1320.35.
(a) It is the intent of the Legislature in enacting this section to understand and reduce biases based on race, ethnicity, gender, age, economic circumstances, and behavioral or developmental disabilities gender, income level, race, or ethnicity in pretrial release decisionmaking.(b) For the purposes of this section, the following terms have the following meanings:
(1) “Pretrial risk assessment tool” means an instrument used to determine the risks associated with individuals in the
pretrial context.
(2)“Pretrial services agency” means an agency or organization designated by the county to conduct pretrial risk assessments and provide other pretrial services to defendants.
(2) “Pretrial services agency” means a local public agency that elects to perform pretrial risk assessments on individuals and provides the assessment information to a court.
(3) “Release
conditions framework” means the guidelines used by the pretrail services agency and the court to categorize varying degrees of risk for purposes of recommending whether to release or detain a person, whether to impose pretrial release conditions on a person, and guidance regarding those conditions.
(3)“Validated” means that the
risk assessment tool has been demonstrated by
(4) “Validate” means using scientifically accepted methods to do measure both of the following:
(A) Accurately and reliably assess The accuracy and reliability of the risk assessment tool in assessing (i) the risk that an assessed person will fail to
appear in court as required and (ii) the risk to public safety due to the commission of a new criminal offense if the person is released before the adjudication of the current criminal offense for which they have been charged.
(B) Minimize disparate impact Any disparate effect or bias in the risk assessment tool based on race, ethnicity, gender, age, economic circumstances, and behavior or developmental disability. gender, race, or ethnicity.
(c) (1) Any pretrial risk assessment
tool used by a pretrial services agency shall be validated by January 1, 2021, and on a regular basis thereafter, but no less frequently than once every three years. A pretrial services agency may coordinate with the Judicial Council to validate a pretrial risk assessment tool.
(2) A pretrial risk assessment tool shall be validated using the most recent data collected by the pretrial services agency within its jurisdiction, or, if that data is unavailable, using the most recent data collected by a pretrial services agency in a similar jurisdiction within California.
(d) (1) In order to increase transparency, a pretrial services agency shall, with regard to a pretrial risk assessment tool that it utilizes, make the following information publicly available:
(A) Line items, scoring, and weighting,
as well as details on how each line item is scored, for each pretrial risk assessment tool that the agency uses.
(B) Validation studies for each pretrial risk assessment tool that the agency uses.
(2) A pretrial services agency shall, when selecting which pretrial risk assessment tool to utilize, ensure that the agency would be able to comply with paragraph (1) if that tool was selected.
(e) The Judicial Council shall maintain a list of pretrial services agencies that have satisfied the validation requirement described in subdivision (c) and complied with the transparency requirements described in subdivision (d).
(f)A pretrial services agency shall collect and retain all of the following information regarding any pretrial risk assessment tool that it uses:
(1)Input information, including, but not limited to, both of the following:
(A)The age, gender, race, ethnicity, and arrest offense for each individual assessed using the pretrial risk assessment tool.
(B)The total number of individuals assessed.
(2)Performance measures, including, but not limited to, all of the following:
(A)The number of interviews and assessments conducted on eligible detained individuals.
(B)The number of assessed individuals, aggregated by risk level.
(C)The number of cases in which the detention or release recommendation by the agency conducting the assessment does not conform with the release or detention decision of the judicial officer.
(D)The rate at which the pretrial
assessment staff conducting the assessments follows pretrial risk assessment criteria when recommending release or detention.
(E)The rate at which judicial officers follow the detention or release recommendation of pretrial risk assessment staff.
(F)The rate at which judicial officers make a pretrial detention decision that is more restrictive than the release recommendation of pretrial risk assessment staff.
(3)Outcome measures, including, but not limited to, all of the following:
(A)The percentage
and raw number
of assessed and released individuals who make all scheduled court appearances.
(B)The percentage and raw number of assessed and released individuals who are not charged with a new offense during the pretrial stage.
(C)The percentage and raw number of assessed and released individuals who are not
arrested for a new offense during the pretrial stage.
(D)The percentage and raw number of assessed and supervised individuals whose supervision level or detention status corresponds with their assessed risk of pretrial misconduct.
(E)The percentage and raw number of assessed and released individuals who violated conditions of release that resulted in revocation of pretrial release.
(F)The percentage and raw number of assessed and released individuals who appear for scheduled court appearances, are not charged with any new offense during pretrial supervision, and did not receive a court remand.
(4)All of the following additional data:
(A)The number of individuals released, aggregated by release types, ordered during a specified timeframe.
(B)The number of supervised defendants divided by the number of pretrial officers or case managers.
(C)The time between the pretrial services agency’s assumption of supervision of an individual
and the end of supervision of the individual.
(D)The proportion of pretrial defendants who are detained at any point throughout pretrial case processing.
(g)Beginning on or before July 1, 2021, and on or before July 1 of each year thereafter, the pretrial services agency shall publish on its internet website a report with aggregate demographic data related to outcomes and potential biases in pretrial release. The report shall, at a minimum, compare rates of release prebooking, prehearing, prearraignment, and pretrial, aggregated by race, ethnicity, gender, age, and offense. The report shall also aggregate by race, ethnicity, gender, age, and offense the number of cases in which the detention or release recommendation by the agency conducting the assessment did not conform
with the release or detention decision of the judicial officer. The report published on or before July 1, 2021, shall include at least 12 months of data. The report published on or before July 1, 2022, shall include at least 24 months of data. The report published on or before July 1, 2023, and reports published each year thereafter shall include at least 36 months of data and include changes or trends in pretrial release rates, aggregated by race, ethnicity, gender, age, and offense, and any changes in policy made by the pretrial services agency to mitigate identified biases in outcomes.
(f) Beginning on or before December 31, 2020, and on or before December 31 of each year thereafter, the Judicial Council shall publish on its internet website a report with data related to outcomes and potential biases in pretrial release. The report
shall, at a minimum, include:
(1) The following information on each county pretrial release program:
(A) The name of the pretrial risk assessment tool that is used to inform release decisions by the court.
(B) The release conditions framework used in the county.
(C) Whether a pretrial services agency is conducting interviews as part of the risk assessment.
(2) The following information by superior court in large and medium courts and otherwise aggregated by superior court size:
(A) Rates of release granted
prearraignment and rates of release granted pretrial, aggregated by gender, race or ethnicity, ZIP Code of residency and offense type.
(B) The percent of released individuals who make their required court appearances, aggregated by offense type and whether they were released on bail or pursuant to a risk assessment. For those released pursuant to a risk assessment, this information shall be aggregated by risk level.
(C) The percent of released individuals who are not charged with a new offense during the pretrial stage, aggregated by offense type and whether they were released on bail or pursuant to a risk assessment. For those released pursuant to a risk assessment, this information shall be aggregated by risk level.
(D) The number of assessed individuals by age, ZIP Code of residency, gender, and race or
ethnicity.
(E) The number of assessed individuals by risk level, ZIP Code of residency, booking charge level, and release decision.
(F) The number and percentage of assessed individuals who receive pretrial supervision by level of supervision.
(G) The number and percentage of assessed individuals, by supervision level, who fail to appear in court as required, are arrested for a new offense during the pretrial period, or have pretrial release revoked.
(3) The following information on each risk assessment tool:
(A) The percent of released individuals who attend all of their required court appearances and are not charged with a new offense during the pretrial stage, aggregated by risk
level.
(B) Risk levels aggregated by race or ethnicity, gender, offense type, ZIP Code of residency, and release or detention decision.
(C) The predictive accuracy of the tool by gender, race or ethnicity, and offense type.
(D) The proportion of cases in which the release or detention recommendation derived from the risk assessment is different than the release or detention decision imposed by the judicial officer.
(4) If feasible, the Judicial Council shall provide information on any disparate effect in the tools based on income level.
(g) (1) Pretrial services agencies and courts shall provide the Judicial Council the requisite data, as determined by the Judicial
Council, to meet the requirements of this section.
(2) The Department of Justice shall work with the Judicial Council to provide the data necessary to fulfill the requirements of this section.
(3) The Judicial Council shall not share any individual-level data with any outside entity unless it has entered into a contract for research purposes with the entity and privacy protections are established to anonymize the data.
(h) The requirements of subdivisions (f) and (g) shall apply to pretrial services agencies that perform risk assessments pursuant to a contractual agreement with the courts, including all of the following:
(1) Agencies funded pursuant to the Budget Act of 2019 as pretrial pilot projects.
(2) Agencies otherwise funded by the state to perform risk assessments.
(3) Other agencies that perform risk assessments as long as sufficient funding is provided to the Judicial Council, the superior courts,
and pretrial services agencies to ensure their ability to meet the data reporting requirements and standards set by the Judicial Council.
(i) (1) By July 1, 2022, the Judicial Council shall provide a report to the courts and the Legislature containing recommendations to mitigate bias and disparate effect in pretrial decisionmaking.
(2) A report to be submitted pursuant to this subdivision shall be submitted in compliance with Section 9795 of the Government Code.