The broad goal of this study was to examine multiple potential predictors of anti-Black bias among counselors. Specifically, in an online survey of 173 trainees and professionals in mental health, this study used 3 measures related to cultural sensitivity as predictors of therapists' expectancies for bond and prognosis with African American clients compared with White clients. The Multicultural Counseling Inventory (MCI; Sodowsky, Taffe, Gutkin, & Wise, 1994) was used to measure global multicultural competence. The Implicit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998) served as a measure of automatic prejudice toward Blacks. Additionally, a new self-report measure of anti-Black clinical prejudice was created specifically for this study. The Balanced Inventory of Desirable Responding (Paulhus, 1984) was included to control for socially desirable responding. Each predictor of cultural sensitivity uniquely explained variance in anti-Black bias in bond ratings, with the IAT accounting for more variance than the 2 self-reports. Our novel measure of clinical prejudice accounted for anti-Black bias in prognosis ratings, but the MCI and the IAT did not. Researchers studying cultural competence are encouraged to consider the roles of automatic and deliberate prejudice in determining disparities in clinical expectancies and cross-racial therapeutic alliances. (PsycINFO Database Record (c) 2014 APA, all rights reserved).