At a Glance
On April 9, 2016, the plaintiffs reached a settlement with the Census Bureau. The parties filed a motion for final approval of the settlement with the District Court on April 19, 2016, that is currently pending, and a settlement fairness hearing has yet to be scheduled.
Outten & Golden
In 2010, the U.S. Bureau of the Census hired over a million temporary employees to conduct census surveys and serve in clerical positions. The Bureau ran the names of all applicants through the FBI Criminal Justice Information Services Division’s Name Index, which contains a “criminal history record” for anyone who has ever been arrested. If an applicant’s name appeared, Census informed the applicant and gave them the opportunity to “to provide official court documentation on any arrest(s) that affect hiring eligibility” with a 30-day deadline.
However, half of the FBI’s records have no information on the final outcome of the arrest, i.e. whether the case actually led to a prosecution or conviction. Even those with expunged records may still have their names in the FBI’s database. Putting the onus on the applicant to produce these records, especially those that may be expunged or destroyed, was a time-consuming, arduous task, especially within the short timeframe (which also violated federal policy). While an arrest did not necessarily preclude an otherwise eligible applicant from being hired, it appeared that, in practice, the Census Bureau intended to use a single arrest, regardless of the underlying circumstances, to disqualify an applicant. Moreover, the inquiry into arrest history – regardless of how old the arrest was, underlying circumstances, or disposition – deterred or excluded many of the best qualified applicants with arrest records that were irrelevant to census-taking (such as minor misdemeanors that are decades old and which could not reasonably be job-related), because these are precisely the kind of records that are the most difficult to find.
Such a strict use of arrest records as an employment screening criteria meant that people of color, in particular Blacks and Latinos, were disproportionately denied employment because they constitute a disproportionate share of criminal arrests nationwide. Because of the enormous scope of hiring for the 2010 Census, the discriminatory impact of this was profound. Ironically, the Census Bureau expressed a desire to reach communities at risk of being undercounted, particularly low-income people of color and immigrants, by hiring within those communities, yet it erected an unnecessary and discriminatory obstacle to achieving that very goal.
Plaintiffs brought this class action suit as a disparate impact challenge pursuant to Title VII of the Civil Rights Act. Under the disparate impact doctrine (which does not require a showing of intentional discrimination), the Census Bureau violated Title VII by using an employment screening device that had a racially disproportionate effect on African American, Latino, and Native American applicants. The plaintiffs argued that the Bureau’s discriminatory practice was not justified by business necessity, and its legitimate interests could be achieved through less discriminatory means.