The Civil Rights Act of 1991 (CRA91) was enacted after a rancorous debate about whether it was a ``quota''
hiring bill or a necessary means of opening labor markets. We analyze the effects of CRA91 on the composition
of firms' workforces. We consider employer behavior when firms vary in their susceptibility to discrimination
suits and when firms can reduce exposure to discrimination claims by employing more protected workers. These
forces lead to a sorting effect, which causes firms that are more susceptible to discrimination
litigation to substitute away from protected workers, and a quota effect, which causes
firms with fewer pre-CRA91 protected workers to substitute toward these workers. Using data from
various sources, we find evidence consistent with CRA91 having had a sorting effect. We find no evidence that
CRA91 led to widespread quota hiring and no evidence that CRA91 helped integrate industries that had employed
relatively few protected workers.