ASSEMBLY THIRD READING

ATR Bill Id:ACR 125

Author:(Jones-Sawyer and Holden)

As Amended Ver:September 4, 2019

Majority vote

SUMMARY:

Urges policymakers in both federal and state government to explore ways to promote the development and use of new technologies to reduce bias and discrimination in hiring and employment.

Specifically, this resolution makes the following findings and declarations:

1. The State of California has one of the world's most diverse populations, with citizens from a wide variety of ethnic, racial, and national backgrounds. No single race or ethnic group constitutes a majority of the state's population, and 39% of residents are Latino, 14% are Asian American, 6% are African American, and 3% identify as multiracial; and
2. The current laws of this state protect and safeguard the right and opportunity of all persons to seek, obtain, and hold employment without discrimination on account of certain protected characteristics, including race, ethnic group, or sex; and
3. The current laws of this state permit employers to utilize tests and other selection procedures for purposes of hiring or promotion, provided the tests and selection procedures do not have an unlawful discriminatory impact; and
4. Under current state law, a test or other selection procedure that has an adverse impact on employment opportunities for individuals on the basis of race, ethnic group, or sex is unlawful unless the employer establishes that the test or selection procedure is job-related and consistent with business necessity. In addition, under current law, an employer that utilizes a test or other selection procedure must, prior to implementation of the test or selection procedure, investigate valid alternatives and must opt for the alternative that poses the lowest risk of discriminatory impact; and
5. Many commonly used selection procedures fail to ensure unbiased decision making and also yield poor outcomes for employers. A 2014 report by the Harvard Business Review found that résumés were in fact the worst predictor of job success of any of the employment selection tools studied; and
6. Despite laws prohibiting employment discrimination, a 2017 study reported in the Harvard Business Review analyzing discrimination trends found that hiring discrimination against Black workers in the United States has not declined in the last quarter century, and has declined only slightly for Latino workers. According to the study, since 1990, White applicants received, on average, 36% more callbacks than Black applicants and 24% more callbacks than Latino applicants with identical résumés; and
7. Research published in the Harvard Business Review found that the country's most prestigious employers limited their campus recruiting efforts to the most elite institutions, which has the effect of excluding many lower income and minority students from consideration; and
8. A 2017 study by Harvard University's Opportunity Insights group found that 25% of the wealthiest American students attend an elite college, while less than 0.5% of children from the bottom fifth of American families attend an elite college. Data from the National Center for Education Statistics found Black and Latino students were more underrepresented at the nation's top colleges and universities in 2017 than they had been 35 years before; and
9. Innovative technologies for hiring and promotion, including artificial intelligence and algorithm-based technologies, have the potential to reduce bias and discrimination in hiring and promotion based on protected characteristics, as well as helping where appropriate to reduce bias based on unprotected characteristics, such as socioeconomic status or status as a formerly incarcerated person. At the same time, these technologies can help employers reach larger and more diverse pools of qualified talent and better identify candidates with the right skills and abilities to succeed; and
10. California should be the leader in the appropriate standards that are necessary to ensure that the technologies being developed and used will reduce employment bias and discrimination, while achieving better employment outcomes and ensuring equal opportunity to economic mobility.

COMMENTS:

See policy Committee analysis.

According to the Author:

"ACR 125 aims to raise awareness of the ongoing racial discrimination and gender disparity that continue to exist within the job hiring process. Much of the tools used by employers have shown to be ineffective in decreasing discrimination and bias. For example, while standardized testing has proven to identify qualified candidates, it often leads to biased outcomes. Likewise, a myriad of studies have shown that decisions made primarily on resumes have evidence of increased bias against women and minorities…

In addition to discriminatory hiring practices, these common selection tools can result in poor outcomes for employers with information found in resumes shown to be a poor predictor of an employee's future success. In another study by the Harvard Business Review, new hires were found to fail 50% of the time, a practice that is not only costly for employers, but also demoralizing for workers. Ultimately, employers resort to hiring candidates of the same talents, ignoring those of diverse backgrounds who could be a better fit for the objectives of the position."

Arguments in Support:

TechNet, states in support, "While the challenges in hiring are not unique to California, our state should be a leader when it comes to developing solutions that will serve workers and employers. New algorithm-based technologies have real potential to allow employers to open their pipelines and consider unlimited numbers of applications efficiently and objectively to identify the best fits for a particular role. Unlike human processes, these technologies can be audited to identify and remove potential bias. But at this time, there are no standards governing the development or use of hiring technologies to ensure they are in fact improving objectivity and helping to reduce disparities."

Arguments in Opposition:

None on file.

FISCAL COMMENTS:

None

VOTES:

ASM LABOR AND EMPLOYMENT: 7-0-0

YES: Kalra, Flora, Carrillo, Diep, Gonzalez, Jones-Sawyer, Luz Rivas

UPDATED:

VERSION: September 4, 2019

CONSULTANT: Megan Lane / L. & E. / (916) 319-2091 FN: 0002205

Date of Hearing: September 9, 2019

ASSEMBLY COMMITTEE ON LABOR AND EMPLOYMENT

Ash Kalra, Chair

ABPCA Bill Id:ACR 125 (

Author:Jones-Sawyer) – As Introduced Ver:August 14, 2019

SUBJECT: Bias and discrimination in hiring reduction through new technology

SUMMARY: Urges policymakers in both federal and state government to explore ways to promote the development and use of new technologies to reduce bias and discrimination in hiring and employment.

Specifically, this resolution makes the following findings and declarations:

1. The State of California has one of the world’s most diverse populations, with citizens from a wide variety of ethnic, racial, and national backgrounds. No single race or ethnic group constitutes a majority of the state’s population, and 39 percent of residents are Latino, 14 percent are Asian American, 6 percent are African American, and 3 percent identify as multiracial; and 
2. The current laws of this state protect and safeguard the right and opportunity of all persons to seek, obtain, and hold employment without discrimination on account of certain protected characteristics, including race, ethnic group, or sex; and
3. The current laws of this state permit employers to utilize tests and other selection procedures for purposes of hiring or promotion, provided the tests and selection procedures do not have an unlawful discriminatory impact; and
4. Under current state law, a test or other selection procedure that has an adverse impact on employment opportunities for individuals on the basis of race, ethnic group, or sex is unlawful unless the employer establishes that the test or selection procedure is job-related and consistent with business necessity. In addition, under current law, an employer that utilizes a test or other selection procedure must, prior to implementation of the test or selection procedure, investigate valid alternatives and must opt for the alternative that poses the lowest risk of discriminatory impact; and
5. Many commonly used selection procedures fail to ensure unbiased decision making and also yield poor outcomes for employers. A 2014 report by the Harvard Business Review found that résumés were in fact the worst predictor of job success of any of the employment selection tools studied; and
6. Despite laws prohibiting employment discrimination, a 2017 study reported in the Harvard Business Review analyzing discrimination trends found that hiring discrimination against Black workers in the United States has not declined in the last quarter century, and has declined only slightly for Latino workers. According to the study, since 1990, White applicants received, on average, 36 percent more callbacks than Black applicants and 24 percent more callbacks than Latino applicants with identical résumés; and
7. Research published in the Harvard Business Review found that the country’s most prestigious employers limited their campus recruiting efforts to the most elite institutions, which has the effect of excluding many lower income and minority students from consideration; and
8. A 2017 study by Harvard University’s Opportunity Insights group found that 25 percent of the wealthiest American students attend an elite college, while less than 0.5 percent of children from the bottom fifth of American families attend an elite college. Data from the National Center for Education Statistics found Black and Latino students were more underrepresented at the nation’s top colleges and universities in 2017 than they had been 35 years before; and
9. Innovative technologies for hiring and promotion, including artificial intelligence and algorithm-based technologies, have the potential to reduce bias and discrimination in hiring and promotion based on protected characteristics, as well as helping where appropriate to reduce bias based on unprotected characteristics, such as socioeconomic status or status as a formerly incarcerated person. At the same time, these technologies can help employers reach larger and more diverse pools of qualified talent and better identify candidates with the right skills and abilities to succeed; and
10. California should be the leader in the appropriate standards that are necessary to ensure that the technologies being developed and used will reduce employment bias and discrimination, while achieving better employment outcomes and ensuring equal opportunity to economic mobility.

In addition, this resolution makes the following resolutions:

The Assembly of the State of California, the Senate thereof concurring, that the Legislature urges policymakers in federal and state government to explore ways to promote the development and use of new technologies that will reduce bias and discrimination in hiring, help employers find the best-suited candidates based on objective factors, and help ensure that all persons who seek, obtain, and hold employment in this state are able to do so free of discrimination.

COMMENTS: According to the author, “ACR 125 aims to raise awareness of the ongoing racial discrimination and gender disparity that continue to exist within the job hiring process. Much of the tools used by employers have shown to be ineffective in decreasing discrimination and bias. For example, while standardized testing has proven to identify qualified candidates, it often leads to biased outcomes. Likewise, a myriad of studies have shown that decisions made primarily on resumes have evidence of increased bias against women and minorities…

In addition to discriminatory hiring practices, these common selection tools can result in poor outcomes for employers with information found in resumes shown to be a poor predictor of an employee’s future success. In another study by the Harvard Business Review, new hires were found to fail 50% of the time, a practice that is not only costly for employers, but also demoralizing for workers. Ultimately, employers resort to hiring candidates of the same talents, ignoring those of diverse backgrounds who could be a better fit for the objectives of the position.”

TechNet, states in support, “While the challenges in hiring are not unique to California, our state should be a leader when it comes to developing solutions that will serve workers and employers. New algorithm-based technologies have real potential to allow employers to open their pipelines and consider unlimited numbers of applications efficiently and objectively to identify the best fits for a particular role. Unlike human processes, these technologies can be audited to identify and remove potential bias. But at this time, there are no standards governing the development or use of hiring technologies to ensure they are in fact improving objectivity and helping to reduce disparities.”

Committee Comments

Committee staff note that a review of the literature on the use of AI in hiring decisions reveals a lack of consensus on the link between AI technology and reducing bias. As pointed out in a recent article by the Harvard Business Review, “many hope that algorithms will help human decision-makers avoid their own prejudices by adding consistency to the hiring process. But algorithms introduce new risks of their own. They can replicate institutional and historical biases, amplifying disadvantages lurking in data points like university attendance or performance evaluation scores. Even if algorithms remove some subjectivity from the hiring process, humans are still very much involved in final hiring decisions. Arguments that cast ‘objective’ algorithms as fairer and more accurate than fallible humans fail to fully recognize that in most cases, both play a role.”^[1] Understanding how algorithm and other AI based technology can be developed and appropriately deployed to reduce bias in employment certainly warrants further study, debate, and the engagement of policy makers.

REGISTERED SUPPORT / OPPOSITION:

Support

California Hispanic Chambers of Commerce
Calinnovates
Code.org
National Diversity Coalition
Technet

Opposition

None on file.

Analysis Prepared by: Megan Lane / L. & E. /

1. Miranda Bogen, “All The Ways Hiring Algorithms Can Introduce Bias,” Harvard Business Review, May 6, 2019. See also Nicole Martin, “Are AI Hiring Programs Eliminating Bias Or Making It Worse?” Forbes, December 13, 2018; Marisa Sanfilippo, “Can Artificial Intelligence Solve Your Bias Problem?” Business News Daily, July 16, 2018. ↑