity [who] did not go to college for financial reasons” (IHEC, 1957, p. 20). Thus, IHEC proposed a need- and merit-based scholarship that would support the training of more Illinoisans, which the legislature would adopt as the State Scholarship Act of 1957. 212 IHEC proposed specific eligibility requirements for the scholarship, including a “good moral character” requirement. IHEC’s only explanation for the requirement was rather uninformative: “For obvious reasons, good moral character should be a condition of eligibility” (IHEC, 1957, p. 158). Nothing in the report indicated where this “obvious” rule came from, but Eber (1982) points to one explanation. According to Eber (1982), “the IHEC patterned the [State Scholarship Program] after a recently created California program” (p. 27). Indeed, California legislators created the Competitive Scholarship in 1955, aptly named for the competitive examinations (or the SAT) that applicants had to take for qualification.213 Applicants also had to demonstrate financial need and “high moral character, good citizenship, and dedication to American ideals.”214 If IHEC did model the State Scholarship off the California Competitive Scholarship, perhaps this is where the good moral character requirement came from. 212 Act of June 21, 1957, 1957 Ill. Laws 855-861. By 1971, the State Scholarship was unfunded, but still today, the most academically strong students from across Illinois are designated as State Scholars by the Illinois Student Assistance Commission. 213 Cal. Stats. 1955, ch. 1846. 214 Id. at § 21702(e). 151 Regardless of where it came from, the good moral character requirement was problematic to implement: it “has been historically troubling for the [Illinois State Scholarship Commission]” (Eber, 1982, p.22). In practice, high school principals determined who had good moral character and had to document cases when students were disqualified based on character. The commission had to clarify parameters for character occasionally, like in the 1960s when they decided that premarital pregnancy should not disqualify a student on moral grounds. According to meeting minutes, commission officials debated the legal standard of “good moral character” at least six times between 1957 and 1973, and the rule was recommended to be abolished several times between 1972 and 1975 (Eber, 1982). It was estimated that six to ten students each year were denied scholarships for moral reasons, though “no systematic study has been done of the reasons high schools have withheld moral character certifications to otherwise qualified State Scholar applicants” (Eber, 1982, p. 23). In 1964, the Illinois Board of Higher Education proposed in its Master Plan for Higher Education that more funding be allocated to the State Scholarship program, and in 1966, the Board went further by proposing a new financial need-based grant (Eber, 1982). In 1967, legislators amended the newly-named Higher Education Assistance Law by adding what is now called the Monetary Award Program (MAP), which is a need-based grant with no initial academic qualifications.215 As such, the eligibility requirements for the State Scholarship, including the good moral character provision, applied to MAP. Soon after, funding for the State Scholarship and the short-lived Upperclass Grant were phased out, and MAP became Illinois’ primary grant program for financially needy students (Eber, 1982), as it remains today. 215 Act of Aug. 3, 1967, 1967 Ill. Laws 2644, 2646. 152 The good moral character requirement continued to be challenging to interpret after the creation of the MAP grant, especially when it came to the question of whether incarcerated students should be disqualified on moral grounds. In 1971, the Illinois State Scholarship Commission (ISSC) Executive Director, Joseph Boyd, asked the Illinois Attorney General to issue an opinion on the matter.216 Mr. Boyd asked: Is there a legal definition of good moral character? Does a person assigned by the courts to any institution of correction automatically disqualify himself for an award as long as he is serving out his/her sentence? This matter has become an issue since a considerable number of public and private institutions in Illinois are now offering credit courses to those persons in Illinois institutions of correction. In March 1972, the Illinois Attorney General William Scott responded with a 12-page, unpublished opinion.217 On the first question, the attorney general stated that there was no consistent legal definition of good moral character and that each case needed to be decided on the facts. “It is the applicant’s actual conduct that is at issue, not the reputation of the applicant.”218 His best advice was that ISSC “must objectively determine whether an applicant is of good moral character by determining whether he measures up to the generally accepted moral standards currently prevalent within the State of Illinois.”219 On the second question, the Attorney General opined: “the fact than an applicant for a scholarship is incarcerated in a correctional institution does not automatically disqualify him for a scholarship.” Though a person who commits a felony was, at the time, not of good moral 216 State officials can seek the legal opinion of the attorney general on matters that pertain to their work, though attorney general opinions are not legally binding. 217 1972 Ill. Att’y Gen. Op. 434 (NP). 218 Id. at 5. 219 Id. at 6. 153 character, he suggested, other factors must be weighed when evaluating the character of a person after the fact, such as evidence of rehabilitation. With these somewhat dissatisfactory answers (Eber, 1982), it is not clear how ISSC handled the good moral character requirement after 1972. However, when ISSC issued its first program regulations in the 1986 Illinois Administrative Code, a person was said to be of good moral character “if the applicant will benefit from postsecondary instruction and is allowed to enroll at an approved postsecondary institution.”220 This definition left open the possibility that incarcerated students enrolled in prison higher education programs could receive state grants, which indeed happened so frequently that it caught the attention of state legislators. After legislators eliminated aid eligibility for incarcerated students in 1989, as described next, legislators eliminated the good moral character rule from the MAP statute in 1992 amidst a broad reorganization of the Higher Education Assistance Act.221 Nonetheless, the 1986 definition of good moral character remains today in the Illinois Student Assistance Commission regulations that apply to all grant programs.222 Aid for incarcerated students. By the end of 1989, incarcerated students would lose eligibility for scholarships administered by the Illinois State Scholarship Commission, later called the Illinois Student Assistance Commission. Transcripts of debates from the Illinois General Assembly provided a rich source of insight on how legislators viewed the issue of funding for incarcerated students enrolled in higher education. According to transcripts from June 16, 1989, the director of the Department of Corrections reported to the House Appropriations Committee that as much as $500,000 in state 220 23 Ill. Admin. Code pt. 1700.20 (1986). 221 P.A. 87-997, 1992 Ill. Laws 2128, 2137. 222 23 Ill. Admin. Code pt. 2700.20 (2018). 154 scholarship funding was going to students who were incarcerated in Illinois prisons (Illinois House of Representatives, June 16, 1989).223 In response, Representatives Richard Mulcahey (D) and Richard Mautino (D) introduced an amendment to Senate Bill 10 to eliminate the eligibility of prisoners to get state scholarships. The bill changed the definition of a qualified higher education institution in the Illinois Higher Education Act to exclude programs for incarcerated students.224 Thus, a college that offered higher education courses in a prison would no longer be eligible to accept state scholarship funds for those students. This sparked a lively debate about the merits of higher education for people involved in the criminal justice system and how they should be funded. After Rep. Mulcahey introduced the amendment, Rep. Mautino explained their position on the issue: “We think it’s unfair. We believe that the individuals who are not incarcerated, that our students should have the first opportunity for this money” (Illinois House of Representatives, June 16, 1989, p. 30).225 After another representative spoke briefly to share her support, Rep. William Shaw (D) posed two questions to the bill sponsors to set up his rebuttal: “In reference to the people that are receiving the grants – the scholarships that you’re talking about – how many 223 It is not clear if the $500,000 figure was an annual expenditure on aid to incarcerated students or if it represents multiple years of expenditures. 224 “For otherwise eligible educational organizations which provide academic programs for incarcerated students, the terms “institution of higher learning”, “qualified institutions”, and “institution” shall specifically exclude academic programs for incarcerated students.” See P.A. 86-1002, 1989 Ill. Laws 6809. 225 The Illinois legislature allocated a dollar amount each year for MAP grants. It was not an entitlement program, which means some eligible students may not have received the award if the allocated money ran out. Qualified incarcerated students, therefore, competed equally with non-incarcerated students for MAP awards. Yet, a 2012 report indicated that as of 2002, there was enough funding for all applicants who applied to get awards (Illinois Student Assistance Commission, 2012); this report, however, does not indicate whether that was true throughout MAP’s history. It may be possible that in 1989, when legislators were debating MAP grants for incarcerated students, there was not enough funding for all applicants. The testimony reported in this section seems to indicate that legislators believed incarcerated students were being awarded grants in lieu of non-incarcerated students. Today, if incarcerated students were restored to eligibility, they would most certainly compete with other students for awards. And there is not enough money for everyone. Since at least 2012, as many as half of qualified applicants do not receive a MAP due to lack of funding (ISAC, 2012). To encourage students to apply for aid early, the Illinois Student Assistance Commission announces a cut-off date for MAP awards based on projections of applicants and available funding. See https://www.isac.org/isac-gift-assistance-programs/map/suspense/ 155 of those people have come out of the prison system and wound back up in the prison system?” (p. 31). Rep. Mulcahey responded, “I have no idea. I couldn’t answer that.” Rep. Shaw continued, “So, you don’t really know… whether the scholarships are worthwhile or not? You don’t know… whether this is keeping people from committing additional crime?” Rep. Mulcahey responded, “I can’t document that. No.” With his questions answered, Rep. Shaw presented his counterargument to the amendment in a brief speech: Certainly, I was at the hearings where the director of Corrections pointed out that [incarcerated people] are getting these scholarships. And certainly I don’t think there’s anybody in here [who] is anymore harder on criminals than I am, but at the same time, unless we do something to educate these people in these institutions – and the director also testified that two-thirds of the population were at a possible sixth grade reading level – and if you just turn those people back on the streets without any formal education, you are sending those people out there to commit more crime. This Amendment should be… looked at and this Amendment should be defeated. Those people are entitled to some type of education. I don’t know whether this is the right way in terms of the scholarships, but those same individuals will be back out in your neighborhood and my neighborhood committing additional crime unless they are educated. And we should make some effort to educate those inmates – the ones that want to be educated. We got enough that don’t want to be educated, and they’re going to be out there anyway burglarizing your house. (pp. 31-32) 156 Unconvinced, Rep. William Black (R) first clarified that the amendment would not eliminate the existing education programs offered by the Department of Corrections – rather it would only eliminate scholarship funding – and then argued his support: In a period when we don’t even fund the merit recognition scholarships,226 I don’t think there are many of you on this floor who realize that inmates are not only eligible but are receiving, not only Pell grants for higher education courses, but Illinois State Scholarships Commission grants to pursue higher education courses. I think that shows a little bit of confusion of our priorities. (pp. 32-33) In response, Rep. Paul Williams (D) reiterated the problem of the under-education of people involved in the criminal justice and the likelihood of recidivism without additional education, and then he argued a new counterpoint: Think about it. If you save what meager amount we might spend on scholarships to inmates and you, at the same time, possibly increase the number of repeat offenders because of the fact that those who might have had an opportunity to change weren’t given that opportunity, what you end up doing is building more prisons, … you end up spending more money for prisons and more care of people who might have changed, if we had given them the opportunity. (p. 34) Upon hearing Rep. Williams characterize the amount of funding as “meager,” Rep. Charles Hartke (D) asked for clarification on the $500,000 amount before pledging his support: You know, we’re trying to … help these individuals, but we’ve got a lot of kids who have been good, abiding by the law and so forth and have been doing a good job. And we’re 226 Merit Recognition Scholarship, P.A. 83-1385, 1984 Ill. Laws 2711, which was periodically unfunded and has not been funded since 2004. 157 penalizing [them] by giving these grants to those individuals from the [correctional] institutions. (p. 35) At this point, the Speaker tried to close debate and call for a vote, but there were still questions about parts of the bill that did not relate to scholarships for prisoners. Then it returned to the scholarships. Rep. Monique Davis (D) asked again about the $500,000 before launching into a back-and-forth with Rep. Mulcahey. She quipped: Well, Representative Mulcahey, if you were talking to the relative of a victim of a prison inmate, who had been released on parole, do you think that victim would prefer that you use that money for imprisoning people or that you use that money to help people not to become prison inmates? (p. 37) Reps. Davis and Mulcahey argued back and forth, with Mulcahey reiterating that prisoners can still access education programs in prison without the scholarships. Rep. Davis argued: We’re talking about learning things that businesses require, that you have knowledge of before they will hire you. Sometimes… you just have to say I have a college diploma. So the intent of your Bill then, is to prevent the furthering of education for people who are incarcerated. (p. 38) “That’s not true, Representative Davis,” retorted Rep. Mulcahey. “Then tell me what [the amendment] is intended to do,” demanded Rep. Davis. Rep. Mulcahey explained one final time: It is intended to take $500,000 – not the programs that are set up by the Department of Corrections for the education of inmates, not the Pell grants, not the federal grants, not the guaranteed student loans that are available for everybody whether you’re on the inside or the outside – it simply is removing the Illinois State Scholarship funding for the inmates that are in prisons right now. And we’re going to give it to the kids around the 158 outside, who have parents who are lugging their lunch bucket to work every day, trying to send their children to school. Kids that can’t afford it, that have the ability to go. It’s going to go to those kids. (p. 38) Rep. Davis got the last word on the subject before the vote was cast: Representative, I’ll say this and I will cease my questions. I do believe that those families of victims would much prefer that the State of Illinois do everything that it can to prevent the incarceration of anyone. And if it means spending a few extra dollars from different pots, then let’s spend those few extra dollars from different pots. That is much more beneficial to this society, rather than to continue incarceration. (pp. 38-39) The amendment passed easily with 87 ayes, 9 nays, and 11 voting “present,” a form of abstention (Illinois House of Representatives, June 16, 1989). Ten days later, on June 26, the Senate unanimously passed Senate Bill 10 with no debate, and the bill was sent to the governor the following month. It would have seemed that the bill was sure to become law, but on September 6, 1989, Governor James Thompson (R) exercised his amendatory veto power, which is a veto that kills a bill unless legislators approve the governor’s amendments. The governor wanted Senate Bill 10 to take effect on July 1, 1990, rather than taking immediate effect, which he apparently did on other bills, too, in order not to disrupt departmental budgets across the state. In this case, it would allow the Department of Corrections about ten months to prepare for the elimination of the $500,000 that funded prison higher education programs coming from the Illinois State Scholarship Commission. The legislature had a few options: it could let the bill die by doing nothing, it could override the governor’s veto and pass it as is, or it could approve the governor’s amendment and extend the effective date. 159 On October 18, a motion was introduced in the Senate to override the governor’s veto on Senate Bill 10 (Illinois Senate, October 18, 1989). Much of the debate was spent trying to understand the governor’s veto and the actions they had to take next (to override the veto or not), but the substance of Bill 10 was also debated. Since the Senate approved the bill the first time unanimously with no debate, this was the first discussion of scholarships for incarcerated students in this chamber. Senator Patrick Welch (D) opened by explaining why the veto should be overridden: The governor altered, slightly, Senate Bill 10, but what he did was he altered it so that a group of individuals, who have in the past been receiving scholarship money, will continue to receive it for an additional nine months. The group that I’m referring to are those currently in prison. And what the Governor did was take part of the bill that says that we are not going to allow anyone in prison to continue to receive Need Based Monetary Awards [MAP Grants], issued by the Illinois Student Assistance Commission, to let them continue to [receive] it. Now it makes no sense to me that one of the criteria for receiving a Need Based Monetary Award is immediately met by someone in prison. A person in prison is usually very much in need, and so they meet that qualification right away. To allow them to continue to do this for another nine months is going to not only compound an injustice, it’s going to take money away from deserving students who don’t participate in this program, because there are not enough funds available. And so, what I’m asking to do is to override that specific recommendation of the governor, that this particular wrong be allowed to continue through the end of this year, and into half of next year. I’m asking that we stop this misspending of state funds right away. I would ask that we override the veto of the governor and implement this termination of sending money to 160 prisoners so that they can go to school and compete with other more deserving students, right away. (Illinois Senate, October 18, 1989, p. 95) Senator Adeline Geo-Karis (R) voiced her opposition to the override, saying: I think the governor’s statement in his amendatory veto is very good, because what he’s saying is you’re taking the money away from needy students… The other part is that all he’s asking, in addition, is that the date be changed to July 1, 1990, in order to give time to the Department of Corrections to implement the program in the other departments. I think it’s only fair. (p. 96) The confusion began to set in, and Senator Howard Brookins (D) asked Senator Welch to explain the bill again. Satisfied, he explained his opposition to the override: I know that, and I do not encourage us to [pander] and soft-peddle the people that have committed crimes and et cetera, but I know that the only way that we’re going to prevent these same people from coming out of the jails and creating other crimes is to give them an opportunity to improve themselves. I can’t go along with this. I think that we need not defeat this motion. I needed to let it go through as is, as has been recommended by the governor. (p. 97) Now several senators are lost, and Senator Welch had to explain how the scholarship funding flows to incarcerated students. A Senator Collins added: Now, I can go along with not giving them money, but I most certainly cannot go along with denying them the opportunity, if they qualify, to continue their education… So I support Senator Brookins and the governor, if that’s the case. (p. 99) Senator Collins’ statement did not help matters because it missed the point of the override motion. Senator Robert Kustra (R) jumped in: 161 I’m reading the veto, trying to figure this thing out myself, and it sounds to me like most of us are in agreement here. The governor agrees with Senator Welch that there should not be these Need Based Monetary Awards used for people who are in prison. But in his veto, he says that he doesn’t think we should do this until July 1, 1990, because it goofs up the Department of Corrections Educational program… Doesn’t seem to me like an unreasonable request to make… I would urge that we defeat Senator Welch’s motion here. (p. 99) This brought more questions about how the scholarship program works. After an exchange between Welch and Senator Arthur Berman (D), Berman landed somewhat equivocally on the side of opposing the override and suggested the whole bill needs reconsidered in the future: I understand Senator Welch’s concern that we’ve said we ought to have money for scholarships, and here part of it is going to kids in jail. Well, I’m not sure…that there’s anything wrong with that. In fact, to great extent, I think that it’s probably pretty well used money, if, in fact, we get that kid out of the life of crime. Now we’re faced with this bill in this posture… I’d probably vote against the bill. So, I think what I’m going to do today is vote with the governor to make it effective July 1 and look at this next spring because I’m not sure that what we’re doing makes sense. (pp. 101-102) Frustrated, Senator Denny Jacobs (D) tried to level with his colleagues: You can sure tell it’s getting late… I voted for this bill with 57 other senators. This bill passed 58 to nothing, and here we are just talking about one little date. I don’t care what you do with it, but I just think that we’re killing this baby. (p. 102) Finally, Senator Welch delivered an impassioned, if not exasperated, closing statement: 162 Well, Ladies and Gentlemen, let me just say that it’s interesting that we’re focusing on the rights of the individual in jail, as opposed to the rights of the individual who hasn’t committed a crime. If you have a poor individual who hasn’t committed a crime, and he applies for a scholarship, he’s thrown in a pool of competition with individuals who have committed crimes. Now, should we say to that individual who has led an upstanding life, hasn’t used drugs, hasn’t stolen a car, hasn’t done any of these other things, … are we saying to them, “Well, we’re not going to take that into consideration. You are now going to compete evenly [with people in jail] to see if you can get this scholarship. Now, you may be competing with some people in jail, but we want to be fair about this, everybody’s on the level playing field.” Well, I don’t think there should be a level playing field. I think that if we want to fund individuals, and rehabilitate them in the Department of Corrections, let’s be honest about it and given an appropriation to that Department… Let’s not mix our feelings about helping incarcerated students with those of helping students who want to go to college but can’t get the scholarships. We’ve talked about scholarship money time and time again, and obviously there isn’t enough. Why split what little money we have with individuals who’ve committed crimes? It doesn’t make sense to me, and I would move for an Aye vote to override the Governor’s veto. (pp. 103-104) Either Senator Welch was unpersuasive, or senators were still confused. The motion to override the governor’s amendatory veto failed by a narrow vote of 27 ayes to 30 nays (Illinois Senate, October 18, 1989). The legislators would have to use other maneuvers to pass their bill. The next day, in fact, Senator Welch yielded his position on the governor’s veto, “against [his] better judgment” (Illinois Senate, Oct. 19, 1989, p. 5). With no discussion, Senate Bill 10 – 163 as amended by the governor – passed in the Senate unanimously with 56 votes, a striking departure from the contentious divide from the day before. This meant the bill to eliminate aid eligibility for incarcerated students effective the next year was on its way to passage. On October 31, Representative Mulcahey filed a motion to accept the amended veto in the House, but a procedural rule relating to gubernatorial vetoes held it up. Thus, a new strategy, again, was needed to pass Senate Bill 10. On November 1, 1989, Representative Michael Curran (D) moved to approve the First Committee Report on Senate Bill 632. Conference committees are formed to reconcile differences in bills between the House and Senate. The committee members, in this case, elected to strip out the original content of Senate Bill 632 and insert the content of four bills that were vetoed by the governor, including Senate Bill 10. Except for some arguing about process, there was no debate on the substance of Senate Bill 632, and the committee report was successfully submitted in the House (Illinois House, November 1, 1989). The next day, the Senate took up the conference committee report on Senate Bill 632 (Illinois Senate, November 2, 1989). Senator Welch explained what happened and summarized the content of the four provisions inserted into Bill 632. Again, except for some arguing over procedure, and a speech calling for protest over the actions of the Speaker for reasons apparently broader than this particular bill, there was no debate about the substance of the bill. In his closing statement, Senator Welch addressed the scholarships issue: Let me just say that the four bills that are included in this particular package are all bills that we supported and passed over to the House. In particular, the bill concerning not giving scholarships to individuals in prisons was the amendatory veto of the governor that we ended up accepting after a great deal of debate. If you want to vote present, and 164 let individuals who are currently in jail continue to receive scholarships at the expense of students who haven’t committed crimes, well, after all our big debate about grand juries and who is tougher on crime, this doesn’t make a heck of a lot of sense to me. You know, go explain that to the student who didn’t get his student scholarship and the fellow in jail did. (Illinois Senate, November 2, 1989, pp. 75-76) The bill narrowly passed the Senate with a vote of 30 ayes and 27 abstentions. The same day, the bill passed unanimously in the House with 114 votes. The governor signed the bill on December 28, 1989 as Public Act 1002, effective July 1, 1990.227 Despite the efforts of a minority of state legislators, incarcerated students in Illinois lost access to the MAP grant and other scholarship programs. It took the Illinois Student Assistance Commission (ISAC) several years of tinkering to get its regulations to where they stand today. In June 1990, ISAC updated its regulations to say that “students enrolled in academic programs while incarcerated are ineligible for ISAC Gift Assistance benefits.”228 In July 1996, ISAC moved that rule to a different section on general eligibility requirements and added a caveat: “students enrolled in academic programs while incarcerated are ineligible for ISAC Gift Assistance benefits, except for Illinois National Guard Grant and Illinois Veteran Grant program recipients” (amendments italicized).229 It is not clear why exceptions were made for these students or how many incarcerated students later were awarded these grants. Finally, in August 1997, ISAC added the language passed in Senate Bill 632 as a caveat to its definition of an eligible institution.230 These two references to incarcerated 227 P.A. 86-1002, 1989 Ill. Laws 6809. 228 14 Ill. Reg. 10538, effective July 1, 1990. 229 “For otherwise eligible educational organizations that provide academic programs for incarcerated students, the term "institution of higher learning" shall specifically exclude academic programs for incarcerat ed students (Section 10 of the Higher Education Student Assistance Act).” See 20 Ill. Reg. 9170, effective July 1, 1996. 230 21 Ill. Reg. 11066, effective July 18, 1997. 165 students – one in the definitions section and the other in the general eligibility section – remain today in ISAC’s regulations. Except for the two noted exceptions, no incarcerated students can receive aid from any of Illinois’ grant programs. Georgia The 1993 Georgia HOPE (Helping Outstanding Pupils Educationally) program is arguably the most famous and most studied state financial aid program (Campbell & Finney, 2005; Chen, 2004; Cornwell & Mustard, 2002; Cornwell, Mustard, & Sridhar, 2006; Dynarksi, 2000; Dynarksi, 2004; Long, 2002; Long, 2004; Resch & Hall, 2002; Sjoquist & Winters, 2013; Zhang, 2011). Biographers of Governor Zell Miller documented how he nearly single-handedly implemented a new lottery and designed the HOPE Scholarship, two of his most important legacies (Eby-Ebersole, 1999; Hyatt, 1997). Lost in these studies and historical accounts is any mention of the fact that incarcerated students lost eligibility for HOPE in 1995. Using these sources, plus interviews with three individuals, I attempted to learn more about how and why incarcerated students lost HOPE and the political contexts of the time. Lottery. The story of the HOPE Scholarship begins with then Lieutenant Governor Zell Miller and his push for a state lottery. The lottery in Georgia was banned by the state constitution, but after neighboring Florida launched its lottery 1988, Miller saw a lottery as an opportunity to fund education programs, one of his passions. He was also dismayed at the millions of dollars that Georgians were spending on Florida lottery tickets (Hyatt, 1997). After deciding to run for governor, Miller needed an issue that made him stand out among his opponents, and he needed a way to fund his education projects. The lottery could do both. Miller’s embrace of the lottery proved to be beneficial; in his own words, “More than any one single thing, it elected me” (Hyatt, 1997, p. 240). Miller was elected Governor in November 166 1990, and upon assuming office in January 1991, he immediately faced a state budget crisis. Though later celebrated for his frugal financial management of the state, at the time, Miller had to put his ideas for new education programs on hold to address the historic state shortfall. Refusing to raise taxes, he cut expenditures by nearly one billion through cutting thousands of state employees and reorganizing state agencies (Eby-Ebersole, 1999). The lottery was a viable revenue stream, but the battle to change the state constitution would be a process. In the 1991 legislative session, Miller narrowly pushed through a resolution in the House to put a constitutional amendment for a lottery on the ballot;231 the Senate easily passed it in April of that year. In the 1992 legislative session, the Georgia Lottery for Education Act was created with all the rules necessary to operate a state lottery, including a stipulation “that net proceeds of lottery games … shall be used to support improvements and enhancements for educational purposes and programs and that such net proceeds shall be used to supplement, not supplant, existing resources for educational purposes and programs,” including a preschool program and “tuition grants, scholarships, or loans to citizens of this state to enable such citizens to attend colleges and universities located within this state.”232 Now, he just had to convince the public to approve it. Many Georgians opposed the lottery on moral grounds, but Miller campaigned across the state on a pro-education message. In the November 1992 election, the lottery ballot initiative narrowly passed with a 52-48 percent result (Hyatt, 1997). Wasting no time, Miller appointed a lottery commission before 1992 ran out, and the first tickets were sold on June 29, 1993 (Eby- Ebersole, 1999). Surpassing national records and all expectations, the Georgia Lottery brought in $1.1 billion in its first year, returning $362 million to the state for education programs (Georgia 231 Ga. Res. No. 50, House Res. No. 7, at 2035 (1991). 232 Georgia Lottery for Education Act, Ga. Code Ann. §20-27-2 (2018). 167 Lottery Commission, 2018). Victorious, Miller would next create his signature education program, the HOPE Scholarship. HOPE Scholarship. With only guesses about how much money the lottery would generate, and with no other existing state scholarship program to use as a model, officials had to work quickly to create the rules for a new HOPE Scholarship program. In the spring and summer of 1993, as the first lottery tickets were being sold, and as Miller was advertising across the state that the first scholarships would be awarded that fall, the Georgia Student Finance Commission and other state officials “self-promulgated” the rules one by one as they learned about issues. Miller – “who thrives on control down to the most minute detail” (Hyatt, 1997, p. 231) – was the chief architect and spoke almost daily with GSFC officials. In fact, one interview participant joked that the GSFC officials did not use the bathroom without permission from Miller, and another noted that not a single sentence changed without Miller’s approval. Early sticking points included how to define a B average when school systems across the state used different grading scales and settling on the original $66,000 income cap to define financial need eligibility. Over the next two years, parents would complain to Miller about being shut out of the HOPE program because of their income. Officials also found that many HOPE recipients lost their scholarship because of their lower-than-expected performance in the first year of college. Because it was funded by the lottery, and because of these unintended complications, “HOPE was not a popular program in Georgia,” according to one official. Facing a Republican critic in the 1994 gubernatorial race who dubbed HOPE the “false hope program,” Miller quickly made changes to the program. The income cap was pushed to $100,000 and then eliminated, and students were given a second chance to earn HOPE back after losing it. It was expanded to cover books and fees and went from a two-year scholarship to a 168 four-year scholarship (Eby-Ebersole, 1999). “So in a matter of I'd say twelve to fifteen months, HOPE went from a program that was very unpopular to a program that you could not take away from the citizens of Georgia,” according to a study participant. In September 1993, the Georgia Student Finance Commission published “Procedural Guidelines for the HOPE Program” containing the first written program rules (GSFC, 1993). There are no rules that indicate incarcerated students, if otherwise eligible, could not get HOPE. In fact, a section of the guidelines explained how incarcerated students could get the HOPE GED Grant, and another section explicitly stated that incarcerated students could get the HOPE Grant, a now defunct HOPE program for students at private institutions. The HOPE Grant’s rules were much simpler than those of the HOPE Scholarship: any full-time student attending a private college – regardless of need or merit – could get $500 for the year (GSFC, 1993). This proved to be a boon to private colleges operating higher education programs within Georgia prisons. According to one official, “we noticed the numbers of the private college students receiving the HOPE Grant was continually increasing tremendously.” After visiting private colleges across the state, commission officials discovered some were offering postsecondary education in prisons. To understand how many incarcerated students were getting the HOPE Grant, the student finance commission and the department of corrections collaborated on a study where they matched the social security numbers of all HOPE Grant recipients against the prison rosters, finding a “fair number” or “a couple thousand” of incarcerated HOPE Grant recipients. This would be a political problem for the governor. The three officials from Georgia who I spoke with did not mince their words about Miller’s criminal justice agenda in the 1990s. The pendulum that swings from a philosophy of rehabilitation to a philosophy of punishment had swung hard towards punishment, as one official 169 analogized. “The vast majority of the citizens wanted to see an eye for an eye; we had a new commissioner of corrections that talked about bringing chain gangs back.”233 Miller postured himself as a “tough-on-crime” governor, and his policies showed it. In 1991, he approved a boot camp program for juvenile offenders that focused on manual labor and tough discipline (Eby- Ebersole, 1999). Ahead of the 1994 election, he passed a sweeping sentencing reform law for certain “serious violent felonies.”234 A person convicted of a first offense was required to serve a 10-year sentence (14 years for murder) (“mandatory minimum”) and serve the entire sentence without the possibility of parole or early release (“truth-in-sentencing”). A second offense resulted in a mandatory life-without-parole sentence (“two strikes and you’re out”). After the more progressive Allan Ault resigned from his post as corrections commissioner, Miller replaced him with Wayne Garner in December 1995, who would earn a reputation of fostering brutality towards Georgia prisoners (Bragg, 1998). Miller also increased penalties for rape, for using minors in drug operations, for driving under the influence of alcohol, and much more (Eby- Ebersole, 1999). Officials knew that dolling out upwards of what could have been $1 million in HOPE Grants to incarcerated students would not reflect the tough-on-crime image that Governor Miller wanted to project ahead of the 1994 election. In October 1994, an official presented the study results to Miller, who reported Miller as saying something like: “I don't want to see it, I've never seen this, but take care of this.” Until after the 1994 election – that Miller won narrowly by just 2.1% – they “kept a lid on” the HOPE Grant issue. What had to be done, though, was clear: the 233 Chain gangs are a form of punishment and control where prisoners are chained together while forced to do physical labor, like building roads or breaking rocks. By mid-century, most chain gang laws were abolished, with Georgia among the last in 1943. In 1995, Alabama became the first state to reinstate prisoner chain gangs, though the practice was abandoned again after just a year. However, the policy created a spark across the country. In 1996, the Georgia House passed a chain gang bill but the Senate rejected it (see Burley, 1997; Glazer, 1996; Peloso, 1997). 234 1994 Ga. Acts No. 1265; Murder, armed robbery, kidnapping, rape, aggravated child molestation, aggravated sodomy, aggravated sexual battery. 170 administration would have to stop administering grants to incarcerated students. Officials later met with all the private college presidents, other private college leaders, and the Senate leadership to explain the issue and the remedy. All understood the political ramifications of this change, and there was no memorable opposition. In an April 3, 1995 guidance letter, the GSFC issued updated rules and regulations for the program. Described as a “regulatory improvement,” all incarcerated students were from then on ineligible for all the HOPE programs (GSFC, 1995). Codification. In the waning years of Governor Miller’s second term, Miller’s allies became concerned about preserving his legacy program – the HOPE Scholarship – for future generations. Because the program existed only in administrative rules, the next governor could easily eliminate it or make changes to it. Miller’s biographer recorded his pre-retirement anxiety: “I doubt anybody will ever do away with the lottery or with HOPE… My concern is that some well-meaning person will come up with another program that sounds good. Then another and another until HOPE and pre-K are diluted or narrowed” (Hyatt, 1997, p. 404). Thus, legislators and other policy staff proposed to put the HOPE program before the legislature to codify it in state statute. But fearing that the political process would bastardize his carefully designed program, Miller strongly opposed the effort. As one legislator remembered Miller saying to him, “You’re going to screw it up!” So, policymakers and Miller struck a deal, or perhaps an ultimatum best describes it: Miller and his staff would write the HOPE bill exactly how they wanted it and the legislature would pass it with not a single change. As both a legislator and a lobbyist recalled, Miller was highly concerned about this process, and both knew that if any changes were made, Miller would veto it. 171 The bill, authored by Representative Mike Polak, delineated definitions, detailed eligibility rules for all the HOPE programs, and delegated authority to the Georgia Student Finance Commission to promulgate additional rules. Included in the bill was a rule preventing incarcerated students from receiving any of the HOPE programs. Though a legislator involved with the bill did not remember this provision, they also were not involved in determining any of the eligibility requirements, since Miller and his staff literally wrote the bill. For several reasons, the bill passed unadulterated. For one, HOPE was highly popular in Georgia, and legislators had little interest in opposing it or Governor Miller. Indeed, the bill would accumulate approximately 120 co-sponsors to get in the “good graces” of the governor. For another, the bill was engrossed in the House upon introduction, a legislative process whereby a two-thirds vote of the House protects the bill from amendments during committee or floor consideration. On February 5, 1998, House Bill 1556 was introduced in the House, the motion to engross was filed, and it was referred to the Committee on University System of Georgia. A day later, it was read a second time. On February 12, it passed from the committee with no changes. With 156 affirmative votes and none in opposition, the HOPE bill passed the House on March 6, 1998. The same day, it was introduced in the Senate and referred to the Committee on Higher Education. On Friday, March 13, the bill passed out of committee unscathed and was read a second time. The following Monday, March 16, the bill passed the Senate with 43 ayes and zero nays. One senator who missed the vote entered a quasi-apologetic statement into the record, perhaps to stay in the good graces of the governor: “On March 16, 1998 due to a momentary distraction I failed to cast a vote on House Bill 1556. Let the record reflect that I strongly support House Bill 1556” (Georgia Senate, 1998, p. 1647). On April 6, 1998, Governor Miller signed his 172 HOPE program into law, thereby preserving his legacy. Now preserved in law was the prohibition on HOPE scholarships for incarcerated students. Even though this statute only applies to HOPE, the Georgia Student Finance Commission over time would promulgate similar bans on aid to incarcerated students in other scholarship programs. Today, no scholarships administered by the GSFC are available to incarcerated students. Indiana The most recent ban on state aid for justice-involved students in the U.S. came in 2011 in the state of Indiana. Where incarcerated students enrolled in college prison programs previously benefitted from the Frank O’Bannon need-based awards, they suddenly lost eligibility after state legislators amended state statutes. This section uncovers the rationale for this policy change through an interview with a state official, news reports, and legislative documents. In 1965, the Indiana legislature created the Indiana Scholarships Act to provide a grant to students under 24 years of age with financial need to attend a public or private college.235 In 1981, the name of the program was changed to Higher Education Awards; at the same time, an original requirement that a student be “a person of good moral character” was struck.236 Then, in 1987, the Indiana legislature amended the Higher Education Award statute to say: The [Commission for Higher Education] may deny assistance under this chapter to a higher education award applicant or recipient who is: (1) convicted of a felony; (2) sentenced to a term of imprisonment for that felony; and (3) confined for that felony at a penal facility” (emphasis added).237 235 Ind. Scholarships Act, P.L. 157-1965. 236 Ind. P.L. 202-1981. 237 Ind. P.L. 235-1987. 173 It is not clear from legislative records why this somewhat nebulous change was added, but some legislators were resistant to it. While the bill passed the Senate 43 to 5, the House passed the bill with a much narrower margin of 56 to 42. In practice, however, the state continued to grant Higher Education Awards to incarcerated students (personal communication, May 30, 2018), raising questions about why the provision was added in the first place. Over two decades later, in 2011, the Indiana legislature removed the imprecision of the “may” from this statute and replaced it with a clear prohibition on granting Higher Education Awards to incarcerated students.238 Understanding why requires an examination of the national and state economic contexts in the preceding years. Great Recession. Economists now refer to the 18 months between December 2007 and June 2009 as the Great Recession, the longest recession in American history since the Great Depression of 1929-1933 (Dworkin, 2010). Because of the sudden collapse of the housing industry, hundreds of banks failed, hundreds of thousands of businesses closed, millions of homes were foreclosed, unemployment rose to 10%, average long-term unemployment reached six months, average household wealth dropped by 20%, and the poverty rate surged to 14.3% (Coghlan, McCorkell, & Hinkley, 2018; Dworkin, 2010). The “revenue-killing recession” also hit state budgets hard (Brown & Ketzenberger, 2010, p. 1; Gordon, 2012). In a recession, when people are unemployed and saving money, states and localities collect less taxes; from 2008 to 2009 nationwide, state taxes dropped by 17% and personal income taxes dropped by 27% (Gordon, 2012). In addition, when people are experiencing financial hardships, they require additional public services, like higher education, Medicaid, and unemployment insurance, causing an increase in state spending (Gordon, 2012). The decrease in 238 Ind. P.L. 229-2011. 174 revenue and increase in welfare spending strain state budgets. To compensate, states must raise revenues (taxes), cut spending, and draw on state reserves, all easier said than done. Indiana faced these same challenges. Though the Great Recession had been officially over for over a year at the end of fiscal year 2010, Indiana was still experiencing the brunt of it. Unemployment was at a staggering 10.2%, growth in Gross Domestic Product was small, tax revenues were at historically low levels, and state spending had been cut by $1.4 billion in fiscal year 2009 and $785 million in 2010 (Brown & Ketzenberger, 2010). Federal stimulus money helped, but with a projected structural budget deficit of over $1.3 billion, more cuts in spending would be needed in the 2011-2013 biennial state budget (Brown & Ketzenberger, 2010). Making cuts. The 2011-2013 budget passed in April 2011 under slightly improved economic conditions, namely because tax collections were increasing (Ketzenberger, 2011). The $1.3 billion deficit projected the year before was looking more like $650 million. Despite an overall 15% reduction in allocations for most executive branch agencies, legislators managed to increase education spending compared to FY 2011, including appropriations increases to the State Student Assistance Commission of Indiana (SSACI) for student financial aid (Horst, 2011). The larger financial picture, however, masks the fact that legislators did, in fact, make cuts to aid spending; by amending eligibility rules in a state statute, they reduced spending on Higher Education Awards for incarcerated students to zero. In the doldrums of the Great Recession, when people are laid off or cannot find work, some pursue higher education for credentials and training, causing a rise in enrollments, especially at community colleges (Barr & Turner, 2013; 2015; National Student Clearinghouse Research Center, 2011). When demand for aid outpaces funding, cuts in award amounts become necessary, and funding must be rationed among different aid programs. In Indiana, funding for 175 the state’s crown jewel merit-aid program, 21st Century Scholars, competed with the need-based program, Frank O’Bannon awards, comprised of a Higher Education Award for students at public colleges and a Freedom of Choice Award for students at private colleges.239 “When funds are not sufficient (as they have not been in recent years) to fully fund both, Indiana has met its firm commitment to 21st Century Scholars first, at the expense of Frank O’Bannon awards, which have been capped at levels below actual four-year tuition and fees” (Johnson & Yanagiura, 2012, p. 26). These “caps” or limits refer to the percentage or dollar amount of a college’s tuition that the award will cover; the lower the cap, the less of a student’s tuition and fees the award can cover. In Indiana, added to the controversy was the fact that the cap – in those days – was different for private and public colleges; need-based awards for students at private colleges were higher because private colleges could charge higher tuition. State officials, in 2009 and 2010, forecasted the looming crisis in state funding for student financial aid. According to one official, it would have taken a “dramatic increase in appropriations” to keep the financial aid programs whole without cutting back the award caps. To understand better where financial aid awards had been going, SSACI and the governor’s office conducted a study of aid expenditures with an eye towards private colleges. What they found surprised them. “A significant amount of funding” was going to incarcerated students, enrolled primarily but not exclusively in private college prison programs. Because the inmates had no income, there were eligible for the highest tier of need-based aid. For some colleges, it was indeed a boon, because “literally fifty percent of their need-based aid revenues were coming from enrolled students that were incarcerated,” according to the state official. 239 Higher Education Awards and Freedom of Choice Awards were administratively combined into what is now called the Frank O’Bannon Grant Program in 2005. 176 In fact, Indiana was educating thousands of inmates each year through on-site instruction by six public and private universities,240 as much as 9% of the whole prison population (Loughlin, 2011). Table 9 displays the number of incarcerated students enrolled in degree programs and the number of those who graduated during three academic years.241 Table 9: Degree Program Enrollments and Completions in Indiana State Prisons Enrolled Completed 2006-07 2,465 2008-09 3,301 944 940 2010-11 2.432 Data not available Funding these students’ college education was expensive. “In years past, the State Student Assistance Commission of Indiana had provided about $12 million to $13 million in financial aid to Indiana state prison inmates” (Loughlin, 2011, para. 7). Another news source estimated the financial aid funding for prisoners to be around $9 million (Clarke, 2012). Either way, cutting aid to incarcerated students meant big savings that could reduce any needed cuts to award caps. The state official summarized the situation this way: “In a time when we are having a hard time funding need-based aid programs for students who played by the rules and done all the right things and we can't fund their programs at the level we need to and we're really cutting those programs by thirty or forty percent, it was hard to justify providing a full ride…to an inmate, someone who had been convicted of a crime.” Satisfied with having found money to save, SSACI and governor’s office officials promoted the idea to the legislature to change the 240 Ball State University, Grace College, Indiana State University, Ivy Tech State College, Oakland City University, and Purdue University. 241 2006-07 and 2008-2009 Statistics gleaned from Internet Archive grabs of the Indiana Department of Corrections website: https://web.archive.org/web/*/http://www.in.gov/idoc/2319.htm 2010-11 statistics from Loughlin, 2011. 177 eligibility rules for incarcerated students as a remedy to the budget crisis. The eligibility rules amendment first appeared in Senate Bill 577 sponsored by Senator Luke Kenley (R). Where SSACI could no longer award grants to incarcerated students, the bill instead gave statutory authority to the Indiana Department of Correction to “provide financial assistance for tuition, books, and supplies” for incarcerated postsecondary students, though the Department of Correction never funedd prison education at near the levels SSACI did. Before Senate Bill 577 passed, however, both provisions were removed and instead added to the biennial budget bill, House Bill 1001.242 The bill amended the state statute on Higher Education Awards, adding: “The commission may not provide assistance under this chapter to a higher education award applicant or recipient who is: (1) convicted of a felony; (2) sentenced to a term of imprisonment for that felony; and (3) confined for that felony at a penal facility (emphasis added).”243 Though the state of Indiana reportedly saved $9 million in 2012-2013, the amendments had swift, catastrophic effects on higher education programs in Indiana prisons (Clarke, 2012). The Department of Correction cancelled contracts with the six college providers and invested instead in vocational and basic education services (Clarke, 2012). “With the loss of SSACI grants, Oakland City [University]244 was forced to discontinue its programs for inmate education and it had to release 150 full- and part-time employees” (Loughlin, 2011, n.p.). Ball State and Indiana State Universities suffered similar losses in teaching staff (Stokes, 2012), and Ball State lost 242 Id. 243 IC 21-12-3-13. 244 In 2012, Oakland City University created a new prison ministries program. Funded by the Department of Corrections, the program offers basic and vocational education services, instead of degree programs as it did prior to the loss of SSACI grants (see https://www.oak.edu/about/prison-ministries). 178 nearly $3 million in revenue from incarcerated students (Associated Press, 2012). To ease the fallout, the Department of Correction allocated $2 million in the 2011-2012 academic year to assist approximately 600 incarcerated students to complete their degrees, a strategy dubbed the “teach out” program (Loughlin, 2011). Today, the Department of Correction offers a variety of literacy, high school equivalency, vocational, self-help, and special topics courses (Indiana Department of Correction, 2018), none of which compare to the full degree programs offered in Indiana prisons prior to 2011. The change in financial aid eligibility rules not only affected the availability of aid to incarcerated students, it fundamentally altered the postsecondary education program offerings in prisons across Indiana. Though Indiana has long since climbed out of the recession, there are no signs that state legislators will reinstate Higher Education Awards to incarcerated students. Oklahoma In 1995, incarcerated students in Oklahoma lost eligibility to receive the state’s main need-based grant, the Oklahoma Tuition Assistance Grant (OTAG).245 Interviews with two legislators involved with the bill and legislative records helped to explain the context for why this might have happened, though a clear answer was not uncovered. On February 6, 1995, Representative Laura Boyd (D) – member of the Appropriations and Budget Committee for Education – introduced House Bill 1075 as a shell bill for higher education. A shell bill is a mostly blank placeholder bill that allows legislators to add substantive provisions to it during the legislative session. This shell bill was needed to increase tuition and fee rates at Oklahoma colleges, universities, and professional schools, which it later did. Thus, the introduced Bill 1075 did not make any mention of eligibility rules for incarcerated students in 245 1995 Okla. Sess. Laws, ch. 247. 179 the OTAG, but that provision was added later during the legislative process. Neither of the bill sponsors who I spoke to could remember exactly how or why the provision on incarcerated students was added, but both offered context and a little speculation as to what might have happened. One legislator – only in their second year of service in the House – described how they would have relied on their “great colleagues” for strategy and direction on the bill: “I had never thought about it from a fiscal or moral standpoint – whether the state should be paying for post- secondary education expenses for inmates – and so I suspect that I was in a position really trying to hear, learn, figure this out.” In the end, this legislator, who described themselves as the most liberal member of the Oklahoma House, speculated that a fiscal rationale was most likely used to justify the ban on OTAG for incarcerated students rather than a moral one. Above all else, it was important to pass the increases in tuition and fees, so the provision on incarcerated students was not going to kill the bill. In retrospect, this legislator lamented the lack of available data that might have changed legislators’ thinking on the issue: “We did not have the sociological data that shows …the number of people who are incarcerated – the vast numbers – that are going to come back to the community; we did not have that. We did not have the sociological data that shows the importance of preparing them for other ways to come back into society.” Another legislator got involved in the bill because they were concerned about the plight of incarcerated people. They wanted to find funding for the institutions in their district, including a prison, a new substance abuse treatment facility, and a college. They wanted to increase funding to incarcerated students, prompting the legislator to wonder if that desire backfired and resulted in the ban on aid to incarcerated students in Bill 1075. Beyond that, this legislator could not remember more about how or why the provision was added. 180 Even without clear memories of the provision itself, both legislators remembered well the climate of criminal justice policymaking in the mid-1990s. Both remarked that Oklahoma had the highest rate of incarceration in the country, which is still true today at 1,079 incarcerated people per 100,000 people compared to the national average of 698 incarcerated per 100,000 (Wagner & Sawyer, 2018). One Democratic legislator said, “Republicans wanted to throw everyone in jail for every little thing but didn’t want to pay for it,” though they also acknowledged that Democrats were “tough on crime,” too. The other – also a Democrat – described it this way: “All during that period time, …each year we would add to penalties. You know, if there is something that was five years, we'd make it eight. Or if there was something that was $250 fine, we’d make it $500. So we were very much into a punitive mindset.” Even without specific memories for why the ban on aid to incarcerated students was adopted, the new provision fit within the pattern of punishing people in prison. One legislator explained: “It was not that big a deal, I don't think. It was not that big a deal, whether it was going to be justified on the basis of fiscal policy, or whether it was going to be justified on existing policy, you know for clarification, or whether it was a part of tough on crime – which we definitely were – it wasn't that big a deal.” Kentucky Two Kentucky policy events in the mid-1990s affected justice-involved students. One law ended all aid for incarcerated students, and the other created a merit-aid program that was off limits to students with felony convictions (including those in prison with felony convictions). Three interviews and legislative records offered some – though incomplete – evidence of how and why these policies were adopted. 181 House Bill 285. In 1996, Kentucky legislators unanimously passed House Bill 285 in both chambers to block all incarcerated students from receiving state financial aid.246 One bill co-sponsor, who self-described as not an “active” co-sponsor, remembered some details about the bill. In the mid-1990s, funding was an issue for education in the state; they were “thin times” in Kentucky, according to the former legislator. The bill sponsor, Representative John Stacey, was described as trying to save as much money as possible for students wanting to attend college. At the time, five financial aid programs were supporting college students in Kentucky, including the Kentucky Tuition Grant Program, the College Access Program, the College Work Study Program, the Kentucky National Guard Tuition Grant Program, and Teacher Scholarships. For the 1996 fiscal year, nearly $31 million was allocated in student aid across these programs (Commonwealth of Kentucky, 1996). Saving money meant identifying where to cut so that more students could benefit from the limited funds. Representative Stacey looked to see where the money was going. Though the co-sponsor could not remember seeing fiscal analyses for how much funding had been going to incarcerated students, somehow, Representative Stacey identified them as “easy targets." Cutting aid to those incarcerated made room for more law- abiding students coming out of high school to get postsecondary education. The unanimous votes indicate that this was not a controversial policy change, for members of either political party. The co-sponsor did not remember opposition to the bill, which is not surprising given the criminal justice policy context of the time. At the time, according the co-sponsor, the attitude toward the prison population was not good. Mandatory sentencing policies were in vogue, and the prison population was rising from the influx in drug offenders. In fact, the Kentucky prison population steadily climbed from just under 3,400 in 1978 to nearly 246 K.R.S. 164.767. 182 13,000 by 1996; it would climb to its peak at nearly 22,500 in 2007 and has hovered around 21,000 ever since (Aiken, 2017), growth that far outpaces the growth in population. Despite the attitudes towards justice-involved people at the time, the co-sponsor reported that the decision to cut aid to incarcerated students was mostly about the money. Kentucky Educational Excellence Scholarship. Two years later, Kentucky policymakers created a new, merit-based scholarship, the Commonwealth Merit Scholarship, which denied aid to any student with a felony conviction. The program was an initiative of the new Democratic Governor Paul E. Patton, who stated in our interview: “I take credit for it.”247 The scholarship, however, is just one aspect of the monumental higher education reforms Patton made during his first term, which would be among his greatest accomplishments as governor. Patton saw higher education as a critical means for economic development in the state. In his own words, “You have to have jobs to keep the educated people in Kentucky. You have to have educated people to be able to create the jobs that you have got to create” (Ellers, 2003, p. 71). Though not an issue raised during his campaign, higher education reform was the focus of his December 12, 1995 inaugural address, much to the surprise of many observers (Ellers, 2003). To higher education leaders, he declared: I challenge you to articulate a new vision, propose a new method, show me a system more devoted to innovation than it is to turf, more concerned about the big picture than it is about its own place in that picture, and I’ll work with you to find the money to do the job. We must have a system of higher education which is more responsive, more efficient and more relevant to today’s realities and tomorrow’s needs. Our people deserve no less, and I will accept no less. (Patton, 1995, n.p.) 247 Governor Patton gave me express permission to identify him this study. 183 After a year of study, Patton found Kentucky’s existing higher education to be inefficient, a poor producer of graduates, and weak on research. He also heard from constituents that Kentucky’s brightest high school graduates were loath to go to college in the state (Ellers, 2003). Patton’s proposals for reform proved highly contentious and unpopular, but in the 1997 legislative session, he prevailed in passing the 94-page Kentucky Postsecondary Education Improvement Act.248 Its most important accomplishments included: separating community and technical colleges from the University of Kentucky and creating a new community/technical college system, situating the Universities of Kentucky and Louisville as major research universities, and creating a clearinghouse of distance/online education programs, later called the Kentucky Virtual Campus (for a complete historical accounting of this law and its passage, see Garn, 2005). Following these successes in higher education reform, Governor Patton pursued his objective of increasing financial aid to higher education students. Though the Kentucky lottery began in 1989, revenues were not dedicated to education, despite popular beliefs. Since the Georgia HOPE program – being lottery-funded – was the “buzz at the time,” Governor Patton proposed to fund state grant programs with lottery proceeds. This proved to be controversial, as both the governor and another state legislator remembered, and the “hang ups” that occurred in passing the bill were related to funding. Previously, revenue from the lottery went into the general fund, but under this proposal, lottery funds would instead be dedicated to new and existing grant programs. This diversion in funds meant a loss of $14 million in 1998-99 and $25 million in 1999-00 from the general fund (Lester, 1998). Despite having a “terrible time” getting the bill through the Senate, according to a bill sponsor, legislators ultimately agreed to phase in 248 1997 Ky. Acts ch. 1. 184 lottery funding for the scholarships; by 2005, nearly 100% of lottery proceeds were allocated to Kentucky’s financial aid programs. The issue of denying the new Commonwealth Merit Scholarship – later rebranded as the Kentucky Educational Excellence Scholarship (KEES) – to students convicted of felonies was apparently much less controversial; both a bill sponsor and the governor had no memory of this provision. In fact, Governor Patton stated: “[I] don’t know how it got in there” and “I would not have supported that,” though he acknowledged he might not have been able to stop it. In other words, even if the governor had been aware of the rule, the bill was in legislators’ hands, despite him being “extremely involved” with the bill. Though important criminal justice reform efforts were implemented during this time, including sentencing first-time, non-violent offenders to alternatives to jail and reforming the juvenile justice system, the state legislators “sure as hell weren’t going to pay for [prisoners’] college,” according to the bill sponsor, who was speaking more generally about the times than reporting memories about the bill. As described above, Kentucky in the late 1990s was still on a path to a burgeoning prison population, and Governor Patton would implement truth-in-sentencing laws requiring people to serve 85% of their sentences, life without parole, a public sex-offender registry website, bootcamps and harsher penalties for juveniles, and other tough-on-crime policies (Ellers, 2003). Though neither governor nor the bill sponsor remembered why convicted felons were excluded from the KEES program, such an action fit with the times. Delaware In 2005, the Delaware SEED program was created with a provision prohibiting students with any felony conviction from receiving the award. An interview with one bill sponsor shed 185 some light on the context of the time, though it remains unclear exactly how and why the felony convictions provision was added. According to one former Delaware legislator, school testing was a big issue in the early 2000s. Schools were funded in part based on test scores, and there was a desire to improve student outcomes. What would later become the Delaware Student Excellence Equals Degree (SEED) program was intended to incentivize better high school student performance. For students earning a 2.5 or higher GPA, SEED covered tuition for two-year degrees at the Delaware Technical and Community College or the University of Delaware. The program proved to be controversial. When first proposed, the program only covered tuition at the community college, but an influential senator disagreed first on the premise that students should go to college for free and second that University of Delaware students were not afforded the same award. A four-year “long drawn out battle” ensued that required Governor Ruth Ann Minner (D) to intervene in the fighting between legislators over the program, according to the legislator. Though the legislator described Georgia HOPE as “big at the time” and noted that someone from Arkansas was pushing legislators to adopt scholarship programs for community college students, the SEED program was not based on these or any other program. And though that legislator claims SEED as among their greatest legislative achievements, it was not exactly what they intended it to be because “politics got in it.” One such political matter may have been the rule that prohibits students with felony convictions from receiving the program. The original House bill introduced on February 4, 2005 by Representative Nancy Wagner (R), and earlier drafts from the previous three years, did not contain restrictions for people with criminal 186 convictions.249 The Senate version introduced in March 24 by Senator Harris McDowell (D), however, included a provision that stated: “The student shall not have been convicted of any crime and no criminal proceedings shall be pending against the student, other than misdemeanor traffic violations, and the student and his or her parent, legal guardian or relative caregiver shall certify such fact.”250 Then on April 28, Sen. McDowell amended the rule to specify just felony convictions were prohibited: “The student shall not have been convicted of any felony and no felony criminal proceedings shall be pending against the student, and the student and his or her parent, legal guardian, or related caregiver shall certify such fact.”251 On May 10, a substitute bill was introduced to replace Senate Bill 30,252 and the same day, the felony convictions rule was amended one final time by Senator Dorinda Connor (R) to remove the statement on pending convictions.253 This final rule remains in statute today. The legislator (a former Representative) could not remember any discussion about the felony convictions rule or when or why it was added, which is understandable considering the legislative record shows it was added in the Senate. The original synopsis of Senate Bill 30 provides a clue as to why the rule may have been added: “The goal of this Act is to encourage Delaware students to stay in school, avoid contact with the criminal justice system, excel academically, and have access to higher education regardless of a family’s financial circumstances.” Perhaps this rule was meant to incentive good behavior with free college as a carrot. Additional research is needed to determine if the SEED program effectively incentivized students to desist from committing crimes. 249 Del. House Bill 53 (2005). 250 Del. Senate Bill 30 (2005). 251 Sen. Amen. No. 3, Sen. Bill 30 (2005). 252 Sen. Sub. 1, Sen. Bill 30 (2005). 253 Sen. Amen. No. 1, Sen. Sub. 1, Sen. Bill 30 (2005). 187 Wyoming Prior to 2006, the state of Wyoming offered little in financial to its residents. Other than federal aid, there was a federally funded, state administered scholarship254 and a one-time, $500 memorial scholarship available to one recipient each year.255 But a surplus in state revenue presented an opportunity for Wyoming to create its first major state scholarship program. An interview with one of the program’s sponsors in the Wyoming legislature provided background on the program’s adoption, though few insights were gained into how and why incarcerated students and students with felony convictions were excluded from eligibility. The early 2000s were strong economic times for Wyoming. In 2003, the state was predicted to have a $1.06 billion surplus for the 2005-06 year, a rarity among states (Associated Press, 2003b). Indeed, “from 2005 to 2007, Wyoming’s economy experienced the fastest growth in more than a decade,” (Wen, 2009, p.1) largely due to growth in existing and new businesses and the mining industry (Liu, 2005; Wen, 2009). Wages were strong, and so was demand for housing; the increase in collected property taxes contributed to the budget surpluses of the time. It was this surplus, according to a state legislator, that many lawmakers had their eyes on. There were plans to invest the money in local municipalities, K-12 and higher education, and reserve funds (Associated Press, 2003a; Neary, 2005). These funds presented a unique opportunity to finally create a state scholarship program in Wyoming. Over a decade after the creation of the Georgia HOPE Scholarship, it was still a popular topic among policymakers. In the mid-2000s, Wyoming legislators heard about the HOPE Scholarship at the education meetings of the National Conference of State Legislators, as well as from a “guy from Georgia 254 Robert C. Byrd Honors Scholarship 255 Douvas Memorial Scholarship 188 [who was] traveling around talking about HOPE.” Georgia HOPE would become only a rough model for what would become Wyoming’s Hathaway Scholarship. Legislators succeeded in the 2005 legislative session to secure some of that surplus money in a higher education endowed account, which would fund scholarships and faculty recruitment and retention at the University of Wyoming and Wyoming’s community colleges.256 It also established a committee for creating the Hathaway merit scholarship program, named for former Governor Stanley Hathaway, who died in October 2005. According to one of the bill sponsors, the development of the scholarship’s eligibility rules was intended to be “apolitical.” They wanted to do what was best for students. Policymakers wanted to create a program that would incentivize high school students to perform better on standardized tests, graduate at higher rates, and be better prepared for college. The program would prove to be popular and uncontroversial, with a unanimous vote in the state house and a near unanimous vote in the senate, save one. 257 The Hathaway Scholarship Program is comprised of four merit scholarships and one need-based scholarship. The merit scholarships are tiered by high school GPA, test scores, and high school curriculum, and any student who receives a merit scholarship can receive an additional scholarship if they have unmet financial met. Students must apply for the scholarship within two years of graduating from high school, so this scholarship does not support returning adult students. Students who are incarcerated and students with any felony conviction are prohibited from getting Hathaway Scholarships under the original state statute. On February 15, 2006, the Joint Education Interim Committee of the Wyoming legislature introduced Senate Bill 85 to establish the rules for the Hathaway Scholarship. In the 256 2005 Sess. Laws of Wyo., ch. 190. 257 2006 Sess. Laws of Wyo., ch. 36. 189 original bill, incarcerated students were excluded from eligibility, but there was no mention of other students with felony convictions. In early March, the House Committee on Education added the provision about students with felony convictions, which received a unanimous vote of nine ayes. A member of that committee remembered the amendment being proposed but did not remember the exact reason for why or if it garnered much debate. It was believed, at the time, that the provision would likely affect a very small number of students – “four or five kids” – though it is not clear if the legislators had data to inform their estimations. Generally, the scholarship program was in part intended to “create good citizens,” so the provisions on incarceration and criminal records seemed to serve as incentives for students to stay crime-free. Today, it is not clear to what extent the Hathaway Scholarship program has succeeded in incentivizing good behavior or how many justice-involved students in Wyoming have been affected. Utah Like Wyoming, Utah offered little financial aid to its residents. Since 1989, students who graduated early from high school received a partial tuition “centennial” scholarship,258 and since 1999, high school students who earned an associate’s degree in high school and who continued their postsecondary education after high school earned the New Century Scholarship,259 both relatively small programs. But in 2008, the leadership of the Utah Senate sought to create a new scholarship to incentivize and reward strong high school student performance. Interviews with a former state legislator and a former state agency official did not uncover exactly why students with criminal records were excluded from eligibility. 258 Ch. 173 § 1, L. Utah 1989. 259 Ch. 273 § 1, L. Utah 1999. 190 Though the performance of Utah students on standardized tests typically ranked near the national average, educational researchers were finding that Utah students were performing last when compared to states with similar demographics (i.e., 8th grade reading, math, science; Utah Foundation, 2007). Given some of Utah’s demographic characteristics (i.e., low poverty, high parental education, low proportion of ethnic minorities), one would expect Utah’s students to perform better than they were (Utah Foundation, 2007). Coming off the good economic times of the previous years, Utah legislators created the Regents’ Scholarship as a potential solution to this performance problem. According to one legislator: We were looking to incentivize students to perform well academically in their high school time period. We were noticing a lack of diligence on the part of some students, and so we wanted to encourage them to not only be diligent but also to attend in-state institutions of higher education. That the Regent’s Scholarship is intended to incentivize strong high school performance is clear in the statutory program description: “to award merit scholarships to students who complete a rigorous core course of study in high school.”260 Students must complete a specified series of courses, earn a 3.0 GPA, and take the ACT (no matter the score) for a one-time Base Award, and students with a 3.5 GPA and a 26 ACT score can earn a renewable Exemplary Academic Achievement Award. Students must enroll in college immediately after high school to receive the scholarship. In addition to these requirements, a Regent’s Scholarship recipient “may not have a criminal record, with the exception of misdemeanor traffic citations.”261 One state legislator could not remember the rationale for why this was included but speculated that very few students 260 Ch. 271 § 3, L. Utah 2008. 261 Id. 191 would be affected. “You know if it had been one of those heavily controversial debates on a particular provision, I would have remembered. I don't remember any controversy on that provision.” The bill easily passed the Senate on March 4, 2008 and the House on March 5, 2008. Two years later, under different economic conditions, the Regents’ Scholarship and the New Century Scholarship were amended with technical changes, one of them being the criminal records rule from the Regents’ Scholarship was added to the New Century Scholarship.262 “The Great Recession was in full swing at the point, and we were looking at ways to make the New Century Scholarship and Regents’ Scholarship survive the onslaught that was occurring to our state budget otherwise.” Legislators added GPA requirements to the New Century Scholarship, among other changes, potentially narrowing the program’s eligibility to save money. The legislator did not remember any discussion of how or why the criminal records rule was added to the New Century Scholarship. The amendments passed both chambers of legislature unanimously and was signed in March 2010. Discussion The state case studies presented above provided the historical accounts of how and why state legislators adopted eligibility rules for justice-involved students in state financial aid grant programs. The following analyses utilized evidence from the case studies to draw new insights about the state policymaking process. Specifically, I next call attention to the factors from the conceptual model that played a clear role in the adoption of state financial aid policies. Policy Adoption Factors The analytical task of this section is two-fold: to identify the factors that led to the adoption of those policies and to evaluate the extent to which those factors are represented in 262 Ch. 270 § 1, L. Utah 2010. 192 Hearn, McLendon, and Linthicum’s (2017) conceptual model of state policy innovation and diffusion in higher education. This analysis resulted from studying each case using the questions from the analytical framework (see Table 8). Table 10 displays a summary of my analysis. Evidence from DE, OK, UT, and WY were insufficient for analysis, so they do not appear in Table 10. Though I make references to them in this discussion section, I do not draw conclusions about how and why legislators in those states created certain policies. Next, I take the four categories of factors from the model in turn to consider if they reflect any of the findings from the case studies. Table 10: Influential Factors on State Policy Adoption from the Cases Factors Strong Evidence Limited Evidence No Evidence Population demographics Education levels Economics State control of institutions Governance arrangements State agency authority Funding for higher education Socioeconomic Context Postsecondary Organizational/ Policy Context Politico-Institutional Context Policy Diffusion Context Political ideology Legislative professionalism Partisanship Electoral conditions Gubernatorial strength and tenure Interest-group climate IN KY GA, IN PA, IL, IN, GA IL, GA, IN KY IL, GA GA GA KY GA IL x x x x x x x x Decision efficiency Competitive advantage Normative pressure Coercive pressure 193 State socioeconomic context. The first category of factors involves the socioeconomic context of a state, which includes factors related to population demographics, education levels and education needs of the state, and economics. Generally, economics proved to be influential in the adoption of several of the state grants in the Phase II sample. For example, during strong economic times, legislators in Wyoming created the Hathaway Scholarship, and during economic downturns, legislators in Indiana, Kentucky, and Utah made cuts to scholarship programs. But this study is an examination of how and why state legislators adopted eligibility rules for justice-involved students, specifically. In one state, Indiana, economics was the primary factor that led to the elimination of Higher Education Awards for incarcerated students. According to the former state official who I interviewed, supplemented with news reports and other sources, the Great Recession hit the Indiana state budget hard, and state scholarship aid was projected to be particularly expensive. Looking for where to cut, state agencies conducted an analysis of state scholarship spending and found that millions in scholarships were going to the thousands of incarcerated students in Indiana prisons, with private colleges receiving a notable share.263 Facing tough economic realities, the former state official who I interviewed described the discovery as “found money,” meaning the funds previously consumed by incarcerated students could easily be diverted to other causes. And they were. The state financial aid agency and the governor’s budget office proposed that incarcerated students no longer get the state’s primary need-based award, and legislators adopted the policy in the 2011 budget bill, saving the state millions. This caused the collapse of most prison higher education programs in the state of Indiana, which were funded primarily by the state scholarships awarded to the incarcerated students. 263 This act of collecting data on the allocation of scholarship funds is also evidence of the factor called “postsecondary policy context” described in the next category. 194 Something similar appears to have happened in Kentucky in 1996 when incarcerated students lost all state aid, but the evidence is less strong. According to one legislator, Kentucky’s economy was weak at the time, and legislators needed to find where to cut scholarship funding. It is not clear how much legislators knew about the amount of funding going to incarcerated students, but their decision to cut that aid was likely made with an economic rationale. These cases confirm the importance of state economic factors in the adoption of state policies, including the adoption of rules that affect justice-involved students. Explicit evidence of the other factors in the socioeconomic context category (e.g., education and demographics) was not found as having direct influence on the adoption of rules affecting justice-involved students. State postsecondary organizational and policy context. The next category includes factors related to higher education governance, the ratio of private versus public institutions, and historical trends in state higher education funding and enrollment. Evidence from the case studies suggest that these factors were somewhat influential in the adoption of policies affecting justice- involved students, but the links are not as explicit. For example, the fact that private institutions in Georgia and Indiana were disproportionately benefitting from state scholarship funding for incarcerated students is important, but that alone did not appear to be the reason that the policies were changed. Other factors were more directly influential, as described elsewhere in this section. The vaguely labeled “policy” factor within this category pertains to patterns and trends in state higher education enrollments and funding, among other things. I also interpret this factor to represent information about where scholarship funds are historically spent (i.e., which institutions), how much (i.e., per institution), and to whom (i.e., types of students). This proved important in three cases, when legislators cut aid to incarcerated students after learning how 195 much scholarship money they were getting. In Illinois, legislators were quite roused with the fact that $500,000 of state scholarship funds were being spent on incarcerated students. For most Illinois legislators at the time, this was too much money not going to law-abiding students. Similarly, Indiana discovered that it would save upwards of $10 million by cutting off incarcerated students from its main need-based scholarship. And in Georgia, though the dollar amount of scholarships spent on incarcerated students was less important than the politics of it, this issue only came to light after an audit of state scholarship spending at private colleges found HOPE Grants being awarded to students in prison. When state legislators obtain detailed data on how scholarship funds are spent, policymakers were prone to take action, including cutting aid to justice-involved students. Another factor within this category involves the authority to create policies. In addition to the laws created by state legislators, “agencies in some states have policymaking authority themselves, independently of other branches of government” (Hearn, McLendon, & Linthicum, 2017, p. 324). This is true within the state higher education policy context, particularly the state financial aid agencies. In Georgia, it was the Georgia Student Finance Commission that issued regulations banning HOPE scholarships to incarcerated students, about four years before state legislators took up the issue. In Indiana, the State Student Assistance Commission of Indiana helped to propose the ban on Higher Education Awards to incarcerated students. In Pennsylvania, it was the Pennsylvania Higher Education Assistance Association that spent decades fighting lawsuits and tinkering with program regulations before banning all grants to incarcerated students. In Illinois, it was the Illinois Student Assistance Commission that had to wrestle with interpreting and enforcing the rules on “good moral character” in its grant programs. In these examples, state financial aid agencies created the policies that affected justice- 196 involved students. This explains how the rules were created, but it remains unclear if the state agencies themselves were the reasons for why the policies were created. In other words, had these states not had agencies with the authority to promulgate eligibility rules for scholarship recipients, would some other policymaking-body have done so instead? Or, in states where agencies have less authority to promulgate these rules, are there fewer instances of agency- created rules for justice-involved students? My research cannot answer these questions, so I cannot conclude if state agencies themselves were causal factors in the adoption of these policies. Nonetheless, state financial aid agencies certainly played a role in creating rules that affected justice-involved students’ ability to get state financial aid. It is not clear how other factors in this category (e.g., enrollment trends, governance structures) have influenced the adoption of rules that affect justice-involve students. Politico-institutional context. The politico-institutional context includes factors related to political ideology, legislative professionalism, partisanship, electoral conditions, gubernatorial strength and tenure, and interest-group climate (Hearn, McLendon, & Linthicum, 2017). Several of these can directly explain how and why policymakers in some states chose to deny aid to justice-involved students. Political ideology. Hearn, McLendon, and Linthicum (2017) defined political ideology “as a coherent and consistent set of orientations or attitudes toward politics, and is usually defined as being situated along a continuum ranging from liberal to conservative” (p. 326). The influence of political attitudes, particularly “tough on crime” attitudes, was described in the literature review of Phase I, and direct evidence of these attitudes were identified in the Phase II case studies. 197 Many of the former policymakers who I spoke to described the political ideology of the 1990s rather consistently; everyone was tough on crime. Participants from Oklahoma, Georgia, and Kentucky described how Democrats and Republicans alike created and adopted state policies that punished criminals. Indeed, the sweeping criminal justice policies of the late 1980s and 1990s created the mass incarceration crisis in America, which is still being rectified as recently as 2018’s First Step Act.264 Evidence from two case studies pointed directly to how political ideology influenced the adoption of rules affecting justice-involved students. The case of Illinois painted the clearest picture. Transcripts of floor debates show that a “tough on crime” political ideology was primarily responsible for stripping students of the ability to get state scholarships. According to the transcripts, the Department of Corrections revealed to legislators that approximately $500,000 of state scholarship funds were going to incarcerated students in Illinois, though it is not known who asked for this information or why. In debating whether to stop such spending, legislators made no mention of tough economic times or the need to save money. Rather, the debate centered around how being tough on crime was incompatible with giving scholarships to students in prison. For example, one legislator said giving aid to incarcerated students “shows a little bit of confusion of our priorities,” and another said doing so was “unfair.” A third described it as an “injustice,” a “particular wrong,” and a “misspending of state funds.” Senator Welch, a bill sponsor, tried to call out the hypocrisy of his colleagues who previously claimed to be tough on crime: If you want to vote present, and let individuals who are currently in jail continue to receive scholarships at the expense of students who haven’t committed crimes, well, after all our big debate about grand juries and who is tougher on crime, this doesn’t make a 264 First Step Act of 2018, Senate Bill 756 (115th Congress). 198 heck of a lot of sense to me. You know, go explain that to the student who didn’t get his student scholarship and the fellow in jail did. (emphasis added, Illinois Senate, November 2, 1989, pp. 75-76) Others made arguments about who deserves to get state scholarships, which directly invokes the theory of social construction of target populations (Schneider, Ingram & deLeon, 2014). Here, Illinois legislators described non-incarcerated college students as “our students,” “kids,” “good,” “abiding by the law,” “more deserving,” and “kids around the outside, who have parents who are lugging their lunch bucket to work every day, trying to send their children to school. Kids that can’t afford it, that have the ability to go.” Senator Welch, exasperated after a late night of confusing debate, made the strongest statement about who is and is not deserving: Well, Ladies and Gentlemen, let me just say that it’s interesting that we’re focusing on the rights of the individual in jail, as opposed to the rights of the individual who hasn’t committed a crime. If you have a poor individual who hasn’t committed a crime, and he applies for a scholarship, he’s thrown in a pool of competition with individuals who have committed crimes. Now, should we say to that individual who has led an upstanding life, hasn’t used drugs, hasn’t stolen a car, hasn’t done any of these other things, … are we saying to them, “Well, we’re not going to take that into consideration. You are now going to compete evenly [with people in jail] to see if you can get this scholarship. Now, you may be competing with some people in jail, but we want to be fair about this, everybody’s on the level playing field.” Well, I don’t think there should be a level playing field. (emphasis added, Illinois Senate, October 18, 1989, pp. 103-104) The language from these policymakers demonstrates the high status of needy college students in the eyes of state legislators. Intentionally, this language juxtaposes them against a less deserving 199 population: people in prison. No legislator said anything derogatory about people in prison, but there is no question about which students most legislators thought were more deserving of scholarships. This evidence lends support to the political ideology factor, generally, and the theory of social constructions of target populations, specifically; politicians negatively constructed incarcerated students and positively constructed non-incarcerated students, resulting in a shift in financial aid benefits away from incarcerated students in order to appease constituents, even though the goal of the program was to support all residents. This debate in Illinois, however, was not one sided. Legislators with an alternative political ideology (and perhaps a different social construction of incarcerated students) showed support for educating prisoners as well as providing them scholarships. Two even tried to appeal to the tough-on-crime sensibilities of their colleagues by painting themselves as equally tough on crime. Before making their counterarguments, one legislator stated, “I don’t think there’s anybody in here [who] is anymore harder on criminals than I am,” and another said, “I do not encourage us to [pander] and soft-peddle the people that have committed crimes.” They and others argued that education was the best intervention available for people involved in the criminal justice system, and since most prisoners eventually get out, it was preferable that they complete college courses while inside. Thus, the scholarship funding was a warranted investment in keeping those individuals from eventually going back to prison. The “tough on crime” camp ultimately won the political battle in Illinois. This case shows how political ideology can directly lead to policy adoption. Elections, a strong governor, and political ideology. In the case of Georgia, a perfect storm of state socio-political factors was at play that caused incarcerated students to lose access 200 to the HOPE programs: tough-on-crime political ideology, electoral conditions, and gubernatorial strength. During the first years of implementation, 1993-1994, the HOPE programs were not politically popular. Perhaps because of how quickly Governor Miller’s administration had to roll out the program, they had to make many rule changes during that time to appease unhappy parents and students. Still, students were taking advantage of the program in large numbers, and officials began to note the unexpected increase in expenditures of the HOPE Grant, a simple $500 grant for anyone enrolled at a private college. After some investigation, program officials learned that many incarcerated students enrolled in private colleges were being awarded these grants. Three factors converged to yield an abrupt policy change. First, gubernatorial strength and tenure have been observed to be influential factors in state policy adoption (Hearn, McLendon, & Linthicum, 2017). Governor Zell Miller was, by all accounts, a strong influence on the HOPE programs. According to his biographers and the three individuals who I interviewed, he was directly involved in every policy decision related to HOPE, and he communicated with program officials daily during those early years of creation and implementation. At that time, the program was not codified in state statutes under the control of the legislature, so Governor Miller had complete control over the program as administered by his administration. Miller, in fact, directed his staff to change the rules so that incarcerated students could not receive HOPE scholarships. This factor explains how incarcerated students lost aid, but two other influential factors explain why. A defining political ideology – the second factor – of Governor Miller was his tough on crime stance. He campaigned on it, and his voters expected him to follow through. His response 201 to learning that many people in prison were being awarded scholarships from his prized HOPE program was predictable. And the timing of this issue could not have been worse. The third factor at play was the influence of elections on state policy adoption, including election competitiveness and the timing of elections (Hearn, McLendon, & Linthicum, 2017). When Miller’s staff learned about what was happening with private colleges in Georgia prisons, it was just weeks before the gubernatorial election of November 1994. News that incarcerated students were receiving thousands of dollars in HOPE Grants could have been a political bombshell for Miller’s campaign. Miller’s opponent was already a loud critic of the HOPE program, and worse, giving aid to students in prison would have appeared “soft” on crime. Miller directed his staff to quietly take care of the problem. They suppressed the data about HOPE expenditures to students in prison, and after Miller narrowly won re-election, they changed the program rules to explicitly deny aid to incarcerated students in the Spring of 1995. Miller averted an elections disaster and maintained his promise to be tough on crime. The influence of the governor, his strong political ideology, and the timing of the election were the direct causes of the 1995 ban on aid to incarcerated students in Georgia. Other factors. The other factors within the politico-institutional context not yet mentioned include legislative professionalism, partisanship, and interest-group climate. Legislative professionalism was not explored, but additional research might find that it is influential. Regarding partisanship, evidence points to this factor not being influential. In most of the cases above, Democrats and Republicans alike agreed to strip financial aid from justice- involved students, often unanimously. Rather than party, the political ideology of legislators, specifically the extent of their “toughness” on crime, seems to be a better explanation of how these policies came to be adopted. Finally, only the case of Georgia HOPE included any mention 202 of interest groups. Lobbyists for private colleges were involved in creating and codifying the HOPE Grant program, but even the lobbyists, it seems, understood the political need to eliminate aid to incarcerated students. Additional research is needed to understand if lobbyists for private colleges were involved in Indiana, where private colleges were particularly affected by the ban on aid to incarcerated students in 2011. State Policy Diffusion Context. The final category of factors comprises the four forces of horizontal policy diffusion: decision efficiency (learning), competitive advantage, normative pressure, and coercive pressure (Hearn, McLendon, & Linthicum, 2017). These forces explain how and why states adopt policies that are modeled off other states’ policies, oftentimes from neighboring states. Among the policy cases, there was not strong evidence of the state-to-state spread of policies that deny financial aid eligibility to justice-involved students. Interview participants in Utah, Wyoming, Kentucky, and Delaware reported being aware of other state grant programs, especially Georgia HOPE, while they were developing new programs in their states, but all indicated they created their own programs from scratch. Their programs, then, were not direct products from policy diffusion. Further, none indicted that their program’s provisions that affect justice-involved students came directly from another state’s program, though a few of the legislators with fuzzy memories thought this was a possibility. Evidence from only one state points to possible diffusion effects on the rules related to justice-involved students. According to Eber (1982), Illinois’ 1957 State Scholarship Program was modelled off California’s 1955 Competitive Scholarship Program. But even if this were true, there is not enough evidence to confirm whether Illinois’ original prohibition on aid to students who did not have “good moral character” came directly from California’s program. More research is needed to confirm where this provision came from. 203 Two New Factors Across the state policy cases, Hearn, McLendon, and Linthicum’s (2017) conceptual model of state policy innovation and diffusion in higher education accurately predicted most of the factors that influenced the adoption of rules that affected justice-involved students. The factors with the strongest evidence of support included economics, state funding, political ideology, gubernatorial strength and tenure, and electoral conditions (see Table 9). But there are a few cases where the available evidence suggests that two other factors influenced state policy adoption that are not represented in the conceptual model. This presents an opportunity to explore these factors and add them to model. Vertical diffusion. The conceptual model’s category devoted to policy diffusion is limited to horizontal diffusion, or the transfer of policies from state to state. But it is also possible for policies to diffuse vertically, from a higher authority down to a lower level of government, or vice versa. Examples of top-down vertical diffusion might include a new federal law or court decision that mandates the adoption of a new state policy, or a federal grant program that incentivizes states to adopt a new policy with fiscal rewards or punishments. A classic example of the latter is the National Minimum Drinking Age Act of 1984, a federal law that penalized states with reduced highway appropriations that did not raise the minimum drinking age to 21 (Liebschutz, 1985). Because of the power hierarchy, vertical diffusion is often described as a coercive diffusion mechanism, where states are forced to adopt certain policies. Across the literature, studying horizontal diffusion among American states has been the historical tradition (Graham, Shipan, & Volden, 2013). As such, vertical diffusion tends to get only brief mentions, though a few studies have explored it in some depth (Allen, Pettus, & 204 Haider-Markel, 2004; Karch, 2006; Shipan & Volden, 2008; Karch, 2012). From this study, the case of Pennsylvania provides a unique opportunity to examine vertical policy diffusion. The case description of Pennsylvania above traced the twenty-year process to find a legally defensible policy on providing state grants to justice-involved students. In the end, state agency officials adopted a new policy to ban only incarcerated students from getting aid, and applicants would otherwise not be asked about their criminal histories. This policy resulted from coercive, vertical diffusion forces. In 1969, both the Pennsylvania legislature and the Pennsylvania Higher Education Assistance Authority (PHEAA) adopted policies that banned people convicted of felonies and of “misdemeanors involving moral turpitude” from receiving state financial aid. It is not clear why this was done, but the policies PHEAA adopted later resulted from mandates from a higher legal authority. In a 1971 lawsuit against PHEAA, a federal court found that denying aid to students convicted of misdemeanors of “moral turpitude” was unconstitutional (Cooperation of Haverford College v. Reeher, 1971). Leaving PHEAA no other choice, it struck this provision from its regulations, instead adopting a new regulation with more specific prohibitions related to criminal convictions. Had it not been for this lawsuit, it is impossible to know if PHEAA would have adopted new policies; thus, the coercive, vertical diffusion influence is clear. Two additional court cases also had vertical diffusion effects on PHEAA’s policies. In fact, PHEAA won subsequent challenges to its new regulations on banning students convicted of felonies (Corporation of Haverford College v. Reeher, 1972; Carbonaro v. Reeher, 1975), but it responded by adopting new policies, nonetheless. PHEAA sensed that it narrowly won these lawsuits. In Smith’s (1990) summary report, a comment highlights this uneasiness: “Clearly the court was focusing on the Agency’s right to deny only felons; had Kelly’s crime been other than 205 a felony, the court would likely have ruled against PHEAA” (p. 1). Feeling pressure to avoid additional litigation, PHEAA made a series of progressive policy changes over the next twenty years. First, PHEAA stopped denying aid to students with pending charges, and second, PHEAA stopped denying aid to students with misdemeanor convictions. Then PHEAA began approving students with certain felony convictions for aid eligibility and made loan programs available to everyone, regardless of criminal history. By 1988, PHEAA had grown so lenient with its decision-making that nearly every student with criminal convictions who was not currently in prison was being approved for grant aid on appeal (Smith, 1990). So, PHEAA’s final policy on the subject was to deny grant aid to incarcerated students, only. The lawsuits of the early 1970s clearly had a profound coercive effect on PHEAA, even though only the first lawsuit legally mandated the adoption of new policies. In addition to the clear case of Pennsylvania, I expected to find vertical diffusion effects from the 1994 ban on federal aid to incarcerated students. A reasonable hypothesis is that after federal legislators banned aid to incarcerated students, state legislators followed suit. At first glance, this would appear to be what happened in New York, New Jersey, Kentucky, California, and others, where incarcerated students lost aid eligibility in the years after the 1994 federal aid ban (see Table 4). However, I was unable to collect enough data to determine how and why the policies in those states were changed. Additional research is needed, but vertical diffusion is a probable explanation. Given the Pennsylvania case, especially, I propose expanding the state policy diffusion context of Hearn, McLendon, and Linthicum’s (2017) conceptual model. Rather than limiting the conceptualization of state policy diffusion to horizontal influences, adding vertical diffusion expands the possibilities for how higher education policies can be diffused. With this new view 206 in mind, a review of the literature will likely find many other examples of federal policies that have shaped state policies through any of the four previously identified forces. For example, in another historical case involving financial aid, a federal court declared unconstitutional the 1971 Tennessee Tuition Grant Program in its entirety.265 The Tennessee legislature repealed it and created a new financial aid program that avoided the unconstitutional provisions.266 This represents another case of coercive, vertical diffusion. In a more recent example, at least four states adopted new laws in response to the U.S. Department of Education’s 2011 Dear Colleague Letter, 267 which recommended new procedures for addressing campus sexual violence and sexual misconduct (Bauer-Wolf, 2017). Though this Obama Administration guidance was not a legal mandate, many college leaders and policymakers felt compelled to create policies that reflected the new standards, perhaps suggesting a mix of coercive and normative influences were at play. Complicating matters, the Trump Administration rescinded the 2011 guidance letter and replaced it with its own regulations, which both carry more legal weight and may now conflict with state laws that were created to match the Obama guidance. A case study of vertical policy diffusion in higher education is developing in plain sight, and policy theorists could learn much from seeing how state policymakers will react to the changing federal guidelines on campus sexual misconduct. Another opportunity to observe vertical diffusion in higher education is through so-called “bottom-up federalism,” the process of policy diffusion from a lower-level of government to a 265 In 1974, the United States District Court for the Middle District of Tennessee struck down the Tennessee Tuition Grant Program because student awards were administered directly from the state to colleges, rather than to students. Because much of this funding went to private, religious institutions, the funding was determined to be institutional aid that supported religious activities, in violation of the Establishment Clause of the First Amendment relating to religious freedom. See Americans United for Sep. of Church & State v. Dunn, 384 F. Supp. 714 (M.D. Tenn. 1974). 266 Ten. Acts 1976, ch. 415. 267 Cal. Edu. Code § 67386; Conn. PA 14-11; Illinois Preventing Sexual Violence in Higher Education Act, PA 099- 0426, 2015 Ill. Laws; Enough is Enough Act, NY Educ. Law § 129-A, 129-B (2018); 207 higher one (Shipan & Volden, 2006). Still vertical in nature, it is the opposite of the top-down diffusion previously discussed. For example, Shipan and Volden (2006) studied how anti- smoking policies adopted by local governments influenced the adoption of state-level anti- smoking laws. Even this anti-smoking policy example presents opportunities for additional study in higher education. Just five states currently have laws requiring their public institutions to be 100% smoke-free, but at the local level, over 2,000 higher education institutions from across the country have adopted 100% smoke-free campus policies (American Nonsmokers' Rights Foundation, 2018). Could university-level policy decisions on smoking diffuse up to the state level? This and other vertical diffusion policy examples deserve to be explored through scholarship on state higher education policy adoption. Policy patching. The factors from the conceptual framework do not offer a compelling explanation for what happened in Michigan in 1979-1981. I propose here that “policy patching” is a better conceptual match; first, I explain the terms. Policy patching is a relatively new term from the literature on “policy mixes” (Howlett & Rayner, 2013), which is a policy design concept used most commonly by researchers outside of the US in the fields of sustainability, energy, and climate science (Rogge & Reichardt, 2016). A policy mix occurs when governments employ a complex mixture of policies, programs, and processes (“policy instruments”) to address a certain policy goal, like improving energy efficiency in a country (Rogge & Reichardt, 2016). For the policy mix to achieve the intended outcomes, a degree of consistency, coherence, and congruence is needed among the various policies (Howlett & Rayner, 2013).268 As policies are added to the mix – called layering – it can lead to unintended consequences, like inconsistencies 268 Consistency: “the ability of multiple policy tools to reinforce rather than undermine each other in the pursuit of individual policy goals.” Coherence: “the ability of multiple policy goals to co-exist with each other in a logical fashion.” Congruence: “the ability of multiple goals and instruments to work together in a uni-directional or mutually supportive fashion.” See Howlett & Rayner, 2013, p. 170. 208 among the policies (Howlett & Rayner, 2013). Thus, policy tools are needed to make corrections, including “policy packaging” and “policy patching,” terms coined by Howlett and Rayner (2013). “Policy packaging refers to a policy design process in which previous policies are discarded and a new policy package is introduced,” (Kern, Kivimaa, & Martiskainen, 2017, p. 13). Alternatively, policymakers engage in policy patching when they make changes to existing policies in order to improve their consistency, coherence, or congruence with the other policies in the policy mix, “much in the same way as software designers issue 'patches' for their operating systems and programmes in order to correct flaws or allow them to adapt to changing circumstances” (Howlett & Rayner, 2013, p. 177). Evaluation literature shows that policy patching can be an effective means for improving policies and may be more feasible than packaging (see Kern, Kivimaa, & Martiskainen, 2017; Wellstead, Rayner, & Howlett, 2016; Wellstead & Howlett, 2017). Policy patching can be described as a form of policy “tinkering” (Wellstead & Howlett, 2017), which is a term more commonly used in US-based policy studies. Policy tinkering is ill- defined across the literature, but generally, it occurs when policymakers make changes to existing policies. Said differently, tinkering is the “manipulation of the dimensions of a policy alternative to generate other alternatives” (Weimer, 1993, p. 111), or “changing requirements of the [policy] instruments (their ‘settings’) and the way that the instruments are implemented (their ‘calibrations’) (Wellstead, Rayner, & Howlett, 2016, p. 1876). However, there is disagreement among scholars about how consequential these changes are. From a study of state higher education policies, tinkering is defined as “slight alterations in policy content or scope” (McLendon, Heller, & Young, 2005, p. 389). From a study of state criminal justice policies, tinkering is defined as “revisions [that change] words without actually changing the meaning or 209 scope of the content in the law” (emphasis added, Lytle, 2015, p. 224). Because of the discrepancies in definitions of tinkering and the lack of conceptualization of tinkering, I prefer the concept of policy patching to describe what happened in the case of Michigan. Next, I trace the events of the Michigan case study using the language of policy mixes to illustrate how and why legislators in Michigan adopted its ban on financial aid to incarcerated students. By the mid-1950s, several states had established a new policy goal that then spread across the country throughout the 1960s and 1970s; they wanted more college graduates for their growing state economies. Policymakers created state-funded scholarship programs as the policy instrument to address the goal. Michigan joined this pursuit in 1964 when it created the Competitive Scholarship program for students who earned strong scores on a state-administered exam. Over time, states began to layer policies by adopting multiple scholarship programs, each unique but with the same overall goal to get more residents into college. Michigan layered on the Tuition Grant program in 1966 for students at private colleges (the amount determined by family income), the Legislative Merit Award program in 1976 for students who earned strong scores on a national exam, and the Differential Grant program in 1978 for students a private college (the amount not determined by family income). Inevitably, the incremental layering of scholarship programs resulted in inconsistencies among the programs, this according to a 1979 audit of the four programs by the Auditor General: “While there is some agreement among the acts in terms of the eligibility requirements and funding provisions they contain, it has been argued that certain discrepancies among the acts confuse or obscure the general policy of the higher education assistance program” (Michigan Senate Analysis Section, 1980a, p. 1). This is exactly the problem that scholars of policy mixes 210 predict in cases of layering. The Auditor General identified the inconsistent provisions, including: 1. Two of the programs prohibited awarding scholarships to students enrolled in theology or divinity programs and two did not. 2. Two of the programs required applicants to be persons of “good moral character” and two did not. 3. There were three different definitions of what constituted required state residency (i.e., living in the state for 3 years, for 12 months, or at the time of high school graduation). 4. The programs specified different dollar amounts for the awards. Rather than repeal these four programs and institute a new package of programs (policy packaging), the Auditor General proposed that legislators correct the inconsistencies in the existing programs (policy patching) (Lee, 1979). In a series of four bills, Michigan legislators patched the policies by amending the program rules (Senate Bills 1275-1278). Once the bills were signed in early 1981, the four programs were more consistent on the initial eligibility requirements. Before that happened, however, lawmakers tinkered further with the rules pertaining to justice-involved students. Initially, legislators solved the problem of the “good moral character” rule by deleting it from the Competitive Scholarship and the Tuition Grant programs, thereby making all four programs consistent on this point. Then, lawmakers thought to replace the good moral character rule in the Competitive Scholarship with a new rule stating that no one incarcerated could receive the award. The Senate Analysis stated that this provision would “sharpen the original intent” of the original character requirement (Michigan Senate Analysis Section, 1980a, p.3), though it is possible that other factors influenced this decision. Nonetheless, it was recognized 211 that adding this provision would again cause inconsistencies with the other three programs. Curiously, lawmakers chose to add the incarceration rule to the Tuition Grant program but did not add it to the Legislative Merit Award or the Differential Grant programs. The policy patching was not as efficient as it could have been, though only a few years later, both the Legislative Merit Award and Differential Grant programs were defunded. Today, the Competitive Scholarship and the Tuition Grant program are still available to students but are not available to incarcerated students. As described in Phase I, Michigan continued to layer on scholarship programs. After 1981, Michigan legislators would codify at least seven new scholarship programs, each with their own governing statutes and eligibility requirements. The inconsistencies in rules for justice- involved students became worse over time. Table 11 displays the layers of scholarship programs and the patches related to justice-involved students in Michigan since 1964. Though five of the programs are now defunct, the remaining six programs have three different iterations of rules for justice-involved students, including: no rules, not be incarcerated, and not be convicted of a felony involving an assault, physical injury, or death. It may be time again for Michigan legislators to study the consistency of their financial aid programs and to patch them accordingly. 212 Table 11: Layering and Patching of Michigan Financial Aid Policies Program Year Layering Patching Defunct Competitive Scholarship 1964 “good moral character” Tuition Grant 1966 “good moral character” Legislative Merit Award 1976 No rules for JI students Differential Grant 1978 No rules for JI students Competitive Scholarship, Tuition Grant 1981 Deleted character rule, replaced with “is not incarcerated” Part-time Independent Students Grant 1982 “is not incarcerated” Tuition Incentive Program 1987 No rules for JI students Police Officer’s and Fire Fighter’s Survivor Tuition Grant 1996 No rules for JI students Michigan Merit Award 1999 Children of Veterans Tuition Grant 2005 “has not been convicted of a felony involving an assault, physical injury, or death” “has not been convicted of a felony involving an assault, physical injury, or death” Michigan Promise Grant 2006 No rules for JI students Tuition Incentive Program 2011 Added “must not be incarcerated” Fostering Futures Scholarship 2014 No rules for JI students x x x x x In the cases of Michigan’s Competitive Scholarship and Tuition Grant, the results of policy patching should not be understood to be inconsequential tinkering. Though the intent of the corrections was to achieve consistency across programs, the effects were quite tangible. It is not clear if incarcerated students were able to receive the Competitive Scholarship or the Tuition Grant prior to 1981, but the policy patching process made it certain that they could not in the future. In effect, Michigan legislators adopted a new policy when they changed the rules for 213 incarcerated students during the policy patching process. Therefore, I propose adding policy patching as a factor to Hearn, McLendon, and Linthicum’s (2017) conceptual model. Because policy patching is inherently a political process, I add it to the politico-institutional context category of the model. Along with other factors related to the legislature and characteristics of the legislators, policy patching is not merely a process, it is a causal factor in the adoption of new policies. It is probable that other cases of policy patching can be identified within state higher education policies. Policymakers regularly “tinker” with program rules, especially at the state- agency level. Changing a word or adding a comma here or there will at times have no substantive effect on the policy, but at other times, such seemingly small changes can have major implications, like when changing a “may” into a “may not,” which happened in the case of Indiana. When policymakers make these changes to bring a policy into alignment with the goals of other similar policies, and when the change represents a substantial change to the policy, policy patching can be understood as a factor that directly influences the adoption of new state policies. Additional research is needed to uncover other examples that support this proposition. Other Findings Somewhat unrelated to the specific research questions, I observed a theme through several of the interviews that deserves discussion. Of the 13 individuals interviewed, 9 remembered little to nothing about how or why justice-involved students were written out of state financial aid grant programs. Perhaps I did not identify the most knowledgeable individuals to interview, although most were bill sponsors or were otherwise somehow directly involved with the policy. Or, perhaps, their inability to remember was a function of memory. Afterall, I asked participants to remember details about events from 10 to 25 years ago. Still, how is it 214 possible that a professional lawmaker could not remember something about a bill with his or her name on it, especially something so targeted as eliminating aid to all incarcerated students, for example? In fact, two interviewees reacted to my questions about their bills with shock and surprise, as if it were the first time they were hearing about the rules related to justice-involved students. It seems there is something else that might explain their inability to remember the policy debates around justice-involved students. It appears that justice-involved students and their plight were, at the time, unremarkable and therefore unmemorable years later. If few policymakers or others were concerned about this population, then the explicit denial of their ability to get financial aid for college would have sparked little emotional reaction, debate, or memorable moments. A few participants’ comments point to this possibility. A former policymaker from Utah said it most directly: “You know if it had been one of those heavily controversial debates on a particular provision, I would have remembered. I don't remember any controversy on that provision.” A couple others also described it as “not a big deal” or “not controversial” to deny justice-involved students, even though they could not remember any details. This lack of controversy again points to Schneider, Ingram, and deLeon’s (2014) theory of social construction of target populations. With no advocates to speak on behalf of justice-involved students, and with no one to ask lawmakers to stop and think about the effects of their decisions for those students, policymakers adopted laws that would affect a powerless population with little to no thought about it. As elaborated in the Phase I study, this inability to see justice-involved students, to understand their needs, and to care about them is an important finding of this study. Should state legislators in the future propose such a policy, advocates should know that a thorough debate of the issue would not be the norm. Rather, these prohibitions on aid are seemingly slipped into 215 bills with little fanfare, explanation, or debate and are likely to pass without resistance. If advocates of justice-involved students want to see their aid remain intact, they will have to push legislators to debate the issue by asking them to formulate and defend a position on it. If nothing else, perhaps a rigorous debate would help legislators to remember the issue, should they be asked about it 15 years later. This incidental finding may also have implications for policy adoption theory and how policy scholars study policymaking. If policymakers cannot be assumed to be rational, as was assumed in this study, understanding how policies are created becomes so much more complex. Rather, as many political scientists have noted, the policymaking process is quite political (see Gordon, Lewis, & Young, 1993), which allows for any number of influences to affect the creation of laws. The policymakers themselves may be entirely unaware of how certain provisions are written into their own bills as the pieces of legislation move through the process. After all, each bill undergoes a long process of negotiation among many people before it becomes a law, presenting many opportunities for seemingly trivial provisions to be slipped in, like eligibility rules for justice-involved students. In future research on state policy adoption, researchers may need to consider in more depth the inherent political and irrational nature of the policymaking process to explain circumstances when even the lawmakers themselves cannot explain or remember important details about a policy. Implications The purpose of the Phase II study was to understand how and why state legislators adopted rules that disenfranchise justice-involved students from receiving state financial aid grants. The intended implications were to advance scholarship on state policy adoption theory through qualitative case study methodology. The most important implications of this study 216 include confirming the strong applicability of Hearn, McLendon, and Linthicum’s (2017) conceptual model and expanding it by two factors. There are also other implications for future policy research, for policymakers, and for advocates of justice-involved students. An Updated Conceptual Model In one of its first empirical tests since publication in 2017, Hearn, McLendon, and Linthicum’s (2017) conceptual model performed well in accurately predicting several of the factors that were strongly influential in the adoption of state higher education policies in several of the cases from this study (see Table 10). This is a testament to the model’s foundation in past research and to its applicability in future research. However, vertical diffusion and policy patching were identified as causal factors that do not appear in the model, though they are discussed elsewhere in the policy design literature. Thus, based on evidence from this study, these two factors should be added to the model, at least tentatively. Additional research is needed to uncover other cases of such factors that can confirm the extent to which the factors belong in the model. Based on case study evidence from this study, I add the two new factors to the model. Policy patching is added to the Politico-Institutional Context along with other factors that involve the politicians and officials who create policy. Vertical diffusion is added to the Policy Diffusion Context. Specifically, it is clarified within this category that both horizontal and vertical diffusion influences are possible and that any of the four existing forces within the category (i.e., decision efficiency, competitive advantage, normative pressure, and coercive pressure) can result in vertical or horizontal policy spread. The definitions of the forces are amended so that they are not limited to describing state-to-state influences. No other changes are 217 made to the model, either removing factors, adding categories, or otherwise rearranging the model. The new factors are underlined in the updated representation of the model (see Figure 7). 218 Adoption of State Postsecondary Policies Socioeconomic Context Demographic: population size, age-group distribution, growth rate, density, % of color, # SMSA’s over one million, etc. Educational: educational attainment levels, postsecondary enrollment rate, etc. Economic: gross state product [GSP], GSP per capita, GSP per-capita growth, median per- capita income, unemployment rate, etc. Organizational and Policy Context Organizational Ecology: number and the proportional distribution of institutions by control and level, the number and proportional distribution of students by institutional sector, student migration patterns, number and strength of research institutions in the state, agency research capacity, etc. Policy: postsecondary funding, tuition, and enrollment patterns. Politico-Institutional Context Political Ideology: state ideology indicators Legislative Professionalism: Squire Index Partisanship: legislative/gubernatorial party affiliations and strength, etc. Electoral Conditions: degree of parties’ electoral competitiveness, election timing, etc. Gubernatorial Strength and Tenure: constitutional authority, time in office, etc. Interest-group Climate: interest-group characteristics (density, strengths, concentration) and governance arrangements (consolidated governing board across institutional sectors, strong coordinating board, weak coordinating board, planning agency) Policy patching: tinkering with an existing policy in order to improve consistency across similar policies that results in a substantial change to the policy, thereby constituting a new policy adoption event Policy Diffusion Context Horizontal and Vertical Diffusion: the spread of policies between states and from higher and lower authorities to states. The following diffusion forces can influence horizontal and vertical policy diffusion. Decision Efficiency: states may learn from other governments’ actual experiences with policy choices, thus informing and ideally improving their policy choices Competitive Advantage: states may adopt policies to position them to better compete with other states or national or local governments Normative Pressure: states may adopt policies favored by regional or national- level peers, intermediary organizations and policy entrepreneurs. Coercive Pressure: states may be forced to adopt policies imposed by critical outside resource providers Figure 7: An updated conceptual model of state policy innovation and diffusion in higher education. 219 From “Conceptualizing State Policy Adoption and Diffusion,” by J.C. Hearn, M. K. McLendon and K. C. Linthicum, 2017, in Higher Education: Handbook of Theory and Research, 32, p. 320. Copyright 2017 by Springer International Publishing. Adapted with permission. For Policy Researchers Qualitative case study methodology proved to be useful for unearthing new factors that influence state policy adoption. Even though some combination of the many policy adoption theories might identify the factors involved in any given case, the reality is that each policy has a unique story of adoption. Only in-depth case study analysis can capture the whole story to identify the historical context and factors that influenced adoption. Specifically, interviewing policymakers proved to be the most direct way to gain insights on the process. Quantitative analyses simply cannot achieve these results. This study identified two new factors that are described elsewhere in the policy adoption literature but were not identified in the model of state policy adoption. Thus, the more higher education policy adoption events that are studied as cases, the more potential there is for expanding existing understandings of the adoption process. For state financial aid policies that affect justice-involved students, there are at least 16 other states identified in Phase I that I was not able to investigate as a case study in Phase II. Perhaps other researchers can pursue this line of research to learn how and why state legislators adopted policies that disenfranchise justice- involved students. The methodology and research questions of Phase I yielded the sample for Phase II. Only currently-funded state grant programs that contained rules affecting justice-involved students were identified. As such, there may be another group of cases that are worth studying. Non- adopter states are states that considered adopting a certain policy but did not. Understanding why states do not adopt a certain policy provides useful insights into the policy adoption process 220 (Ingle, Cohen‐Vogel, & Hughes, 2007). It is possible that some states considered adding rules to state grant programs that would have disenfranchised justice-involved students but did not. Examining these cases would provide an alternative narrative to the cases I described in Phase II and might offer new insights into how legislators view justice-involved college students. Research on policy mixes – and its earlier derivative “integrated policy strategies” – has existed since at least the 1990s (see Rayner & Howlett, 2009; Rogge & Reichardt, 2016), and scholars in the past two decades have advanced the theory considerably (see Howlett & Rayner, 2007). The application of policy mixes theory appears limited to research on energy and climate policy mostly outside the U.S., as in the study of oil sands reclamation policy in Canada (Wellstead, Rayner, & Howlett, 2016), off-shore wind policy in Germany (Reichardt, Negro, Rogge, & Hekkert, 2016), forest policy in Canada and the U.S. (Wellstead & Howlett, 2017), and energy efficiency policy in Finland and the United Kingdom (Kern, Kivimaa, & Martiskainen, 2017), to name only a few. This dissertation may be the first to apply tenets of policy mixes to a solely U.S. context and to education policy. The concepts hold great promise for application to the study of U.S. higher education policy, as illustrated by the case of Michigan’s multi-layered and patched financial aid policies. Though policy scholars in the U.S. already have volumes of policy design theories at their fingertips, policy mixes deserve to be explored in additional U.S. policy contexts. For Policymakers and Advocates The analytical framework (Table 8) points out that information and data are key factors in the policy adoption process, and there are several ways that data can help or hinder the plight of justice-involved students. The cases of Illinois, Georgia, and Indiana showed that when policymakers obtained detailed data on how much scholarship funding was spent on incarcerated 221 students, they reacted by taking away that aid. This helps to explain why some state financial aid agency officials are hesitant to talk about their programs that support justice-involved students. They fear that state legislators will catch wind of the programs and de-fund them, especially depending on the political make-up of a legislature at any given moment (this according to a conversation I had with one official who shared these concerns). Similarly, it is conventional wisdom among prison higher education practitioners that keeping a low profile is the best way to maintain current funding levels. Raising too much attention to certain programs or asking for more funding poses a risk of backfire by having legislatures instead slash funding. Even public discussions of the successful Second Chance Pell experimental program make advocates nervous that the Trump Education Department will eliminate it (these views according to discussions on listservs and at conferences). In the few states that provide financial aid to justice-involved students, policy advocates must walk on eggshells to maintain the status quo, making program expansion difficult. Perhaps the remedy is to fight fire with fire. When legislators get data on scholarship program expenditures, advocates should provide additional data that gives context to those expenditures. Student success stories, program completion metrics, learning outcomes data, post- release employment outcomes, and other data can help explain the importance of spending money on justice-involved students. Without such information, legislators may be prone to see such expenditures as money that can be better spent. Policy advocates should be especially vigilant during economic downturns or recessions. During these times, state officials scrutinize state spending, and in the cases of Indiana, and perhaps Kentucky, state legislators cut financial aid to justice-involved students to save money or to redirect funds to non-justice-involved students. Thus, advocates should be watching for such 222 proposals during recessions and should be prepared to oppose them through organized, concerted efforts. Though developing a complete strategy for such is beyond the scope of this study and my expertise, supplying policymakers with evidence to support the use of state funds for justice- involved students may be persuasive. Policymakers reported they lacked key data that might have been important to have during debates on the merits of funding postsecondary education for justice-involved students. For example, when asked, most policymakers could not remember having data on: the estimated impact of taking aid away from justice-involved students, the rehabilitative effects of higher education, or the number of justice-involved students enrolled in higher education that would be affected. One legislator from Oklahoma specifically lamented, in retrospect, the lack of available data at the time about the positive effects of prison higher education. Having more data about justice-involved students and their educational needs might have been persuasive during the debates on whether to maintain scholarship eligibility for these students. Fortunately, social science research has improved in quality and has increasingly investigated the merits of higher education for people in prison. For decades, evidence has mounted on the conclusion that prison higher education works. Namely, evidence repeatedly shows that participating in higher education programs in prison increases employability and reduces criminal recidivism (Davis et al., 2013). New scholarly pursuits are aimed at measuring other outcomes of prison higher education, like learning outcomes, critical thinking, problem solving, self-awareness, positive identity, and resilience (Castro & Zamani-Gallaher, 2018; Institute for Higher Education Policy, 2018; Pike & Hopkins, 2019). The evaluations that will be published in the near future from the Second Chance Pell program will offer some of the best available data on the effects of financial aid on incarcerated students. These types of data, which 223 were not widely available to legislators in the 1990s, could be incredibly influential if presented to state legislators who are debating eliminating aid for justice-involved students now. That means, of course, advocates must get this information in front of legislators in formats that can be effectively persuasive. Finally, legislators have shown a historical lack of enthusiasm in advocating for justice- involved students and to debating issues of their funding. In most of the cases that I observed, there was little, if any, debate on the merits of keeping or eliminating aid for justice-involved students, with Illinois being the exception in 1989. The requirements were quietly added to bills or program rules, prompting few questions or reactions. As some legislators described, it was not a “big deal,” and in other cases, political attitudes led to active support for taking away aid. This might be due to how policymakers had socially constructed justice-involved students as a target population. Schneider, Ingram and deLeon (2014) suggested that even when good evidence of a policy’s effectiveness is presented, policymakers will struggle to shake their conceptions about negatively constructed populations: Social constructions emerge from emotional and intuitive reactions and then are justified with selective attention to evidence. Policymakers, especially elected politicians, respond to (and exploit) these emotional and intuitive judgements in their rationales and selection of policy elements. (emphasis added, Schneider, Ingram and deLeon, 2014, p. 57) Unless advocates get involved, it is plausible to predict that legislators may try to eliminate aid eligibility for justice-involved students in the future, with little substantive debate. It is up to advocates to force policymakers to debate the issues by presenting a counterargument to the “not a big deal” argument and by challenging their preconceived notions about justice-involved 224 people. The use of data, social science research, voices of justice-involved students, and higher education leaders may be useful in stirring debate on the issue. 225 CHAPTER 4: CONCLUSION At the time of this writing, in early 2019, criminal justice reform is poised to be one of the few bipartisan policy platforms that state and federal legislators can agree upon. Though some politicians still speak in tough-on-crime rhetoric reminiscent of decades ago, more seem to acknowledge the failures of mass incarceration and the need to fix bad policies. The federal First Step Act and the voter-approved amendment to Florida’s constitution restoring voting rights to people with felony convictions are just two of 2018’s headline-grabbing criminal justice policy reforms (Grawert & Lau, 2019; Mak, 2018). Many other reforms, especially at the state level, have been implemented in recent years (American Civil Liberties Union, 2012; Porter, 2018). Perhaps the philosophical pendulum has swung once again from punishment to rehabilitation in America. Fortunately, there is evidence that these criminal justice reforms are spilling over onto higher education. Arguments to reverse the ban on Pell Grants for incarcerated students are gaining strength. In January 2019, a new report estimated that as many as 463,000 incarcerated people would be eligible for Pell Grants, and if half of them attended college in prison with the help of Pell Grants, post-release employment rates would increase by 10%, increasing earnings for those individuals by over $45 million in the first year (Oakford et al., 2019). The resulting reductions in criminal recidivism were estimated to save states over $365 million per year in incarceration costs (Oakford et al., 2019). The near constant pressure to reverse the Pell ban since it passed in 1994, paired with the successes of the Second Chance Pell program, may be starting to persuade politicians. Congressional Democrats have proposed to reverse the ban in its forthcoming Higher Education Act reauthorization bill, and some Republicans are expressing openness to the 226 proposal (Douglas-Gabriel, 2019). It may not be long before students can once again receive Pell Grants to pay for credit-bearing postsecondary courses while in prison. Much less discussion is devoted to state financial aid for justice-involved students, a missed opportunity for social justice advocates. Evidence from this dissertation shows that states have banned justice-involved students from receiving state scholarships since at least the late 1960s, with many implementing bans on aid to incarcerated students beginning in the 1990s. But even without the persistent pressure from advocates that has been directed towards the federal Pell ban, cracks are forming in these state-level bans on aid. New Jersey and Tennessee may become the first states to reverse their bans on some or all aid to incarcerated students. Although scholars and advocates have paid close attention to the federal Pell Grant ban for prisoners, this dissertation contends with the history of state financial aid policies that disenfranchise justice-involved students from aid eligibility, thus filling in a wide gap in the financial aid literature. I conducted a descriptive inventory of all such state policies in Phase I, and in Phase II, I explored the factors that influenced some of these policies’ adoption. In the final sections of this dissertation, I summarize key findings and implications. The State Policy Landscape For over 50 years, state governments have funded scholarship programs for state residents while denying eligibility to people with criminal histories. Until this study, this fact has gone largely ignored. Previously, there was no comprehensive inventory of state financial aid programs that are unavailable to justice-involved students. Perhaps this was understandable. As I discovered during data collection for Phase I, the eligibility requirements of state grant programs are not so easily found. While some states provide accessible and complete accountings of their programs’ eligibility rules, I had to search state statutes, regulations, program websites and 227 contact agency officials directly to find answers for most other states. For justice-involved students, their advocates, and policy researchers, the difficulty in finding and interpreting eligibility requirements is a barrier. The findings from this study break down this barrier by identifying the state grant programs that disenfranchise justice-involved students and by displaying them all in one place (see Table 12). The purpose of the Phase I descriptive study was to uncover the state financial aid policy landscape for justice-involved students. This meant identifying grant programs, recording their eligibility requirements, and analyzing the requirements that explicitly targeted students with criminal histories. In total, 131 grant programs across 26 states are currently off-limits to at least one population of justice-involved student, including currently incarcerated students, students with any criminal conviction, or students with a drug conviction. Over a third of state grant programs, therefore, cannot be accessed by otherwise eligible college students, a staggering proportion considering how rarely researchers of state financial aid programs ever acknowledge these requirements. Additional research is needed to estimate the number of justice-involved students impacted by these state-level bans on aid. I also researched the history of these policies by finding when they were first created and when the rules for justice-involved students were first introduced. Unlike the ban on Pell Grants for prisoners that occurred in a single, swift moment in 1994, justice-involved students across the country lost state aid eligibility over time, spanning from the 1960s to as recently as 2011. With each policy decision occurring in its own time and context, it is difficult to detect any clear patterns in when or where these policies were adopted, though most of the bans on aid to justice- involved students occurred during the 1990s in the South and the Midwest. 228 At the end of Chapter 2, I outlined the major implications of the Phase I results. Chiefly, the denial of aid eligibility to justice-involved students is strikingly common. That so many researchers have ignored these eligibility requirements in their analyses of state grant programs is evidence that the justice-involved student population is among the most marginalized in American higher education. With each of the policies and their eligibility requirements carefully described, I offer to financial aid scholars a new research agenda. We need to understand the social, financial, and behavioral impacts of denying aid to this population. Better yet, what are the benefits of restoring aid eligibility to justice-involved students, and would the benefits outweigh the costs? New estimates of the effects of restoring Pell Grants to prisoners suggest that restoring state aid would also have major positive effects (Oakford et al., 2019). In addition to implications for research, students and their advocates can benefit from this study. Having identified the programs and their rules, individuals no longer have to search themselves (though there remains a need to keep this information updated). My intention is to share this information in a variety of formats such that anyone with internet access – not just academics – can find it. Several organizations are currently researching similar issues, and together, and we can distribute helpful information to college students and their advocates. Finally, policymakers can benefit from Phase I findings. Primarily, just less than half of states do not deny eligibility for state grants to justice-involved students. The half that do can see from these other states that it is politically and financially possible to give state aid to justice- involved students. Furthermore, some states are intentional in making aid available to justice- involved students, like Wisconsin and the states that support exonerees (see Table 5). For the first time, two states are considering reversing the ban on aid to incarcerated students, which suggests that it is possible to restore aid eligibility to justice-involved students. 229 State Policy Adoption The Phase I results identified and described the state grant policies that disenfranchise justice-involved students, but this descriptive information is not enough to draw conclusions about how or why state legislators created these policies. This was the pursuit of Phase II. Guided by Hearn, McLendon, and Linthicum’s (2017) conceptual model of state policy innovation and diffusion in higher education, I examined the factors that influenced the adoption of a sample of state financial aid grant programs. Using multiple case study design, I collected policy documents, historical records, media reports, and interviewed key policy actors to understand how and why the policies were adopted. Of the seven states in the sample, the clearest insights came from Illinois, Pennsylvania, Michigan, Georgia and Indiana. The conceptual model accurately predicted several of the key factors that influenced policy adoption (see Table 10). In Illinois, tough-on-crime political ideology and issues of state funding for higher education led state legislators to implement a state-wide ban on all aid to incarcerated students in 1989. In Georgia, tough-on-crime political ideology, gubernatorial strength, electoral conditions, and state funding levels influenced Governor Zell Miller’s decision to ban incarcerated students from receiving HOPE Grants in 1995. In Indiana, the Great Recession and associated economic and state funding issues led to the ban on Higher Education Awards to incarcerated students in 2011. These cases contribute supporting evidence for the conceptual model and add unique case studies to the literature on state higher education policy adoption. Two of the state cases, however, required explanations that were not represented in the model. In Pennsylvania, a series of court cases in the early 1970s prompted state officials to change the eligibility requirements for state grant applicants several times up until 1988, when 230 officials decided to stop considering criminal history altogether except for students currently incarcerated. This type of coercive influence from a higher governmental authority is called vertical diffusion. While horizontal diffusion is represented in the model, vertical diffusion is not. I propose adding vertical diffusion to the model and suggest that scholars consider other cases where vertical diffusion has influenced state higher education policymaking. In Michigan, incarcerated students lost aid eligibility for two scholarship programs – the Michigan Tuition Grant and the Competitive Scholarship – in 1981 during a “policy patching” process. Policy patching occurs when policymakers tinker with policies within a policy mix to make them more consistent with the central policy goal (Howlett & Rayner, 2013). In the case of Michigan, four separate scholarship programs were in existence by 1978, and a state auditor determined that there were many inconsistencies in the eligibility requirements amongst them, one of which was a rule requiring applicants to be of good moral character. During the tinkering process, legislators turned these rules into explicit bans on aid to incarcerated students. This type of policy tinkering is not represented in the conceptual model and is seldom described in education policy research. I propose adding policy patching to the conceptual model and considering other cases where such tinkering led to substantive changes in state policies. As described in Chapter 3, there are several implications of the Phase II study. The main purpose of Phase II to was to study the extent to which the conceptual model predicted the factors that influenced the adoption of these state grant programs. In summary, the model held up well, though two additional factors were proposed to expand the model. The two factors – vertical diffusion and policy patching – are not new to the policy design literature at large but have not been explored in depth in the higher education policy literature. Additional research is needed to find whether these factors are commonly influential in the creation of higher education 231 policies. Policy patching, especially, is a newer concept and can benefit from additional in-depth case studies to better define it and to extend its theoretical implications. This study also adds to the literature on state policy adoption in higher education because of its use of qualitative methods. The longstanding tradition in policy adoption research was to utilize quantitative methods – like event history analysis – to explore policy diffusion and innovation. Like several scholars before me, I instead interviewed policymakers to hear first- hand accounts of the policy adoption process to better understand the factors and contexts at play. This method proved successful and should be considered in future policy adoption studies. In addition to the implications for policy theory and research, practical implications are apparent for policymaking. Advocates for justice-involved students can learn from these cases as to how and why legislators adopted bans on aid eligibility for justice-involved students. This knowledge can be used to safeguard against future attacks on aid. For example, when legislators learned how much scholarship money was being spent on incarcerated students, they cut that aid in Illinois, Georgia, and Indiana. Should this information come to light in other states, policy advocates must be prepared to advocate against cutting the aid by providing evidence of the positive effects of that aid. Similarly, there appeared to be little resistance or even debate on the policy proposals to eliminate aid to justice-involved students. Today, advocates should not let such proposals pass without a fight. Policy Recommendations In addition to the recommendations that I described throughout Chapters 2 and 3, I conclude my dissertation with a more definitive proposal for policy changes. In short, I advocate for the complete elimination of eligibility rules in state statutes and regulations that prevent 232 justice-involved students from receiving state financial aid. More precisely, I propose the following: 1. California, New York, New Jersey, Pennsylvania, Illinois, and Indiana are among the top ten states that spend the most money annually on undergraduate need- or need-and-merit- based financial aid (NASSGAP, 2019). Yet, in these states, incarcerated students are ineligible for their largest programs, including the Cal Grant, New York Tuition Assistance Program, New Jersey Tuition Aid Grant, Pennsylvania State Grant, Illinois Monetary Award Program, and Indiana Frank O’Bannon Grant. Reversing the policies in these states that disenfranchise incarcerated students from financial aid should be the priority of higher education advocates. Restoring state grant aid to the thousands of qualified incarcerated students in these states could yield vast benefits for society and individuals and would likely have small impacts on state budgets (see Oakford et al., 2019). If New Jersey succeeds in doing so (O’Dea, 2018), these other states should be first to follow. 2. Nine states could benefit from eliminating bans on financial aid to drug offenders. Of all the eligibility rules explored in this study, the drug offense rules are the most cumbersome to interpret and enforce and have the most variations. Some are one-year bans following a felony drug conviction, others include misdemeanors, and still others are permanent bans on drug convictions. Some require students to pledge to stay “drug- free,” though state officials were largely unable to explain what triggers the suspension of aid or their process for doing so, if there is any enforcement action at all. Other states tie state eligibility to federal financial aid eligibility under the 2005 drug conviction rule (Deficit Reduction Act, 2005). Eliminating the drug rules from financial aid programs in 233 AR, GA, IN, OK. MD, MO, SC, TN, and TX would likely eliminate confusion and burdensome enforcement actions at state agencies. Additional research is needed to estimate how many students would become available for aid in these states if the drug rules were eliminated, though the financial impact is likely relatively small. 3. I call on state agency officials to improve their websites and promotional materials on their state grant programs. Some states make no mention of the fact that their grant programs are unavailable to certain justice-involved students. First, all the eligibility requirements should be plainly listed. Second, detailed Frequently Asked Questions (FAQ) should be provided to help justice-involved students to understand if they qualify. This is especially important for states that have vaguely worded rules, like those on “good moral character” or rules that affect only people with certain drug convictions at certain times. Some of the state agency officials who I talked to by phone or email did not have clear answers to my questions about eligibility for justice-involved students, which certainly poses a problem for students who are trying to apply for aid. Officials must ensure they have clear interpretations of their own program rules and have fair, established procedures for processing applications from justice-involved students. Specifically, the following state agencies must improve their websites and program materials: a. Indiana Commission for Higher Education: https://www.in.gov/che/2745.htm b. Louisiana Office of Student Financial Assistance: https://www.osfa.la.gov/tops_mainlink.html c. Maryland Higher Education Commission: https://mhec.maryland.gov/preparing/Pages/FinancialAid/descriptions.aspx 234 d. Mississippi Office of Student Financial Aid: https://www.msfinancialaid.org/help/ e. New Jersey Higher Education Student Assistance Authority: https://www.hesaa.org/Pages/NJGrantsHome.aspx f. New York Higher Education Services Corporation: https://www.hesc.ny.gov/ g. Ohio Department of Higher Education: https://www.ohiohighered.org/sgs h. South Carolina Higher Education Tuition Grants Commission: https://sctuitiongrants.org/ i. Tennessee Higher Education Commission & Student Assistance Corporation: https://www.tn.gov/collegepays/financial-aid.html j. Utah System of Higher Education: https://stepuputah.com/ If state governments are serious about improving educational access, implementing these proposals would be historic. Now more than ever, restoring aid and providing better support to justice-involved people for college are politically and sociably feasible. Conclusion For the first time, this dissertation described the state financial aid landscape for justice- involved students in the United States. In 26 states, justice-involved students are ineligible for 131 grants, with some states or individual programs being more restrictive than others. For student success advocates, this should be alarming. With so much attention paid to the loss of Pell Grants for prisoners in 1994, it is remarkable that so little attention has been paid to the states that also ban incarcerated students from getting any state grants, including Pennsylvania, New York, New Jersey, Georgia, Tennessee and Kentucky. It is time to give state programs more attention, since restoring state grant aid to justice-involved students could make the difference between attending and not attending college for many people. 235 The policy descriptions, historical analyses, and first-hand accounts from policymakers presented in this study were intended to build a new knowledge base, contribute to the research literature on financial aid and state policy adoption, advance the use of a new conceptual model, and most importantly, advance a policy agenda whereby justice-involved students can once again get state financial aid for college. Whether students lost this aid because of economic and financial pressures, political or ideological motives, coercive influences, or other factors, 2019 could be the year when the trend starts to reverse. Rather than more states denying aid to justice- involved students, New Jersey and Tennessee could be the first to restore aid eligibility to some or all incarcerated students. I hope this study raises awareness of the plight of justice-involved students such that more states will restore grant aid eligibility to justice-involved students. 236 APPENDICES 237 Table 12: State Grant Policies with Rules for Justice-Involved Students APPENDIX A State Financial Aid Policy State Year Adopted269 Year of JI Rule270 Academic Challenge Scholarship Cal Grant Competitive Award Cal Grant Transfer Entitlement Award Cal Grant C Award Cal Grant High School Entitlement Award Middle Class Scholarship Student Excellence Equals Degree Act Florida Academic Scholars Florida Medallion Scholars Gold Seal Vocational Scholars Gold Seal CAPE Scholars HOPE Grant HOPE GED Grant HOPE Scholarship at Private Institutions Zell Miller Scholarship Program HOPE Career Grant Zell Miller Grant Program Georgia HERO Scholarship AR CA CA CA CA CA DE FL FL FL FL GA GA GA GA GA GA GA 1991 1955 1968 1973 1975 2013 2005 1997 1997 1997 2016 1993 1993 1996 2011 2013 2014 2005 1991/2009 2000 2000 2000 2000 2013 2005 1997 1997 1997 2016 1995 1995 1996 2011 2013 2014 2015 Eligibility Rules of Justice-Involved Students (Target Populations) No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge x x x x x x x x x x x x x x x x x x x x x x x x x x 269 The year the grant program was first created in state law or regulations. 270 The year that rules affecting justice-involved students first appeared in state law or regulations. An entry with two dates indicates that two rules affecting justice-involved students were created in different years. For example, the Arkansas Academic Challenge Scholarship was created in 1991 with a drug-free pledge rule, but the incarcerated rule was not added until 2009. 238 Table 12 (cont’d) State Financial Aid Policy State Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge Tuition Equalization Program HOPE Scholarship at Public Institutions GA Public Safety Memorial Grant Monetary Award Program Grant Program for Dependents of Police of Fire Officers Grant Program for Dependents of Correctional Officers MAP Plus Grant Program for Child Raised by Grandparent Community College Transfer Grant Grant Program for Medical Assistants in Training 21st Century Scholarship Higher Education Award Freedom of Choice Grant Kentucky Tuition Grant College Access Program (CAP) Go Higher Grant Kentucky Educational Excellence Scholarship (KEES) TOPS Opportunity Award TOPS Performance Award TOPS Honors Award TOPS Tech Award Senatorial Scholarship GA GA GA IL IL IL IL IL IL IL IN IN IN KY KY KY KY LA LA LA LA MD x x x x x x 1971 1990/2011 1993 1993/1995 1994 1967 1971 1973 2006 2006 2010 2015 1990 1965 1973 1972 1990 2007 1998 1989 1989 1989 1998 1978 1994/2015 1989 1989 1989 2006 2006 2010 2015 1990 2011 2011 1996 1996 1996 1998 1989 1989 1989 1998 1991 239 x x x x x x x x x x x x x x x x x x x x State Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge Table 12 (cont’d) State Financial Aid Policy Delegate Scholarship Edward T. and Mary A. Conroy Memorial Scholarship & Jean B. Cryor Memorial Scholarship Jack F. Tolbert Memorial Grant Howard P. Rawlings Guaranteed Access Grant Howard P. Rawlings Education Assistance Grant Howard P. Rawlings Campus Based EAG Part-Time Grant 2+2 Transfer Scholarship Veterans of the Afghanistan and Iraq Conflicts Scholarship Guaranteed Access Partnership Program Michigan Competitive Scholarship Michigan Tuition Grant Children of Veterans Tuition Grant Tuition Incentive Program MD MD MD MD MD MI MI MI MI MO National Guard Tuition Assistance MO Access Missouri Financial Assistance Program Higher Education Academic Scholarship Advanced Placement Incentive Grant MO MO MO A+ Scholarship Program MO x x x x x x x x x x x x x x x x x x x x MD 1978 1991 MD 1978 1991 MD MD 1979 1991 1991 1991 MD 1991 1991 1991 1991 2000 2006 2016 1964 1966 1935 1987 1994 2007 1986 2011 1993 1991 1991 2000 2006 2016 1981 1981 2005 2010 1999 2007 2009 2011 1993/2011 240 Table 12 (cont’d) State Financial Aid Policy Higher Education Legislative Plan for Needy Students Grant Educational Opportunity Fund Grant Tuition Aid Grant Program Survivor Tuition Benefits Program Law Enforcement Officer Memorial Scholarship World Trade Center Scholarship Part-Time TAG for County College Students NJ STARS NJ STARS II Governor's Urban Scholarship Governor's Industry Vocations Scholarship Regents Awards for Children of Deceased and Disabled Veterans Veterans Tuition Award Aid for Part-Time Students (Tuition Awards for Part Time) NYS Memorial Scholarship Tuition Assistance Program (TAP) Scholarships for Academic Excellence World Trade Center Memorial Scholarship Military Enhanced Recognition Incentive and Tribute American Airlines Flight 587 Memorial Scholarship State MS NJ NJ NJ NJ NJ NJ NJ NJ NJ NJ NY NY NY NY NY NY NY 1997 1968 1969 1979 2001 2002 2003 2004 2006 2012 2013 1974 1984 1984 1989 1995 1996 2002 1997 1999 1999 1999 2001 2002 2003 2004 2006 2012 2013 1995 1995 1995 1995 1995 1996 2002 NY 2003 2003 NY 2005 2005 241 Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge x x x x x x x x x x x x x x x x x x x x Table 12 (cont’d) State Financial Aid Policy State Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge Part-Time TAP Flight 3407 Memorial Scholarship Achievement and Investment in Merit Scholarship NYS Part-Time Scholarship Enhanced Tuition Awards Excelsior Scholarship Ohio College Opportunity Grant Ohio War Orphans Scholarship Ohio Safety Officers College Memorial Fund Ohio Army National Guard Scholarship Choose Ohio First Scholarship Oklahoma Tuition Aid Grant Program Oklahoma's Promise (Higher Learning Access Program) Oregon Barber and Hairdresser Grant Pennsylvania State Grant Program Blind or Deaf Higher Education Beneficiary Grant POW/MIA Education Program Ready To Success Scholarship PA-TIP PATH Program Legislative Incentives for Future Excellence (LIFE) Dayco Products, Inc. Scholarship NY NY NY NY NY NY OH OH OH OH OH OK OK OR PA PA PA PA PA PA SC SC x x x x x x x x 2006 2009 2015 2017 2017 2017 1969 1955 1976 1977 2007 1971 2006 2009 2015 2017 2017 2017 1981/2003 2003 2003 2003 2007 1995 1992 1992/2003 1985 1966 1949 1972 2014 2012 1989 1998 1988 1985 1969 1988 1988 2014 2015 2016 1998 1999 242 x x x x x x x x x x x x x x x x x x x Table 12 (cont’d) State Financial Aid Policy Higher Education Tuition Grants Palmetto Fellows Scholarship South Carolina Hope Scholarship LIFE Enhancement Palmetto Fellows Enhancement S.C. Need-Based Grant Program Christa McAuliffe Scholarship Tennessee HOPE Access Grant Wilder-Naifeh Technical Skills Grant Tennessee HOPE Scholarship General Assembly Merit Scholarship Aspire Award TN HOPE Scholarship Nontraditional TN HOPE Foster Child Tuition Grant Helping Heroes Grant (EL) Tennessee STEP UP Scholarship Tennessee Promise Scholarship Tennessee Reconnect Grant Tennessee Student Assistance Award Toward Excellence Access and Success Grant Texas Educational Opportunity Grant Regent's Scholarship New Century Scholarship College Bound Scholarship Higher Education Adult Part-Time Student Grant State Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions Federal Drug-Free Rules Drug-Free Pledge SC SC SC SC SC SC TN TN TN TN TN TN TN TN TN TN TN TN TN TX TX UT UT WA WV 1970 1988 2001 2007 2007 1996 1986 2003 2003 2003 2003 2003 2005 2005 2008 2013 2013 2017 1976 1999 2001 2008 1999 2007 1999 2000 2000 2001 2007 2007 1996/2000 1990 2003 2003 2003 2003 2003 2005 2005 2008 2013 2013 2017 1990/1992 1999 2001 2008 2010 2007 1999 243 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Table 12 (cont’d) State Financial Aid Policy State Year Adopted Year of JI Rule No Criminal Convictions Not be No Drug Incarcerated Convictions WY Hathaway Scholarship - Honors Hathaway Scholarship - Performance WY WY Hathaway Scholarship - Opportunity Hathaway Scholarship - Provisional WY Hathaway Scholarship - Need Based WY 2006 2006 2006 2006 2006 2006 2006 2006 2006 2006 x x x x x x x x x x Federal Drug-Free Rules Drug-Free Pledge 244 APPENDIX B Brief Recruitment Email Subject: Research Interview on [PROGRAM NAME] Dear NAME, My name is Bradley Custer, and I am a doctoral candidate in higher and adult education at Michigan State University. I am writing to see if you would be willing to be interviewed for my dissertation research. I am studying the history of state financial aid programs, particularly how state legislators determined their eligibility criteria. In [YEAR], you sponsored the [PROGRAM NAME] which is why talking with you is so critical to my research. Attached to this email is a consent document, which provides more information about my study. I am only asking for about 30 minutes of your time for a confidential interview at your convenience. I am happy to answer any questions you have about my study before you agree to participate. Please let me know if you are willing to talk with me, and thank you for your consideration. Sincerely, Bradley D. Custer, MA PhD Candidate: Higher, Adult, & Lifelong Education Michigan State University custerbr@msu.edu Academia Profile 245 APPENDIX C 1. What was your role (if any) in the development of the [PROGRAM]? Interview Protocol for State Policymakers 2. The [PROGRAM] includes an eligibility rules that denies aid to [JI STUDENTS]. What was the purpose of including this eligibility rule? a. What was the motivation for including this rule? b. What information was considered? c. What was the expected impact of adopting this eligibility rule? 3. Can you tell me about the first-time you heard about this type of eligibility rule in Michigan or elsewhere? a. Who came up with the idea for that eligibility rule? b. Were you familiar with other financial aid programs in Michigan or outside of Michigan that contain this rule? i. If so, how did you learn about it? ii. How did this information influence your thinking about the program in Michigan? 4. Who were the major players in determining this eligibility rule? a. To what extent was the governor aware of this rule? What opinions, if any, did he/she express about it? b. Were there people or organizations who opposed this rule? i. If so, who were they and what were their arguments? c. Were there people or organizations who supported this rule? i. If so, who were they and what were their arguments? 5. Why do you believe the idea for that eligibility rule emerged in Michigan when it did? a. What else was going in Michigan at the time that this policy was being discussed? i. In the realm of higher education in Michigan? ii. In the realm of criminal justice in Michigan? b. What was the political climate like at the time, and to what extent did political ideology or partisanship play a role in the adoption of this rule? 6. At the time, what did you think about [JI PEOPLE] being able to get state grant aid? 7. Who do you believe was ultimately behind the passage of the [PROGRAM] with the eligibility rule? 8. If you were doing this project, what else would you want to know regarding how this eligibility rule in the [PROGRAM] came to be adopted? 9. Who else could I interview who would have first-hand knowledge about this topic? 246 APPENDIX D 1. What do you know about how the history of the [PROGRAM]? Interview Protocol for Other Policy Actors 2. What do you know about the rule in the [PROGRAM] that states that [JI PEOPLE] are ineligible? a. How did Michigan policymakers decide to include (or not include) that rule? b. What do you think the purpose was? 3. Who were the major players involved in determining the eligibility criteria? a. Individuals (elected officials, policy experts, etc.) b. Organizations (interest groups, higher education institutions, etc.) 4. What else was going in Michigan at the time that this policy was being discussed? a. In the realm of higher education in Michigan? b. In the realm of criminal justice in Michigan? 5. What was the political climate like at the time? 6. At the time, what did you think about [JI PEOPLE] being able to get state grant aid? 7. If you were doing this project, what else would you want to know regarding the story behind this eligibility rule? 8. Who else could I interview who would have first-hand knowledge about this topic? 247 REFERENCES 248 REFERENCES Aiken, 2017. Table 5. Jail and prison incarcerated populations by state over time [data table]. Retrieved from https://www.prisonpolicy.org/reports/jailsovertime_table_5.html Alexander, M. (2012). The new Jim Crow: Mass incarceration in the age of colorblindness. New York: The New Press. Allen, M. D., Pettus, C., & Haider-Markel, D. P. (2004). Making the national local: Specifying the conditions for national government influence on state policymaking. State Politics & Policy Quarterly, 4(3), 318-344. Alon, S. (2011). Who benefits most from financial aid? The heterogeneous effect of need‐based grants on students' college persistence. Social Science Quarterly, 92(3), 807-829. doi: 10.1111/j.1540-6237.2011.00793.x American Bar Association. (2017). List of House of Delegates resolutions. Retrieved from https://www.americanbar.org/news/reporter_resources/midyear-meeting-2017/house-of- delegates-resolutions.html American Civil Liberties Union. (2012). Promising beginnings: Bipartisan criminal justice reform in key states. Retrieved from https://www.aclu.org/report/promising-beginnings- bipartisan-criminal-justice-reform-key-states American Nonsmokers' Rights Foundation. (2018). Smokefree and tobacco-free U.S. and tribal colleges and universities. Retrieved from https://no-smoke.org/wp- content/uploads/pdf/smokefreecollegesuniversities.pdf Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does correctional treatment work? A clinically relevant and psychologically informed meta‐ analysis. Criminology, 28(3), 369-404. Aschenbrener, M. (2016). Federal financial aid policy: Then, now, and in the future. Retrieved from NASPA website: https://www.naspa.org/constituent-groups/posts/federal-financial- aid-policy-then-now-and-in-the-future Associated Press. (2003a, Oct. 17). Budget surplus tops $1b in Wyoming. Billings Gazette. Retrieved from www.billingsgazette.com Associated Press. (2003b, Nov. 4). Lawmakers discuss ideas for $1B budget surplus. Billings Gazette. Retrieved from www.billingsgazette.com Associated Press. (2012, May 22). Indiana prison education cuts cost Ball State jobs. Diverse Issues in Higher Education. Retrieved from https://diverseeducation.com/article/17088/ 249 Banks, C. (2005). Punishment in America. Santa Barbara, CA: ABC-CLIO. Barr, A., & Turner, S. E. (2013). Expanding enrollments and contracting state budgets: The effect of the Great Recession on higher education. The ANNALS of the American Academy of Political and Social Science, 650(1), 168-193. doi:10.1177/0002716213500035 Barr, A., & Turner, S. (2015). Out of work and into school: Labor market policies and college enrollment during the Great Recession. Journal of Public Economics, 124, 63-73. doi: 10.1016/j.jpubeco.2014.12.009 Bauer-Wolf, J. (2017, Sept. 6). Looming state-federal conflict on sex assault. Inside Higher Ed. Retrieved from www.insidehighered.com Baumgartner, F. R., Jones, B. D., & Mortensen, P. B. (2014). Punctuated equilibrium theory: Explaining stability and change in public policymaking. In P. A. Sabatier & C. M. Weible (Eds.), Theories of the public policy process (3rd ed.), (pp. 64-104). Boulder, CO: Westview Press. Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Berry, F. S., & Berry, W. D. (1990). State lottery adoptions as policy innovations: An event history analysis. American Political Science Review, 84(2), 395-415. Berry, F. S., & Berry, W. D. (2014). Innovation and diffusion models in policy research. In P. A. Sabatier & C. M. Weible (Eds.), Theories of the public policy process (3rd ed.), (pp. 290- 341). Boulder, CO: Westview Press. Berry, W. D., Ringquist, E. J., Fording, R. C., & Hanson, R. L. (1998). Measuring citizen and government ideology in the American states, 1960-93. American Journal of Political Science, 42(1), 327-348. Bettinger, E. (2012). Financial aid: A blunt instrument for increasing degree attainment. In A. P. Kelly & M. Schneider (Eds.), Getting to graduation: The completion agenda in higher education (pp. 157-174). Baltimore: Johns Hopkins University Press. Bettinger, E., Gurantz, O., Kawano, L., & Sacerdote, B. (2016). The long run impacts of merit aid: Evidence from California’s Cal Grant (Research Report No. w22347). Retrieved from National Bureau of Economic Research website: http://www.nber.org/papers/w22347 Bozick, R., Steele, J., Davis, L., & Turner, S. (2018). Does providing inmates with education improve postrelease outcomes? A meta-analysis of correctional education programs in the United States. Journal of Experimental Criminology, 14(3), 389-428. doi: 10.1007/s11292-018-9334-6 250 Bragg, R. (1998, Feb. 4). Georgia nears accord with inmates who say they were beaten. The New York Times. Retrieved from www.nytimes.com Burley, L. M. (1997). History repeats itself in the resurrection of prisoner chain gangs: Alabama's experience raises Eighth Amendment concerns. Law & Inequality, 15(1), 127- 155. Brown, M. D., & Ketzenberger, J. (2010). Resetting Indiana’s future: How the recession and the state’s new fiscal ties will drive the General Assembly’s budget-making (Policy Brief). Indianapolis: Indiana Fiscal Policy Institute. California Student Aid Commission. (2016). Cal Grant Handbook. Retrieved from https://www.csac.ca.gov/post/cal-grant-handbook Campbell, N., & Finney, R. Z. (2005). Mitigating the combined distributional consequences of the Georgia Lottery for Education and the HOPE scholarship. Social Science Quarterly, 86(3), 746-758. Carbonaro v. Reeher, 392 F. Supp. 753 (E.D. Pa. 1975). Castleman, B. L. (2014). The impact of partial and full merit scholarships on college entry and success: Evidence from the Florida Bright Futures scholarship program (No. 17). Retrieved from EdPolicy Works website: http://curry.virginia.edu/uploads/resourceLibrary/17_Castleman_All_or_Nothing.pdf Castro, E. L., Hunter, R. K., Hardison, T., & Johnson-Ojeda, V. (2018). The landscape of postsecondary education in prison and the influence of Second Chance Pell: An analysis of transferability, credit-bearing status, and accreditation. The Prison Journal, 98(4), 405–426. doi:10.1177/0032885518776376 Castro, E. L., & Zamani-Gallaher, E. M. (2018). Expanding quality higher education for currently and formerly incarcerated people: Committing to equity and protecting against exploitation. ASHE-NITE Paper Series. Retrieved from https://www.indiana.edu/~cece/wordpress/wp-content/uploads/2017/02/Expanding- Quality-Higher-Ed-FINAL-11.26.pdf Century Foundation. (2018). Summary and history of California financial aid programs. Retrieved from https://tcf.org/app/uploads/2018/03/CSAC_Appendix-6.pdf Chen, V. (2004). The Georgia HOPE Scholarship. Policy Perspectives, 11(1), 9-19. Clarke, M. (2012, March 15). Indiana cuts prison college courses. Prison Legal News. Retrieved from www.prisonlegalnews.org 251 Coghlan, E., McCorkell, L., & Hinkely, S. (2018, September). What really caused the Great Recession? (Policy Brief). Institute for Research on Labor and Employment. Retrieved from http://irle.berkeley.edu/what-really-caused-the-great-recession/ Cohen‐Vogel, L., & Ingle, W. K. (2007). When neighbours matter most: Innovation, diffusion and state policy adoption in tertiary education. Journal of Education Policy, 22(3), 241- 262. doi: 10.1080/02680930701269152 Cohen-Vogel, L., Ingle, W. K., Levine, A. A., & Spence, M. (2008). The “spread” of merit- based college aid: Politics, policy consortia, and interstate competition. Educational Policy, 22(3), 339-362. Commonwealth of Kentucky. (1996). FY 1996 Operative Budget. Retrieved from https://osbd.ky.gov/Archives/Pages/default.aspx Corporation of Haverford College v. Reeher, 329 F. Supp. 1196 (E.D. Pa. 1971). Corporation of Haverford College v. Reeher, 53 F.R.D. 374 (E.D. Pa. 1972). Cornwell, C., & Mustard, D. (2002). Race and the effects of Georgia's HOPE scholarship. In D. E. Heller, & P. Marin (Eds.), Who should we help? The negative social consequences of merit scholarships (pp. 57-72). Cambridge, MA: The Civil Rights Project at Harvard University. Cornwell, C., Mustard, D. B., & Sridhar, D. J. (2006). The enrollment effects of merit-based financial aid: Evidence from Georgia’s HOPE program. Journal of Labor Economics, 24(4), 761-786. Corrections to College California. (2018). Formerly incarcerated college student fund application. Retrieved from https://correctionstocollegeca.org/resources/formerly- incarcerated-college-student-fund-application Crawford, K. B. (2005). Collateral sanctions in higher education: A constitutional challenge to the Drug-Free Student Loan Provision of the Higher Education Act of 1998. University of Toledo Law Review, 36, 755-787. Cullen, F. T., & Jonson, C. L. (2011). Rehabilitation and treatment programs. In J. Q. Wilson & J. Petersilia (Eds.), Crime and Public Policy (pp. 293-344). New York: Oxford University Press. Cullen, F. T., Pealer, J. A., Fisher, B. S., Applegate, B. K., & Santana, S. A. (2002). Public support for correctional rehabilitation in America: change or consistency? In J. V. Roberts & M. Hough (Eds.), Changing attitudes to punishment: Public opinion, crime and justice (pp. 128-147). Portland: Willan. 252 Custer, B. D. (2013). Why college admissions policies for students with felony conviction are not working at one institution. College & University, 88(4), 28-36. Custer, B. D. (2016). College admission policies for ex-offender students: A literature review. Journal of Correctional Education, 67(2), 35-43. Custer, B. D. (2018a). Reconsidering policy barriers for justice-involved college students. Journal of College Access, 4(1), 51-63. Custer, B. D. (2018b). Criminal background checks in university housing: An analysis of housing discrimination and state laws. Journal of College & University Student Housing, 44(3), 14-29. Davis, L. M., Bozick, R., Steele, J. L., Saunders, J., & Miles, J. N. V. (2013). Evaluating the effectiveness of correctional education: A meta-analysis of programs that provide education to incarcerated adults (Report No. RR-266-BJA). Retrieved from RAND Corporation website: https://www.rand.org/pubs/research_reports/RR266.html Deficit Reduction Act of 2005, Pub. L. No. 109-171, § 8021, 120 Stat. 178 (2005). Delaney, J. A., & Ness, E. C. (2009, November). A state-level merit aid typology. Paper presented at the annual meeting of the Association for the Student of Higher Education, Vancouver, Canada. DeLoughry, T. J. (1992, August 5). The Higher Education Amendments of 1992: What they mean for colleges and students. The Chronicle of Higher Education. Retrieved from www.chronicle.com DeLuca, H. R., Miller, T. J., & Wiedemann, C. F. (1991). Punishment vs. rehabilitation: A proposal for revising sentencing practices. Federal Probation, 55, 37-45. Denial of Federal Benefits to Drug Traffickers and Possessors, 21 U.S.C. § 862 (1988). Deutsch, M. E., Douglass, A. A., & Strayer, G. E. (1948). A report of a survey of the needs of California in higher education. Retrieved from https://oac.cdlib.org/view?docId=hb2p3004kd&brand=oac4&chunk.id=meta Douglas-Gabriel, D. (2019, January 16). How lifting a federal ban on student aid for inmates could bolster state economies. The Washington Post. Retrieved from www.washingtonpost.com Dowden, C., & Andrews, D. A. (2000). Effective correctional treatment and violent reoffending: A meta-analysis. Canadian Journal of Criminology, 42, 449-467. Doyle, W. R. (2006). Adoption of merit-based student grant programs: An event history analysis. Educational Evaluation and Policy Analysis, 28(3), 259-285. 253 Dworkin, J. B. (2010). The Great Recession: Impacts on Indiana and beyond. Journal of the Indiana Academy of the Social Sciences, 14(4), 1-10. Dynarski, S. (2000). Hope for whom? Financial aid for the middle class and its impact on college attendance. National Tax Journal, 53(3), 629-662. Dynarski, S. (2002). Race, income, and the impact of merit aid. In D. E. Heller, & P. Marin (Eds.), Who should we help? The negative social consequences of merit scholarships (pp. 73-92). Cambridge, MA: The Civil Rights Project at Harvard University. Dynarski, S. (2004). The new merit aid. In C. M. Hoxby (Ed.), College choices: The economics of where to go, when to go, and how to pay for it (pp. 63-100). Chicago: University of Chicago Press. Eber, P. R. (1982). The ISSC’s statutory evolution: Scholarship and grant programs 1957-1982. A report of the Illinois Higher Education Student Assistance Commission. Eby-Ebersole, S. (Ed.). (1999). Signed, sealed, and delivered: Highlights of the Miller record. Macon, GA: Mercer University Press. Ellers, F. (2003). Progress and paradox: The Patton Years, 1995-2003. Louisville: Butler Book Publishing Services. Federal Bureau of Prisons. (2017). Inmate statistics. Retrieved from https://www.bop.gov/about/statistics/ Felony. (n.d.). In Wex, Legal Information Institute’s legal dictionary. Retrieved from https://www.law.cornell.edu/wex/felony Fuller, M. B. (2014). A history of financial aid to students. Journal of Student Financial Aid, 44(1), 42-68. Garn, M. A. (2005). Power, politics, and the 1997 restructuring of higher education governance in Kentucky. Retrieved from University of Kentucky Doctoral Dissertations (No. 353). Gehring, J. (2001, June 20). Groups decry student-aid law's anti-drug provision. Education Week. Available from ProQuest. Gehring, T. (1997). Post-secondary education for inmates: An historical inquiry. Journal of Correctional Education, 48(2), 46-55. Georgia House of Representatives. (1990). Journal of the house of representatives of the state of Georgia at regular session, Vol. 1. Retrieved from Georgia Government Publications website: http://dlg.galileo.usg.edu/cgi/ggpd?query=ca%3A%20%28%22ga%20l404%22%29&_cc =1 254 Georgia House of Representatives. (1998). Journal of the house of representatives of the state of Georgia at regular session, Vol. 2. Retrieved from Georgia Government Publications website: http://dlg.galileo.usg.edu/cgi/ggpd?query=ca%3A%20%28%22ga%20l404%22%29&_cc =1 Georgia Lottery Commission. (2018). History. Retrieved from: https://www.galottery.com/en- us/about-us/history.html Georgia Senate. (1998). Journal of the senate of the state of Georgia at regular session, Vol. 2. Retrieved from Georgia Government Publications website: http://dlg.galileo.usg.edu/ggp/ Georgia Student Finance Commission. (1993). Procedural guidelines for the HOPE program. Retrieved from the Georgia Student Finance Commission. Georgia Student Finance Commission. (1995). HOPE scholarship program: Highlights of improvements for 1995-96. Retrieved from Georgia Student Finance Commission. Georgia Student Finance Commission. (2011). Tuition equalization grant program. Regulations 600 et seq. https://gsfc.georgia.gov/regulations-2011-2012 Georgia Student Finance Commission. (2012). Tuition equalization grant program. Regulations 600 et seq. Retrieved from https://gsfc.georgia.gov/regulations-2012-2013 Georgia Student Finance Commission. (2015a). Georgia HERO scholarship program. Regulations 800 et seq. Retrieved from https://gsfc.georgia.gov/regulations-2015-2016 Georgia Student Finance Commission. (2015b). Georgia public safety memorial grant program. Regulations 1200 et seq. Retrieved from https://gsfc.georgia.gov/regulations-2015-2016 Gibbert, M., & Ruigrok, W. (2010). The “what” and “how” of case study rigor: Three strategies based on published work. Organizational Research Methods, 13(4), 710-737. doi: 10.1177/1094428109351319 Glater, J. D. (2018). To the rich go the spoils: Merit, money, and access to higher education. Journal of College and University Law, 43(2), 195-226. Glazer, Y. (1996). The chains may be heavy, but they are not cruel and unusual: Examining the constitutionality of the reintroduced chain gang. Hofstra Law Review, 24(4), 1195-1224. Goldrick-Rab, S., Kelchen, R., Harris, D. N., & Benson, J. (2016). Reducing income inequality in educational attainment: Experimental evidence on the impact of financial aid on college completion. American Journal of Sociology, 121(6), 1762-1817. doi: 0002- 9602/2016/12106-0003$10.00 255 Gonzales, J. (2018, August 27). Free Tennessee community college for adults program shatters expectations in its first year. Tennessean. Retrieved from www.tennessean.com Gordon, T. (2012). State and local budgets and the Great Recession (Policy Brief). Stanford, CA: Stanford Center on Poverty and Inequality. Retrieved from https://www.stanford.edu/group/recessiontrends-dev/cgi- bin/web/sites/all/themes/barron/pdf/StateBudgets_fact_sheet.pdf Gordon, I., Lewis, J., & Young, K. (1993). Perspectives on policy analysis. In M. Hill (ed.), The policy process: A reader (pp. 5-9). Hertfordshire, England: Harvester Wheatsheaf. Governor’s Commission on Drug Awareness and Prevention. (1990). Anti-drug legislative and budgetary review. Retrieved from Digital Library of Georgia website: https://dlg.usg.edu/record/dlg_ggpd_s-ga-bg630-pd71-bm1-b1990-bl4 Graham, E. R., Shipan, C. R., & Volden, C. (2013). The diffusion of policy diffusion research in political science. British Journal of Political Science, 43(3), 673-701. doi: 10.1017/S0007123412000415 Grawert, A., & Lau, T. (2019, January 4). How the FIRST STEP Act became law – and what happens next. Retrieved from Brennan Center for Justice website: https://www.brennancenter.org/blog/how-first-step-act-became-law-and-what-happens- next Hanson, R. K., Bourgon, G., Helmus, L., & Hodgson, S. (2009). The principles of effective correctional treatment also apply to sexual offenders a meta-analysis. Criminal Justice and Behavior, 36(9), 865-891. Harkreader, S., Hughes, J., Tozzi, M. H., & Vanlandingham, G. (2008). The impact of Florida’s Bright Futures Scholarship Program on high school performance and college enrollment. Journal of Student Financial Aid, 38(1), 5-16. Harvey, W. S. (2011). Strategies for conducting elite interviews. Qualitative Research, 11(4), 431-441. doi:10.1177/1468794111404329 Hearn, J. C., & Holdsworth, J. M. (2004). Federal student aid. In E. P. St. John & M. D. Parsons (Eds.), Public funding of higher education: Changing contexts and new rationales (pp. 40-59). Baltimore: Johns Hopkins University Press. Hearn, J. C., McLendon, M. K., & Linthicum, K. C. (2017). Conceptualizing state policy adoption and diffusion. In M. B. Paulsen (Ed.), Higher education: Handbook of theory and research (pp. 309-354). Springer International Publishing. doi: 10.1007/978-3-319- 48983-4_7 Heller, D. E. (2002). State merit scholarship programs: An introduction. In D. E. Heller & P. Marin (Eds.), Who should we help? The negative social consequences of merit 256 scholarships (pp. 15-23). Cambridge, MA: The Civil Rights Project at Harvard University. Higher Education Amendments of 1992, Pub. L. No. 102-325, § 401, 106 Stat. 481 (1992). Higher Education Amendments of 1998, Pub. L. No. 105-244, § 483, 112 Stat. 1736 (1998). Higher Education Opportunity Act of 2008, Pub. L. No. 110-315, § 401, 122 Stat. 3189 (2008). Horst, A. M. (2011). State of Indiana, List of appropriations. Retrieved from Indiana State Budget Agency website: www.in.gov/sba/files/ap_2011_all.pdf Howlett, M., & Rayner, J. (2007). Design principles for policy mixes: cohesion and coherence in ‘new governance arrangements’. Policy and Society, 26(4), 1-18. doi: 10.1016/S1449- 4035(07)70118-2 Howlett, M., & Rayner, J. (2013). Patching vs packaging in policy formulation: Assessing policy portfolio design. Politics and Governance, 1(2), 170-182. doi: 10.12924/pag2013.01020170 Hyatt, R. (1997). Zell: The governor who gave Georgia hope. Macon, Georgia: Mercer University Press. Illinois Higher Education Commission. (1957). Illinois looks to the future in higher education. Retrieved from https://www.hathitrust.org/ Illinois House of Representatives. (1989, June 16). Transcription Debate – 58th Legislative Day. Retrieved from http://www.ilga.gov/house/transcripts/htrans86/hts.html Illinois House of Representatives. (1989, Nov. 1). Transcription Debate – 75th Legislative Day. Retrieved from http://www.ilga.gov/house/transcripts/htrans86/hts.html Illinois Senate. (1989, Oct. 18). Senate Transcript – 63rd Legislative Day. Retrieved from http://www.ilga.gov/senate/transcripts/strans86/sts.html Illinois Senate. (1989, Oct. 19). Senate Transcript – 64th Legislative Day. Retrieved from http://www.ilga.gov/senate/transcripts/strans86/sts.html Illinois Senate. (1989, Nov. 2). Senate Transcript – 68th Legislative Day. Retrieved from http://www.ilga.gov/senate/transcripts/strans86/sts.html Illinois Student Assistance Commission. (2012). Monetary Award Program. Task force report. Retrieved from https://www.isac.org/about-isac/monetary-award-program-MAP-task- force/index.html 257 Indiana Commission on Higher Education. (2017). Indiana state financial aid manual: A guide to Indiana’s state financial aid programs. Retrieved from https://www.in.gov/che/files/Financial%20Aid%20Manual_10_23_17.pdf Indiana Department of Correction. (2018). Programs. Retrieved from https://www.in.gov/idoc/2799.htm Ingle, W. K., Cohen‐Vogel, L., & Hughes, R. (2007). The public policy process among southeastern states: Elaborating theories of regional adoption and hold‐out behavior. Policy Studies Journal, 35(4), 607-628. Ingle, W. K., & Petroff, R. A. (2013). Public entrepreneurs and the adoption of broad-based merit aid beyond the Southeastern United States. Education Policy Analysis Archives, 21(58), 1-26. Ingle, W. K., & Ratliff, J. R. (2015). Then and now: An analysis of broad-based merit aid initial eligibility policies after twenty years. Kentucky Journal of Higher Education Policy and Practice, 3(2), 1-31. Innocence Project. (2017). Compensation statutes: A national overview. Retrieved from https://www.innocenceproject.org/compensating-wrongly-convicted/ Institute for Higher Education Policy. (2018, July 19). New IHEP research initiative to assess impact of prison-based postsecondary education programs. Retrieved from IHEP website: http://www.ihep.org/press/news-releases/new-ihep-research-initiative-assess- impact-prison-based-postsecondary-education Jackson, R. L., & Covell, C. N. (2013). Sex offender civil commitment: Legal and ethical issues. In K. Harrison & B. Rainey (Eds.), The Wiley-Blackwell handbook of legal and ethical aspects of sex offender treatment and management (pp. 406-423). Chichester, West Sussex, UK: John Wiley & Sons, Ltd. Jacobs, J. B. (2015). The eternal criminal record. Cambridge, MA: Harvard University Press. Jeffries, E. (2010, February 17). FY 2010-11 higher education budget: S.B. 1157 governor’s recommendation. Michigan Senate Fiscal Agency. Retrieved from: https://www.legislature.mi.gov/documents/2009-2010/billanalysis/Senate/pdf/2009-SFA- 1157-G.pdf Jen, K. I. (2010, February 19). Higher education: FY 2010-11. Michigan House Fiscal Agency. Retrieved from: https://www.legislature.mi.gov/documents/2009- 2010/billanalysis/House/pdf/2009-HLA-1157-1.pdf Johnson, R., & Toch, H. (Eds.). (2000). Crime and punishment: Inside views. Los Angeles: Roxbury. 258 Johnson, N., & Yanagiura, T. (2012). Evaluation of Indiana’s financial aid programs and policies. Retrieved from State of Indiana website: https://www.in.gov/che/files/2012-09- 12_Indiana_Financial_Aid_Program_Evaluation(1).pdf Judge Upholds Ban on Pell Grant for Prisoners. (1995, March 10). The Chronicle of Higher Education. Retrieved from http://www.chronicle.com/article/Judge-Upholds-Ban-on- Pell/83979 Kaeble, D., & Glaze, L. (2016). Correctional populations in the United States, 2015 (NCJ 250374). Retrieved from U.S. Department of Justice, Bureau of Justice Statistics website: https://www.bjs.gov/content/pub/pdf/cpus15.pdf Karch, A. (2006). National intervention and the diffusion of policy innovations. American Politics Research, 34(4), 403-426. doi: 10.1177/1532673X06288202 Karch, A. (2012). Vertical diffusion and the policy-making process: The politics of embryonic stem cell research. Political Research Quarterly, 65(1), 48-61. doi: 10.1177/1065912910385252 Kern, F., Kivimaa, P., & Martiskainen, M. (2017). Policy packaging or policy patching? The development of complex energy efficiency policy mixes. Energy Research & Social Science, 23, 11-25. doi: 10.1016/j.erss.2016.11.002 Kreighbaum, A. (2018, March 7). New push to drop drug offenses as barrier to student aid. Inside Higher Ed. Retrieved from www.insidehighered.com Laird, C. A. (1971). A study of the college-level educational program of the Texas department of correction (Doctoral dissertation). University of Houston, Houston, TX. Lapp, K. (2012). Reforming the good moral character requirement for US citizenship. Indiana Law Journal, 87(4), 1571-1637. Lee, A. (1979). Audit Report: Student Financial Assistance Services, et al., Department of Education. Retrieved from the State of Michigan Office of the Auditor General. Lester, P. (1998). State fiscal note statement: Senate Bill 21. Retrieved from http://www.lrc.ky.gov/lrcsearch Levitt, S. D. (2004). Understanding why crime fell in the 1990s: Four factors that explain the decline and six that do not. The Journal of Economic Perspectives, 18(1), 163-190. Liebschutz, S. F. (1985). The national minimum drinking-age law. Publius: The Journal of Federalism, 15(3), 39-52. 259 Lipsey, M. W., & Cullen, F. T. (2007). The effectiveness of correctional rehabilitation: A review of systematic reviews. Annual Review of Law and Social Science, 3, 297-320. doi: 10.1146/annurev.lawsocsci.3.081806.112833 Liu, W. (2005). 10 year outlook. Retrieved from the State of Wyoming’s website: http://eadiv.state.wy.us/wef/outlook2005.pdf Loeb, S., Dynarksi, S., McFarland, D., Morris, P., Reardon, S., & Reber, S. (2017). Descriptive analysis in education: A guide for researchers. (NCEE 2017–4023). Washington, DC: U.S. Department of Education. Retrieved from https://ies.ed.gov/ncee/pubs/20174023/ Long, B. T. (2002). Do state financial aid programs cause colleges to raise prices? The case of the Georgia HOPE Scholarship. In D. E. Heller, & P. Marin (Eds.), Who should we help? The negative social consequences of merit scholarships (pp. 93-110). Cambridge, MA: The Civil Rights Project at Harvard University. Long, B. T. (2004). How do financial aid policies affect colleges? The institutional impact of the Georgia HOPE scholarship. Journal of Human Resources, 39(4), 1045-1066. Loughlin, S. (2011, June 3). IDOC cuts back on inmate education. Tribune Star. Retrieved from www.tirbstar.com Love, M. C., Roberts, J., & Klingele, C. (2013). Collateral consequences of criminal convictions: Law, policy and practice. Washington, DC: NACDL Press. Lovenheim, M. F., & Owens, E. G. (2014). Does federal financial aid affect college enrollment? Evidence from drug offenders and the Higher Education Act of 1998. Journal of Urban Economics, 81, 1-13. doi: 10.1016/j.jue.2014.01.005 Lytle, R. (2015). Variation in criminal justice policy-making: An exploratory study using sex offender registration and community notification laws. Criminal Justice Policy Review, 26(3), 211-233. doi: 10.1177/0887403413507274 Mahaffie, L. B. (2014, December 8). [Dear Colleague Letter]. Federal Pell Grant eligibility for students confined or incarcerated in locations that are not federal or state penal institutions (GEN-14-21). Retrieved from https://www2.ed.gov/policy/gen/guid/correctional-education/pell-letter.pdf Mak, T. (2018, November 7). Over 1 million Florida felons win right to vote with Amendment 4. NPR. Retrieved from www.npr.org Martinson, R. (1974). What works? Questions and answers about prison reform. The Public Interest, (35), 22-54. Martinson, R. (1979). New findings, new views: A note of caution regarding sentencing reform. Hofstra Law Review, 7(2), 243-258. 260 Marx, B. M., & Turner, L. J. (2015). Borrowing trouble? Student loans, the cost of borrowing, and implications for the effectiveness of need-based grant aid (No. 20850). Retrieved from National Bureau of Economic Research website: http://www.nber.org/papers/w20850 Mastrorilli, M. E. (2016). With Pell grants rising: A review of the contemporary empirical literature on prison postsecondary education. The Journal of Correctional Education, 67(2), 44-60. McKinney, L. (2009). An analysis of policy solutions to improve the efficiency and equity of Florida's Bright Futures scholarship program. Florida Journal of Educational Administration & Policy, 2(2), 85-101. McLendon, M. K., & Cohen-Vogel, L. (2008). Understanding education policy change in the American states: Lessons from political science. In B. S. Cooper, J. G. Cibulka, & L. D. Fusarelli (Eds.), Handbook of education politics and policy (pp. 30-51). New York: Routledge. McLendon, M. K., Cohen-Vogel, L., & Wachen, J. (2015). Understanding education policy making and policy change in the American states: Learning from contemporary policy theory. In B. S. Cooper, J. G. Cibulka, & L. D. Fusarelli (Eds.), Handbook of education politics and policy (2nd ed.), (pp. 86-117). New York: Routledge. McLendon, M. K., Heller, D. E., & Young, S. P. (2005). State postsecondary policy innovation: Politics, competition, and the interstate migration of policy ideas. The Journal of Higher Education, 76(4), 363-400. doi: 10.1080/00221546.2005.11772289 Michigan Department of Treasury. (2010). Tuition incentive program fact sheet: Academic year 2010-11. Retrieved from https://web.archive.org/web/20101203123906/http://www.michigan.gov/documents/Fact SheetTIP_161201_7.pdf Michigan Senate Analysis Section. (1980a, November 10). S.B. 1275, 1276, 1277, & 1278: First Analysis. Library of Michigan, Lansing, MI. Michigan Senate Analysis Section. (1980b, December 22). S.B. 1275, 1276, 1277, & 1278: Second Analysis. Library of Michigan, Lansing, MI. Mills, D. L. (1996). United States v. Johnson: Acknowledging the shift in the juvenile court system from rehabilitation to punishment. DePaul Law Review, 45(3), 903-941. MI Student Aid. (2017). Michigan Merit Award. Retrieved from the State of Michigan’s website: https://web.archive.org/web/20171129023233/http://www.michigan.gov/mistudentaid/0,4 636,7-128-60969_61016-274560--,00.html 261 Natapoff, A. (2018). Punishment without crime: How our massive misdemeanor system traps the innocent and makes America more unequal. New York: Basic Books. National Association of State Student Grant and Aid Programs. (2018). NASSGAP program quick finder [Data file]. Retrieved from https://www.nassgapsurvey.com/ National Association of State Student Grant and Aid Programs. (2019). Table 8 history: 2004- 2017 [Data file]. Retrieved from https://www.nassgapsurvey.com/ National Student Clearinghouse Research Center. (2011). National postsecondary enrollment trends: Before, during, and after the Great Recession. Retrieved from https://files.eric.ed.gov/fulltext/ED536112.pdf Neary, B. (2005, Nov. 24). Freudenthal calls for saving state surplus. Billings Gazette. Retrieved from www.billingsgazette.com Ness, E. C. (2008). Merit aid and the politics of education. New York: Routledge. Ness, E. C. (2010). The politics of determining merit aid eligibility criteria: An analysis of the policy process. The Journal of Higher Education, 81(1), 33-60. doi: 10.1353/jhe.0.0080 Newell, W. (2013). The legacy of Nixon, Reagan, and Horton: How the tough on crime movement enabled a new regime of race-influenced employment discrimination. Berkeley Journal of African-American Law & Policy, 15(1), 3-36. Norris, R. J. (2012). Assessing compensation statutes for the wrongly convicted. Criminal Justice Policy Review, 23(3), 352-374. doi: 10.1177/0887403411409916 Norton, I. (2008, August 5). Congress bans Pell Grants for sex offenders. The Chronicle of Higher Education. Retrieved from www.chronicle.com Oakford, P., Brumfield, C., Goldvale, C., Tatum, L., diZerega, M., & Patrick, F. (2019). Investing in futures: Economic and fiscal benefits of postsecondary education in prison. Retrieved from Vera Institute website: https://www.vera.org/publications/investing-in- futures-education-in-prison O’Dea, C. (2018, July 2). NJ close to giving prisoners access to state financial aid for college courses. NJ Spotlight. Retrieved from www.njspotlight.com Oliver, W. M., & Hilgenberg, J. F., Jr. (2006). A history of crime and criminal justice in America. Boston: Pearson. Orfield, G. (2002). Foreword. In D. E. Heller, & P. Marin (Eds.), Who should we help? The negative social consequences of merit scholarships (pp. xi-xiii). Cambridge, MA: The Civil Rights Project at Harvard University. 262 Page, J. (2004). Eliminating the enemy: The import of denying prisoners access to higher education in Clinton’s America. Punishment & Society, 6(4), 357-378. Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108(5), 937- 975. Palmer, T. (1975). Martinson revisited. Journal of Research in Crime and Delinquency, 12(2), 133-152. Patrick. F. Taylor Foundation. (2018). Louisiana TOPS. Retrieved from http://www.taylorplan.com/resources/louisiana-tops/ Patton, P. (1995). Inaugural address. Retrieved from http://governors.e- archives.ky.gov/_govpatton/_governor1997/address.htm Peloso, W. I. (1997). Les Miserables: Chain gangs and the cruel and unusual punishments clause. Southern California Law Review, 70, 1459-1511. Pennsylvania Higher Education Assistance Authority. (2017). Pennsylvania state grant program policy manual. Retrieved from https://www.pheaa.org/documents/grants/ph/2017-18- policy-manual.pdf Pfaff, J. F. (2012). The micro and macro causes of prison growth. Georgia State University Law Review, 28(4), 1-35. Pike, A., & Hopkins, S. (2019). Transformative learning: positive identity through prison-based higher education in England and Wales. International Journal of Bias, Identity and Diversities in Education, 4(1), 48-65. doi: 10.4018/IJBIDE.2019010104 Pilkington, E. (2018, September 9). US inmates mark end of prison strike with push to regain voting rights. The Guardian. Retrieved from www.theguardian.com Pittman, N. A. (2017). Arkansas Scholarship Lottery. In G. Lancaster (Ed.), The Encyclopedia of Arkansas History & Culture. Retrieved from http://www.encyclopediaofarkansas.net/encyclopedia/entry-detail.aspx?entryID=7843 Porter, N. D. (2018). Top trends in state criminal justice reform, 2017. Retrieved from The Sentencing Project website: https://www.sentencingproject.org/publications/top-trends- state-criminal-justice-reform-2017/ Pratt, T. C. (2009). Addicted to incarceration: Corrections policy and the politics of misinformation in the United States. Thousand Oaks, CA: SAGE. Rampey, B.D., Keiper, S., Mohadjer, L., Krenzke, T., Li, J., Thornton, N., & Hogan, J. (2016). Highlights from the U.S. PIAAC survey of incarcerated adults: Their skills, work experience, education, and training: Program for the international assessment of adult 263 competencies: 2014 (NCES 2016-040). Retrieved from https://nces.ed.gov/pubs2016/2016040.pdf Rampolla, M. (2012). A pocket guide to writing in history (7th ed.). Boston: Bedford/ St. Martin’s. Rayner, J., & Howlett, M. (2009). Introduction: Understanding integrated policy strategies and their evolution. Policy and Society, 28(2), 99-109. doi:10.1016/j.polsoc.2009.05.001 Reichardt, K., Negro, S. O., Rogge, K. S., & Hekkert, M. P. (2016). Analyzing interdependencies between policy mixes and technological innovation systems: the case of offshore wind in Germany. Technological Forecasting and Social Change, 106, 11-21. doi: 10.1016/j.techfore.2016.01.029 Resch, T., & Hall, H. (2002). Attrition, completion, and graduation rates in Georgia technical colleges before and after the initiation of the HOPE grant. Journal of Vocational Education Research, 27(3), 277-291. Rhode, D. L. (1985). Moral character as a professional credential. Yale Law Journal, 94(3), 491- 603. Rogers, D. L., & Ferguson, C. J. (2011). Punishment and rehabilitation attitudes toward sex offenders versus nonsexual offenders. Journal of Aggression, Maltreatment & Trauma, 20(4), 395-414. doi: 10.1080/10926771.2011.570287 Rogge, K. S., & Reichardt, K. (2016). Policy mixes for sustainability transitions: An extended concept and framework for analysis. Research Policy, 45(8), 1620-1635. doi: 10.1016/j.respol.2016.04.004 Sample, L. L., & Kadleck, C. (2008). Sex offender laws: Legislators' accounts of the need for policy. Criminal Justice Policy Review, 19(1), 40-62. doi: 10.1177/0887403407308292 Scher, M. R. (2017). From penal to postsecondary: Navigating the college-corrections partnership through Second Chance Pell (Undergraduate Honors Thesis, Wesleyan University). Retrieved from http://wesscholar.wesleyan.edu/etd_hon_theses/1724/ Schneider, A., & Ingram, H. (1993). Social construction of target populations: Implications for politics and policy. American Political Science Review, 87(2), 334-347. Schneider, A. L., Ingram, H., & deLeon, P. (2014). Democratic policy design: Social construction of target populations. In P. A. Sabatier & C. M. Weible (Eds.), Theories of the public policy process (3rd ed.), (pp. 105-150). Boulder, CO: Westview Press. Sentencing Project, The. (2017). Criminal justice facts. Retrieved from https://www.sentencingproject.org/criminal-justice-facts/ 264 Shepherd, J. M. (2002). Police, prosecutors, criminals, and determinate sentencing: The truth about truth‐in‐sentencing laws. Journal of Law and Economics, 45(2), 509-533. Shipan, C. R., & Volden, C. (2006). Bottom‐up federalism: The diffusion of antismoking policies from US cities to states. American Journal of Political Science, 50(4), 825-843. Shipan, C. R., & Volden, C. (2008). The mechanisms of policy diffusion. American Journal of Political Science, 52(4), 840-857. Shipan, C. R., & Volden, C. (2012). Policy diffusion: Seven lessons for scholars and practitioners. Public Administration Review, 72(6), 788-796. doi: https://doi.org/10.1111/j.1540-6210.2012.02610.x Sjoquist, D. L., & Winters, J. V. (2013). The effects of HOPE on post-college retention in the Georgia workforce. Regional Science and Urban Economics, 43(3), 479-490. Sjoquist, D. L., & Winters, J. V. (2015). State merit‐based financial aid programs and college attainment. Journal of Regional Science, 55(3), 364-390. Smith, G. D. (1990). Brief history of the satisfactory character/crime policy. Retrieved from Pennsylvania Higher Education Assistance Authority. South Carolina Commission on Higher Education. (n.d.). Dayco Products, Inc. scholarship guidelines. Retrieved from http://www.che.sc.gov/Students,FamiliesMilitary/PayingForCollege/FinancialAssistance Available/ScholarshipsGrantsforSCResidents/DaycoScholarshipProgram.aspx SpearIT. (2016). Keeping it REAL: Why Congress must act to restore Pell Grant funding for prisoners. University of Massachusetts Law Review, 11, 26-43. St. John, E. P., Musoba, G. D., & Simmons, A. B. (2003). Keeping the promise: The impact of Indiana's twenty-first century scholars program. The Review of Higher Education, 27(1), 103-123. St. John, E. P., Musoba, G. D., Simmons, A. B., & Chung, C. G. (2002). Meeting the access challenge: Indiana's Twenty-First Century Scholars Program. New Agenda Series, vol. 4, no. 4. Indianapolis: Lumina Foundation for Education. St. John, E. P., Musoba, G. D., Simmons, A., Chung, C. G., Schmit, J., & Peng, C. Y. J. (2004). Meeting the access challenge: An examination of Indiana's Twenty-first Century Scholars Program. Research in Higher Education, 45(8), 829-871. Stokes, K. (2012, July 4). What Indiana will miss with the state prisons’ college programs gone. Indiana Public Media. Retrieved from https://indianapublicmedia.org/stateimpact/2012/07/04/what-indiana-will-miss-with-the- state-prisons-college-programs-gone/ 265 Stranahan, H. A., & Borg, M. O. (2004). Some futures are brighter than others: The net benefits received by Florida Bright Futures scholarship recipients. Public Finance Review, 32(1), 105-126. SUCCESS Act, S. 2557, 114th Cong. (2016). Sweeten, G., & Apel, R. (2007). Incapacitation: Revisiting an old question with a new method and new data. Journal of Quantitative Criminology, 23(4), 303-326. Swisher, K. (2008). The troubling rise of the legal profession's good moral character. St. John's Law Review, 82(3), 1037-1085. Taylor, J. M. (1999). Pell Grants for prisoners. In B. G. Chevigny (Ed.), Doing time: 25 years of prison writing (pp. 107-112). New York: Arcade. Taylor, J. M. (2008). Pell Grants for prisoners. Why should we care? Straight Low Magazine, 9(2), 55-58. Retrieved from http://realcostofprisons.org/writing/Taylor_Pell_Grants.pdf Tellis, W. M. (1997). Application of a case study methodology. The Qualitative Report, 3(3), 1- 19. Tewksbury, R., Erickson, D. J., & Taylor, J. M. (2000). Opportunities lost: The consequences of eliminating Pell Grant eligibility for correctional education students. Journal of Offender Rehabilitation, 31(1-2), 43-56. doi: 10.1300/J076v31n01_02 Tewksbury, R., & Taylor, J. M. (1996). The consequences of eliminating Pell Grant eligibility for students in post-secondary correctional education programs. Federal Probation, 60(3), 60-63. Tracy, S. J. (2010). Qualitative quality: Eight “big-tent” criteria for excellent qualitative research. Qualitative Inquiry, 16(10), 837-851. doi: 10.1177/1077800410383121 Ubah, C. B. (2004). Abolition of Pell Grants for higher education of prisoners: Examining antecedents and consequences. Journal of Offender Rehabilitation, 39(2), 73-85. doi: 10.1300/J076v39n02_05 University of Georgia. (2018). Researching the legislative history of an enacted Georgia statute. Retrieved from the University of Georgia School of Law website: https://www.law.uga.edu/researching-legislative-history-enacted-georgia-statute U.S. Census Bureau. (2016). Quick facts. Retrieved from https://www.census.gov/quickfacts/fact/table/US/PST045216 266 U.S. Department of Education. (2016). Beyond the box: Increasing access to higher education for justice-involved individuals. Retrieved from http://www2.ed.gov/documents/beyond- the-box/guidance.pdf U.S. Department of Education. Office of Federal Student Aid. (2014). Federal student aid eligibility for students confined in adult correctional or juvenile justice facilities. Retrieved from https://studentaid.ed.gov/sa/sites/default/files/aid-info-for-incarcerated- individuals.pdf U.S. Department of Education. Press Office. (2016, June 24). 12,000 incarcerated students to enroll in postsecondary educational and training programs through Education Department's new Second Chance Pell pilot program. Retrieved from https://www.ed.gov/news/press-releases/12000-incarcerated-students-enroll- postsecondary-educational-and-training-programs-through-education-departments-new- second-chance-pell-pilot-program U.S. Department of Justice. Bureau of Justice Assistance. (n.d.). Denial of Federal Benefits (DFB) Program. Benefits that may be denied. Retrieved from https://www.bja.gov/Programs/dfbbenefits.html U.S. Government Accountability Office. (2005). Drug offenders: Various factors may limit the impacts of federal laws that provide for denial of selected benefits (GAO-05-238). Retrieved from https://www.gao.gov/new.items/d05238.pdf Utah Foundation. (2007, October). School testing results, 2006 & 2007: How Utah compares to other states (Research Report No. 681). Retrieved from http://www.utahfoundation.org/reports/school-testing-results-2006-2007-how-utah- compares-to-other-states/ Violent Crime Control and Law Enforcement Act of 1994, Pub. L. No. 103-332, § 20411, 108 Stat. 1828 (1994). Wagner, P., & Sawyer, W. (2018). States of incarceration: The global context 2018. Retrieved from the Prison Policy Initiative website: https://www.prisonpolicy.org/global/2018.html Walsh, M. (2008, April 29). Law restricting college aid for drug offenders upheld. Education Week: The School Law Blog. Retrieved from http://blogs.edweek.org/edweek/school_law/ Washington Student Achievement Council. (2017). College bound scholarship report. Retrieved from https://www.wsac.wa.gov/sites/default/files/2017.CBS.Report.pdf Weible, C. (2014). Introducing the scope and focus of policy process research and theory. In P. A. Sabatier & C. M. Weible (Eds.), Theories of the public policy process (3rd ed.), (pp. 3- 24). Boulder, CO: Westview Press. 267 Weimer, D. L. (1993). The current state of design craft: Borrowing, tinkering, and problem solving. Public Administration Review, 53(2), 110-120. Wellstead, A., & Howlett, M. (2017). Assisted tree migration in North America: policy legacies, enhanced forest policy integration and climate change adaptation. Scandinavian Journal of Forest Research, 32(6), 535-543. doi: 10.1080/02827581.2016.1249022 Wellstead, A., Rayner, J., & Howlett, M. (2016). Alberta's oil sands reclamation policy trajectory: the role of tense layering, policy stretching, and policy patching in long-term policy dynamics. Journal of Environmental Planning and Management, 59(10), 1873- 1890. doi: 10.1080/09640568.2015.1098594 Welsh, M. F. (2002). The effects of the elimination of Pell Grant eligibility for state prison inmates. Journal of Correctional Education, 53(4), 154-158. Wen, S. Y. (2009). Wyoming’s new business formation in 2006 and 2007. Wyoming Labor Force Trends, 46(3). Retrieved from http://doe.state.wy.us/LMI/0309/a1.htm Wisconsin Higher Educational Aids Board. (2018). Policies and procedures manual. Retrieved from http://heab.state.wi.us/finadmin/manual/index.html#4 Wormith, J. S., Althouse, R., Simpson, M., Reitzel, L. R., Fagan, T. J., & Morgan, R. D. (2007). The rehabilitation and reintegration of offenders: The current landscape and some future directions for correctional psychology. Criminal Justice and Behavior, 34(7), 879-892. Yin, R. K. (2009). Case study research: Design and methods (4th ed.). Thousand Oaks, CA: SAGE. Zhang, L. (2011). Does merit-based aid affect degree production in STEM fields? Evidence from Georgia and Florida. The Journal of Higher Education, 82(4), 389-415. Zhang, L., Hu, S., & Sensenig, V. (2013). The effect of Florida’s Bright Futures program on college enrollment and degree production: An aggregated-level analysis. Research in Higher Education, 54(7), 746-764. doi:10.1007/s11162-013-9293-8 Zook, J. (1994, August 10). Crime bill would bar Pell Grants to federal and state prisoners. The Chronicle of Higher Education. Retrieved from www.chronicle.com 268