Diversity management
Statistics as evidence of discrimination
Federal Employment Court ruling
Comment
Diversity management
The sales team of a German mechanical engineering company comprises seven women, 40 men, two employees with disabilities, two Spanish employees, one Chinese employee, two Japanese employees, one Indian employee, one Kenyan employee and two US employees of Mexican origin, as well as German employees. Is this staff structure a reflection of the freedom to make entrepreneurial decisions and human resources (HR) policy? Or is it the result of discrimination? Such questions must be asked in light of the ongoing debate regarding diversity. Employers must be clear whether HR policy should aim to ensure that the different employee groups are represented to a certain percentage in different positions and cross different management levels, or whether decisions relating to hiring and promotion may continue to be based on performance, know-how, sales experience, the interests of customers and the employer's personal assessment.
The main goal must be to avoid making HR decisions based on statistical requirements and subsequent litigation arising from alleged discrimination.
Statistics as evidence of discrimination
A ruling of the 15th Chamber of the Berlin-Brandenburg Regional Employment Court on November 27 2008(1) made clear that such statistics can be used for this purpose – and possibly also abused. In the ruling the court assumed that the statistical distribution of the sexes on the individual levels of hierarchy of a company could indicate that sex discrimination plays a role in decisions on promotion. In the case at hand, the claimant (a woman) and a male employee (Mr R) were employed as personnel directors at two of the defendant's head offices. They both reported to the HR director of the company, who informed the claimant that R was intended to be his successor. The claimant found fault with this decision: she felt that she was being discriminated against because she was a woman, and because the 27 positions on the top three management levels of the company were occupied solely by men, whereas 69% of the staff as a whole were women.
On appeal on a point of law, in its ruling of July 22 2010(2) the Federal Employment Court further defined the appropriateness of the use of statistics as evidence of discrimination and in doing so set the course for the prevention of abusive lawsuits based on statistics.
Federal Employment Court ruling
The court first affirmed that statistics may be used as evidence of sex discrimination within the meaning of Section 22 of the General Equal Treatment Act. However, the court held that statistics can be used to back up a presumption within the meaning of Section 22 of the act only if they:
- relate specifically to the employer in question; and
- are relevant with regard to the employer's conduct.
Important and effective limitations of the possibilities for filing abusive lawsuits were thus formulated.
The court first explicitly stated that ratios in other companies were irrelevant, and then stated that the sex quota of the entire staff was also irrelevant. A rule that the sexes must be similarly distributed on all levels of hierarchy of a company is not customary; nor are there legal terms of reference for such a rule.
If statistics exist concerning the treatment of comparable employees by the same employer, they must be relevant with regard to the employer's conduct. In particular, this means that promotion possibilities and probabilities – as well as social conditions – are to be taken into account. As a result, little room is left for abusive lawsuits claiming sex discrimination based solely on statistics. The following finding of the court puts the importance of statistics as evidence of sex discrimination into perspective:
"There must be other reasons in addition to mere statistics to presume that women are discriminated against because of their sex in decisions on advancement."
Comment
Therefore, future claims will not be allowed, as a rule, solely on the basis of the evidence of statistics. This must be acceded: mathematics alone is inappropriate for proving discrimination. If statistics are to have significance, not only must the relevant reference groups be clear, but the significance of the statistics must also be substantiated by the claimant in the individual case.
For further information on this topic please contact Andrea Bonanni, Bjoern Gaul, Daniel Ludwig or Bernd Roock at CMS Hasche Sigle by telephone (+49 221 7716 128), fax (+49 221 7716 110) or email ([email protected] [email protected], [email protected] and [email protected]).
Endnotes
(1) Decision 15 Sa 517/08.
(2) Decision 8 AZR 1012/08.