The counseling profession has become diverse and it is not uncommon for a person of color to be included in the supervisory dyad. Therefore, counselor supervisors are required to become competent in supervising people who are racially different from themselves (Daniels, D’Andrea, & Kim, 1999; Young, 2004). Most cross-racial supervision literature examines the supervisory working alliance, but the purpose of this study was to investigate the predictability of social influence, racial identity attitudes, and gender self-confidence on supervisors’ self-efficacy to work with racially-different supervisees. The sample in this study included 149 supervisors who conduct cross-racial supervision. The complete assessments were the Counselor Supervisor Self-Efficacy Scale, the Interpersonal Power Inventory, the Multigroup Ethnic Identity Measure, the Hoffman Gender Scale, and a demographic questionnaire. Multiple regressions were conducted to analyze the data. Findings indicated that: (a) supervisors with flexible attitudes toward power usage and gender self-confidence conducted cross-racial supervision with more efficacy, (b) supervisors with rigid attitudes toward race and power usage broached race more than their counterparts with flexible attitudes, and (c) African-American supervisors broached race less when they worked with racially-different supervisees. Knowledge regarding predictors of supervisor self-efficacy in cross-racial supervision is essential for counselor educators and supervisors. The outcomes of the study beg for a more in-depth examination of the issues surrounding self-efficacy in cross-racial supervision, particularly with ethnic minority supervisors.