Bail Decisions Research Articles

Confused Purposes and Inconsistent Adjudication: An Assessment of Bail Decisions in Delhi's Courts

Abstract A persistent issue in the Indian criminal justice system has been the over-incarceration of pre-trial detainees, which is inextricably linked with the practices of trial courts in deciding on the detention and release of accused persons pending trial. The present statutory law and judicial discourse provides little guidance on the process for deciding bail matters. There has also not been much effort to empirically research decision-making in such matters and/or the impact of these decisions on the pre-trial detainee population in prisons. The present study is an attempt to plug this gap, analysing ‘regular’ bail orders of Sessions Courts of Delhi available from the eCourts system between 2017 and 2019 for the offences of theft and rape. The data reveals a failure to recognise the fundamentally preventive purpose of bail, as well as to develop individualised and specialised processes in compliance with this purpose. Such failure in guidance has resulted in judges importing factors from the trial process that remain unjustified outside of the punitive detention process, such as guilt. In conclusion, we argue that a fundamental reimagination of the approach to bail is required – one that is distinct from the trial process and focuses instead on the individualised assessment of risk.

Learning Fair Policies for Multi-Stage Selection Problems from Observational Data

We consider the problem of learning fair policies for multi-stage selection problems from observational data. This problem arises in several high-stakes domains such as company hiring, loan approval, or bail decisions where outcomes (e.g., career success, loan repayment, recidivism) are only observed for those selected. We propose a multi-stage framework that can be augmented with various fairness constraints, such as demographic parity or equal opportunity. This problem is a highly intractable infinite chance-constrained program involving the unknown joint distribution of covariates and outcomes. Motivated by the potential impact of selection decisions on people’s lives and livelihoods, we propose to focus on interpretable linear selection rules. Leveraging tools from causal inference and sample average approximation, we obtain an asymptotically consistent solution to this selection problem by solving a mixed binary conic optimization problem, which can be solved using standard off-the-shelf solvers. We conduct extensive computational experiments on a variety of datasets adapted from the UCI repository on which we show that our proposed approaches can achieve an 11.6% improvement in precision and a 38% reduction in the measure of unfairness compared to the existing selection policy.

Content Analysis of Judges’ Sentiments Toward Artificial Intelligence Risk Assessment Tools

Objective: to analyze the positions of judges on risk assessment tools using artificial intelligence.Methods: dialectical approach to cognition of social phenomena, allowing to analyze them in historical development and functioning in the context of the totality of objective and subjective factors, which predetermined the following research methods: formal-logical and sociological.Results: Artificial intelligence (AI) uses computer programming to make predictions (e.g., bail decisions) and has the potential to benefit the justice system (e.g., save time and reduce bias). This secondary data analysis assessed 381 judges’ responses to the question, “Do you feel that artificial intelligence (using computer programs and algorithms) holds promise to remove bias from bail and sentencing decisions?”Scientific novelty: The authors created apriori themes based on the literature, which included judges’ algorithm aversion and appreciation, locus of control, procedural justice, and legitimacy. Results suggest that judges experience algorithm aversion, have significant concerns about bias being exacerbated by AI, and worry about being replaced by computers. Judges believe that AI has the potential to inform their decisions about bail and sentencing; however, it must be empirically tested and follow guidelines. Using the data gathered about judges’ sentiments toward AI, the authors discuss the integration of AI into the legal system and future research.Practical significance: the main provisions and conclusions of the article can be used in scientific, pedagogical and law enforcement activities when considering the issues related to the legal risks of using artificial intelligence.

Social Justice and Judicial Working of Bail Decisions in India: a Review by Advocate Parth Singh on the Legacy of Dr. B.R. Ambedkar

Social Justice and Judicial Working of Bail Decisions in India: a Review by Advocate Parth Singh on the Legacy of Dr. B.R. Ambedkar

The Right to Bail in Seychelles: Implementing Article 9(3) of the International Covenant on Civil and Political Rights

Abstract Seychelles acceded to the International Covenant on Civil and Political Rights (iccpr) in 1992. Article 9(3) of the iccpr provides for circumstances in which an accused may be released on bail. Article 18(7) of the Constitution of Seychelles (2008) provides for the grounds on which a court may decline to release an arrested or accused person on bail. Article 48 of the Constitution requires Seychellois courts to, inter alia, refer to international treaties and the practice/jurisprudence of international human rights bodies when interpreting constitutional rights. This article illustrates the drafting history of Article 9(3) of the iccpr and how it has been interpreted by the Human Rights Committee. It also highlights the right to bail under the African human rights system. Based on over 250 cases of the magistrate’s court, Supreme Court, Constitutional Court and Court of Appeal decided between 1998 and 2023, the article discusses the extent to which the courts’ jurisprudence complies with the practice and/or jurisprudence of the Human Rights Committee on the following issues: the right to bail, courts with jurisdiction to grant bail, factors courts consider in making bail decisions, burden of proof in bail matters, bail pending appeal and court’s power to revoke bail. It is argued, inter alia, that although the jurisprudence on bail substantially complies with the standards set by the Human Rights Committee, there are instances in which some of the courts’ decisions are contrary to these standards. The article recommends that courts should ensure that the jurisprudence on bail complies with the principles developed by the Human Rights Committee.

“The roughest form of social work:” How court officials justify bail decisions

AbstractGrowing research has analyzed quantitative patterns of bail decisions and outcomes, but we know far less about how court officials justify their bail decisions. To enhance understanding of how bail decisions—and their resulting pretrial outcomes—are generated, we interviewed 104 judges, prosecutors, and public defenders in a northeastern state. Court officials in our study reported three primary justifications at bail: ensuring defendants return to court, preventing crime, and lessening harm. The first two justifications have been suggested in the literature, but the latter is novel and encompasses two secondary justifications: lessening criminal legal system harm and lessening societal harm. We show how these justifications and the decisions they enable blend risk management with rehabilitation and emerge from court officials’ shared assumption of defendants’ social marginality but varied beliefs about what to do about such marginality pretrial. Each justification allows for distinct, but at times overlapping, bail decisions. We discuss the implications of our findings for theories of court official decision‐making, research on racial and socioeconomic inequality, and bail reform policy.

Content Analysis of Judges' Sentiments Toward Artificial Intelligence Risk Assessment Tools

Artificial intelligence (AI) uses computer programming to make predictions (e.g., bail decisions) and has the potential to benefit the justice system (e.g., save time and reduce bias). This secondary data analysis assessed 381 judges’ responses to the question, “Do you feel that artificial intelligence (using computer programs and algorithms) holds promise to remove bias from bail and sentencing decisions?” The authors created apriori themes based on the literature, which included judges’ algorithm aversion and appreciation, locus of control, procedural justice, and legitimacy. Results suggest that judges experience algorithm aversion, have significant concerns about bias being exacerbated by AI, and worry about being replaced by computers. Judges believe that AI has the potential to inform their decisions about bail and sentencing; however, it must be empirically tested and follow guidelines. Using the data gathered about judges’ sentiments toward AI, we discuss the integration of AI into the legal system and future research.

Machine learning in bail decisions and judges' trustworthiness.

The use of AI algorithms in criminal trials has been the subject of very lively ethical and legal debates recently. While there are concerns over the lack of accuracy and the harmful biases that certain algorithms display, new algorithms seem more promising and might lead to more accurate legal decisions. Algorithms seem especially relevant for bail decisions, because such decisions involve statistical data to which human reasoners struggle to give adequate weight. While getting the right legal outcome is a strong desideratum of criminal trials, advocates of the relational theory of procedural justice give us good reason to think that fairness and perceived fairness of legal procedures have a value that is independent from the outcome. According to this literature, one key aspect of fairness is trustworthiness. In this paper, I argue that using certain algorithms to assist bail decisions could increase three different aspects of judges' trustworthiness: (1) actual trustworthiness, (2) rich trustworthiness, and (3) perceived trustworthiness.

Introduction, Summary, Questions for the Future

Introduction, Summary, Questions for the Future

The early origins of judicial stringency in bail decisions: Evidence from early childhood exposure to Hindu-Muslim riots in India

We estimate the causal effects of judges’ exposure to communal violence during early childhood on pretrial detention rates by exploiting novel administrative data on judgments and detailed resumes of judicial officers born during 1955–1991. Our key result is that judges exposed to communal violence between ages 0 and 6 years are 16% more prone to deny bail than the average judge, with the impact being stronger for the experience of riots between ages 3 and 6 years. The observed judicial stringency is driven by childhood exposure to riots with a higher duration of state-imposed lockdowns and low riot casualties.

Pretrial release judgments and decision fatigue

Abstract Field studies in many domains have found evidence of decision fatigue, a phenomenon describing how decision quality can be impaired by the act of making previous decisions. Debate remains, however, over posited psychological mechanisms underlying decision fatigue, and the size of effects in high-stakes settings. We examine an extensive set of pretrial arraignments in a large, urban court system to investigate how judicial release and bail decisions are influenced by the time an arraignment occurs. We find that release rates decline modestly in the hours before lunch and before dinner, and these declines persist after statistically adjusting for an extensive set of observed covariates. However, we find no evidence that arraignment time affects pretrial release rates in the remainder of each decision-making session. Moreover, we find that release rates remain unchanged after a meal break even though judges have the opportunity to replenish their mental and physical resources by resting and eating. In a complementary analysis, we find that the rate at which judges concur with prosecutorial bail requests does not appear to be influenced by either arraignment time or a meal break. Taken together, our results imply that to the extent that decision fatigue plays a role in pretrial release judgments, effects are small and inconsistent with previous explanations implicating psychological depletion processes.

Measuring Racial Discrimination in Bail Decisions

We develop new quasi-experimental tools to measure disparate impact, regardless of its source, in the context of bail decisions. We show that omitted variables bias in pretrial release rate comparisons can be purged by using the quasi-random assignment of judges to estimate average pretrial misconduct risk by race. We find that two-thirds of the release rate disparity between White and Black defendants in New York City is due to the disparate impact of release decisions. We then develop a hierarchical marginal treatment effect model to study the drivers of disparate impact, finding evidence of both racial bias and statistical discrimination. (JEL J15, K42)

The Early Origins of Judicial Stringency in Bail Decisions: Evidence from Early-Childhood Exposure To Hindu-Muslim Riots in India

The Early Origins of Judicial Stringency in Bail Decisions: Evidence from Early-Childhood Exposure To Hindu-Muslim Riots in India

Police and Vulnerability in Bail Decisions

Since bail legislation was enacted in the 1970s, Australia has experienced a continual increase in the number of prisoners on remand. Amendments to bail legislation and police discretion have been shown to contribute to this increase. Further, an accused’s vulnerability affects whether they are granted or denied bail, with vulnerable people being more likely to be denied bail. Vulnerability in the criminal justice system refers to factors such as race, age, sex and socioeconomic status. Many vulnerable people have multiple intersecting vulnerabilities, which further compounds their contact with the justice system. This study employed a qualitative content analysis of bail legislation for the Australian states of New South Wales (NSW), Tasmania, and Victoria, along with key correlating second reading speeches. The aim was to better understand the way in which bail decision-makers, such as police, consider vulnerability when making decisions about bail, in particular, if and how they are legislated to consider factors relating to vulnerability. The research found that only police in NSW and Victoria were required to consider an accused’s vulnerability explicitly under the law. Although legislation may cater for varying vulnerabilities, intersecting vulnerabilities are not considered.

Measuring Racial Discrimination in Algorithms

Algorithmic decision-making can lead to discrimination against legally protected groups, but measuring such discrimination is often hampered by a fundamental selection challenge. We develop new quasi-experimental tools to overcome this challenge and measure algorithmic discrimination in pretrial bail decisions. We show that the selection challenge reduces to the challenge of measuring four moments, which can be estimated by extrapolating quasi-experimental variation across as-good-as-randomly assigned decision-makers. Estimates from New York City show that both a sophisticated machine learning algorithm and a simpler regression model discriminate against Black defendants even though defendant race and ethnicity are not included in the training data.

The Early Origins of Judicial Bias in Bail Decisions: Evidence from Early Childhood Exposure to Hindu-Muslim Riots in India

We estimate the causal effects of exposure to communal violence during early childhood on pretrial detention rates. We exploit novel administrative data on judgments and detailed resumes of judicial officers born during 1955-1991. Our baseline result is that judges exposed to communal violence between ages 0- and 6 years are 16 percent more prone to deny bail than the average judge. Heterogeneity analyses show that the impact is stronger for the experience of riots between the age 3- and 6 years and the age at first exposure at 4 and 5 years. The early exposure effect is further driven by judges exposed to riots that have resulted in lower casualties or duration. We conjecture that exposure to the state's riot de-escalation efforts during formative years has lasting effects on social preferences for strong law and order. We provide suggestive evidence in support of this interpretation. Our results, which are robust across generations, highlight the link between early experiences related to social disorder and later life decisions on law and order.

Uncertainty, risk and the use of algorithms in policy decisions: a case study on criminal justice in the USA

Algorithms are increasingly used in different domains of public policy. They help humans to profile unemployed, support administrations to detect tax fraud and give recidivism risk scores that judges or criminal justice managers take into account when they make bail decisions. In recent years, critics have increasingly pointed to ethical challenges of these tools and emphasized problems of discrimination, opaqueness or accountability, and computer scientists have proposed technical solutions to these issues. In contrast to these important debates, the literature on how these tools are implemented in the actual everyday decision-making process has remained cursory. This is problematic because the consequences of ADM systems are at least as dependent on the implementation in an actual decision-making context as on their technical features. In this study, we show how the introduction of risk assessment tools in the criminal justice sector on the local level in the USA has deeply transformed the decision-making process. We argue that this is mainly due to the fact that the evidence generated by the algorithm introduces a notion of statistical prediction to a situation which was dominated by fundamental uncertainty about the outcome before. While this expectation is supported by the case study evidence, the possibility to shift blame to the algorithm does seem much less important to the criminal justice actors.

Measuring Racial Discrimination in Algorithms

There is growing concern that the rise of algorithmic decision-making can lead to discrimination against legally protected groups, but measuring such algorithmic discrimination is often hampered by a fundamental selection challenge. We develop new quasi-experimental tools to overcome this challenge and measure algorithmic discrimination in the setting of pretrial bail decisions. We first show that the selection challenge reduces to the challenge of measuring four moments: the mean latent qualification of white and Black individuals and the race-specific covariance between qualification and the algorithm’s treatment recommendation. We then show how these four moments can be estimated by extrapolating quasi-experimental variation across as-good-as-randomly assigned decision-makers. Estimates from New York City show that a sophisticated machine learning algorithm discriminates against Black defendants, even though defendant race and ethnicity are not included in the training data. The algorithm recommends releasing white defendants before trial at an 8 percentage point (11 percent) higher rate than Black defendants with identical potential for pretrial misconduct, with this unwarranted disparity explaining 77 percent of the observed racial disparity in algorithmic recommendations. We find a similar level of algorithmic discrimination with regression-based recommendations, using a model inspired by a widely used pretrial risk assessment tool. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Changing the Rules on Bail: An Analysis of Recent Legislative Reforms in Three Australian Jurisdictions

Bail decisions are a high-volume and hugely consequential component of the Australian criminal justice system, and yet, laws governing access to bail have rarely been the subject of systematic analysis. This article sheds new light on how bail laws have changed and what this reveals about how and why governments employ the criminal law as a public policy tool. Working with a dataset of 71 statutes enacted in New South Wales, Queensland and Victoria during the 10-year period between 2009 and 2018, we employ a combination of quantitative and qualitative analysis to illuminate key features and patterns. Our main findings are that bail law remains an active site of statutory reform, and that the object of mitigating harm- risk routinely takes priority over the fundamental rights of the accused. As a consequence, the strong trajectory of contemporary bail law reform has been to restrict rather than expand access to bail.

Measuring Racial Discrimination in Bail Decisions

We develop new quasi-experimental tools to measure racial discrimination, due to either racial bias or statistical discrimination, in the context of bail decisions. We show that the omitted variables bias in observational release rate comparisons can be purged by using the quasi-random assignment of judges to estimate average race-specific misconduct risk. We find that approximately two-thirds of the average release rate disparity between white and Black defendants in New York City is due to racial discrimination. We then develop a hierarchical marginal treatment effects model to study the drivers of discrimination, finding evidence of both racial bias and statistical discrimination. Outcome-based tests of racial bias therefore omit an important source of racial discrimination in bail decisions, and cannot be used to rule out all possible violations of U.S. anti-discrimination law.