Return to Home PageThe U.S. Equal Employment Opportunity Commission
EEOC Policy Statement on the Use of Statistics in Charges Involving the Exclusion of Individuals with Conviction Records from Employment (7/29/87)
Green v. Missouri Pacific Railroad Company, 523 F.2d 129010 EPD ¶ 10,314 (8th Cir. 1975), is the leading Title VII case on the issue of conviction records. In Green, the court held that the defendant's policy of refusing employment to any person convicted of a crime other than a minor traffic offense had an adverse impact on Black applicants and was not justified by business necessity. In a second appeal following remand, the court upheld the district court's injunctive order prohibiting the defendant from using an applicant's conviction record as an absolute bar to employment but allowing it to consider a prior criminal record as long as it constituted a business necessity. Green v. Missouri Pacific Railroad Company, 549 F.2d 1158, 1160,13 EPD ¶ 11,579 (8th Cir. 1977). See also Commission Decision No. 72- 1497, CCH EEOC Decisions (1973) ¶ 6352, and Commission Decision Nos. 74-89, 78-10, 78-35, and 80-10, CCH EEOC Decisions (1983) ¶¶ 6418, 6715, 6720, and 6822, respectively.
It is the Commission's position that an employer's policy or practice of excluding individuals from employment on the basis of their conviction records has an adverse impact on Blacks(1) and Hispanics(2) in light of statistics showing that they are convicted at a rate disproportionately greater than their representation in the population. Policy Statement on the Issue of Conviction Records Under Title VII (February 4, 1987). However, when the employer can present more narrowly drawn statistics showing either that Blacks and Hispanics are not convicted at a disproportionately greater rate or that there is no adverse impact in its own hiring process resulting from the convictions policy, then a no cause determination would be appropriate.
1. Where the Employer's Policy is Not Crime-Specific
An employer's policy of excluding from employment all persons convicted of any crime is likely to create an adverse impact for Blacks and Hispanics based on national and regional conviction rate statistics. However, it is open to the respondent/employer to present more narrow local, regional, or applicant flow data, showing that the policy probably will not have an adverse impact on its applicant pool and/or in fact does not have an adverse impact on the pool. As the Supreme Court has stated,
Although 'a statistical showing of disproportionate impact need not always be based on an analysis of the characteristics of actual applicants,' Dothard v. Rawlinson, 433 U.S. 321, 330, 'evidence showing that the figures for the general population might not accurately reflect the pool of qualified job applicants' undermines the significance of such figures. Teamsters v. United States, 431 U.S. 324, 340 n. 20.
New York City Transit Authority v. Beazer, 440 U.S. 568, 586 n. 29,19 EPD ¶ 9027 at p. 6315 (1979). See also Costa v. Markey, 30 EPD ¶ 33,173 at p. 27,638 (1st Cir. 1982), vacated on other grounds, 706 F.2d 796, 32 EPD ¶ 32,622 (1st Cir.), cert. denied, 104 S. Ct. 547, 32 EPD ¶ 33,955 (1983).
If the employer provides applicant flow data, information should be sought to assure that the employer's applicant pool was not artificially limited by discouragement. For example, if many Blacks with conviction records did not apply for a particular job because they knew of the employer's policy and they therefore expected to be rejected, then applicant flow data would not be an accurate reflection of the conviction policy's actual effect. See Dothard v. Rawlinson, 433 U.S. 321, 330 (1977). (Section 608, Recruitment, of Volume II of the Compliance Manual will provide a more detailed discussion of when and how to investigate for discouragement.)
2. Where the Employer's Policy is Crime-Specific
In the past, when the Commission has evaluated an employer's "no convictions" policy dealing with a subcategory of crimes; e.g., theft, robbery, or drug-related crimes; the Commission has relied upon national or regional conviction statistics for crimes as a whole. See, e.g., Commission Decision No. 73- 0257, CCH EEOC Decisions (1973) ¶ 6372, and Commission Decision Nos. 76-110 and 80-17, CCH EEOC Decisions (1983) ¶¶ 6676 and 6809, respectively. However, these statistics only show a probability of adverse impact for Blacks and Hispanics, while more narrow data may show no adverse impact.
If the employer can present more narrow regional or local data on conviction rates for all crimes showing that Blacks and Hispanics are not convicted at disproportionately higher rates, then a no cause determination would be proper.(3) Alternatively, the employer may present national, regional, or local data on conviction rates for the particular crime which is targeted in its crime-specific convictions policy. If such data shows no adverse impact, then a no cause determination would be appropriate. Finally, the employer can use applicant flow data to demonstrate that its conviction policy has not resulted in the exclusion from employment of a disproportionately high number of Blacks and Hispanics.
1. See, e.g., Commission Decision No. 72-1497, CCH EEOC Decisions (1973) ¶ 6352, and Commission Decision Nos. 74-89, 78-10, 78-35, and 80- 10, CCH EEOC Decisions (1983) ¶¶ 6418, 6715, 6720, and 6822, respectively.
2. See Commission Decision No. 78-03, CCH EEOC Decisions (1983) ¶ 6714.
3. However, if even more narrow statistics, such as regional or local crime-specific data, show adverse impact, then a cause finding would be appropriate absent a justifying business necessity.
This page was last modified on September 20, 2006.